Annotation of imach/src/imach.c, revision 1.346
1.346 ! brouard 1: /* $Id: imach.c,v 1.345 2022/09/16 13:40:11 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.346 ! brouard 4: Revision 1.345 2022/09/16 13:40:11 brouard
! 5: Summary: Version 0.99r41
! 6:
! 7: * imach.c (Module): 0.99r41 Was an error when product of timevarying and fixed. Using FixedV[of name] now. Thank you Feinuo
! 8:
1.345 brouard 9: Revision 1.344 2022/09/14 19:33:30 brouard
10: Summary: version 0.99r40
11:
12: * imach.c (Module): Fixing names of variables in T_ (thanks to Feinuo)
13:
1.344 brouard 14: Revision 1.343 2022/09/14 14:22:16 brouard
15: Summary: version 0.99r39
16:
17: * imach.c (Module): Version 0.99r39 with colored dummy covariates
18: (fixed or time varying), using new last columns of
19: ILK_parameter.txt file.
20:
1.343 brouard 21: Revision 1.342 2022/09/11 19:54:09 brouard
22: Summary: 0.99r38
23:
24: * imach.c (Module): Adding timevarying products of any kinds,
25: should work before shifting cotvar from ncovcol+nqv columns in
26: order to have a correspondance between the column of cotvar and
27: the id of column.
28: (Module): Some cleaning and adding covariates in ILK.txt
29:
1.342 brouard 30: Revision 1.341 2022/09/11 07:58:42 brouard
31: Summary: Version 0.99r38
32:
33: After adding change in cotvar.
34:
1.341 brouard 35: Revision 1.340 2022/09/11 07:53:11 brouard
36: Summary: Version imach 0.99r37
37:
38: * imach.c (Module): Adding timevarying products of any kinds,
39: should work before shifting cotvar from ncovcol+nqv columns in
40: order to have a correspondance between the column of cotvar and
41: the id of column.
42:
1.340 brouard 43: Revision 1.339 2022/09/09 17:55:22 brouard
44: Summary: version 0.99r37
45:
46: * imach.c (Module): Many improvements for fixing products of fixed
47: timevarying as well as fixed * fixed, and test with quantitative
48: covariate.
49:
1.339 brouard 50: Revision 1.338 2022/09/04 17:40:33 brouard
51: Summary: 0.99r36
52:
53: * imach.c (Module): Now the easy runs i.e. without result or
54: model=1+age only did not work. The defautl combination should be 1
55: and not 0 because everything hasn't been tranformed yet.
56:
1.338 brouard 57: Revision 1.337 2022/09/02 14:26:02 brouard
58: Summary: version 0.99r35
59:
60: * src/imach.c: Version 0.99r35 because it outputs same results with
61: 1+age+V1+V1*age for females and 1+age for females only
62: (education=1 noweight)
63:
1.337 brouard 64: Revision 1.336 2022/08/31 09:52:36 brouard
65: *** empty log message ***
66:
1.336 brouard 67: Revision 1.335 2022/08/31 08:23:16 brouard
68: Summary: improvements...
69:
1.335 brouard 70: Revision 1.334 2022/08/25 09:08:41 brouard
71: Summary: In progress for quantitative
72:
1.334 brouard 73: Revision 1.333 2022/08/21 09:10:30 brouard
74: * src/imach.c (Module): Version 0.99r33 A lot of changes in
75: reassigning covariates: my first idea was that people will always
76: use the first covariate V1 into the model but in fact they are
77: producing data with many covariates and can use an equation model
78: with some of the covariate; it means that in a model V2+V3 instead
79: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
80: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
81: the equation model is restricted to two variables only (V2, V3)
82: and the combination for V2 should be codtabm(k,1) instead of
83: (codtabm(k,2), and the code should be
84: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
85: made. All of these should be simplified once a day like we did in
86: hpxij() for example by using precov[nres] which is computed in
87: decoderesult for each nres of each resultline. Loop should be done
88: on the equation model globally by distinguishing only product with
89: age (which are changing with age) and no more on type of
90: covariates, single dummies, single covariates.
91:
1.333 brouard 92: Revision 1.332 2022/08/21 09:06:25 brouard
93: Summary: Version 0.99r33
94:
95: * src/imach.c (Module): Version 0.99r33 A lot of changes in
96: reassigning covariates: my first idea was that people will always
97: use the first covariate V1 into the model but in fact they are
98: producing data with many covariates and can use an equation model
99: with some of the covariate; it means that in a model V2+V3 instead
100: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
101: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
102: the equation model is restricted to two variables only (V2, V3)
103: and the combination for V2 should be codtabm(k,1) instead of
104: (codtabm(k,2), and the code should be
105: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
106: made. All of these should be simplified once a day like we did in
107: hpxij() for example by using precov[nres] which is computed in
108: decoderesult for each nres of each resultline. Loop should be done
109: on the equation model globally by distinguishing only product with
110: age (which are changing with age) and no more on type of
111: covariates, single dummies, single covariates.
112:
1.332 brouard 113: Revision 1.331 2022/08/07 05:40:09 brouard
114: *** empty log message ***
115:
1.331 brouard 116: Revision 1.330 2022/08/06 07:18:25 brouard
117: Summary: last 0.99r31
118:
119: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
120:
1.330 brouard 121: Revision 1.329 2022/08/03 17:29:54 brouard
122: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
123:
1.329 brouard 124: Revision 1.328 2022/07/27 17:40:48 brouard
125: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
126:
1.328 brouard 127: Revision 1.327 2022/07/27 14:47:35 brouard
128: Summary: Still a problem for one-step probabilities in case of quantitative variables
129:
1.327 brouard 130: Revision 1.326 2022/07/26 17:33:55 brouard
131: Summary: some test with nres=1
132:
1.326 brouard 133: Revision 1.325 2022/07/25 14:27:23 brouard
134: Summary: r30
135:
136: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
137: coredumped, revealed by Feiuno, thank you.
138:
1.325 brouard 139: Revision 1.324 2022/07/23 17:44:26 brouard
140: *** empty log message ***
141:
1.324 brouard 142: Revision 1.323 2022/07/22 12:30:08 brouard
143: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
144:
1.323 brouard 145: Revision 1.322 2022/07/22 12:27:48 brouard
146: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
147:
1.322 brouard 148: Revision 1.321 2022/07/22 12:04:24 brouard
149: Summary: r28
150:
151: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
152:
1.321 brouard 153: Revision 1.320 2022/06/02 05:10:11 brouard
154: *** empty log message ***
155:
1.320 brouard 156: Revision 1.319 2022/06/02 04:45:11 brouard
157: * imach.c (Module): Adding the Wald tests from the log to the main
158: htm for better display of the maximum likelihood estimators.
159:
1.319 brouard 160: Revision 1.318 2022/05/24 08:10:59 brouard
161: * imach.c (Module): Some attempts to find a bug of wrong estimates
162: of confidencce intervals with product in the equation modelC
163:
1.318 brouard 164: Revision 1.317 2022/05/15 15:06:23 brouard
165: * imach.c (Module): Some minor improvements
166:
1.317 brouard 167: Revision 1.316 2022/05/11 15:11:31 brouard
168: Summary: r27
169:
1.316 brouard 170: Revision 1.315 2022/05/11 15:06:32 brouard
171: *** empty log message ***
172:
1.315 brouard 173: Revision 1.314 2022/04/13 17:43:09 brouard
174: * imach.c (Module): Adding link to text data files
175:
1.314 brouard 176: Revision 1.313 2022/04/11 15:57:42 brouard
177: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
178:
1.313 brouard 179: Revision 1.312 2022/04/05 21:24:39 brouard
180: *** empty log message ***
181:
1.312 brouard 182: Revision 1.311 2022/04/05 21:03:51 brouard
183: Summary: Fixed quantitative covariates
184:
185: Fixed covariates (dummy or quantitative)
186: with missing values have never been allowed but are ERRORS and
187: program quits. Standard deviations of fixed covariates were
188: wrongly computed. Mean and standard deviations of time varying
189: covariates are still not computed.
190:
1.311 brouard 191: Revision 1.310 2022/03/17 08:45:53 brouard
192: Summary: 99r25
193:
194: Improving detection of errors: result lines should be compatible with
195: the model.
196:
1.310 brouard 197: Revision 1.309 2021/05/20 12:39:14 brouard
198: Summary: Version 0.99r24
199:
1.309 brouard 200: Revision 1.308 2021/03/31 13:11:57 brouard
201: Summary: Version 0.99r23
202:
203:
204: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
205:
1.308 brouard 206: Revision 1.307 2021/03/08 18:11:32 brouard
207: Summary: 0.99r22 fixed bug on result:
208:
1.307 brouard 209: Revision 1.306 2021/02/20 15:44:02 brouard
210: Summary: Version 0.99r21
211:
212: * imach.c (Module): Fix bug on quitting after result lines!
213: (Module): Version 0.99r21
214:
1.306 brouard 215: Revision 1.305 2021/02/20 15:28:30 brouard
216: * imach.c (Module): Fix bug on quitting after result lines!
217:
1.305 brouard 218: Revision 1.304 2021/02/12 11:34:20 brouard
219: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
220:
1.304 brouard 221: Revision 1.303 2021/02/11 19:50:15 brouard
222: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
223:
1.303 brouard 224: Revision 1.302 2020/02/22 21:00:05 brouard
225: * (Module): imach.c Update mle=-3 (for computing Life expectancy
226: and life table from the data without any state)
227:
1.302 brouard 228: Revision 1.301 2019/06/04 13:51:20 brouard
229: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
230:
1.301 brouard 231: Revision 1.300 2019/05/22 19:09:45 brouard
232: Summary: version 0.99r19 of May 2019
233:
1.300 brouard 234: Revision 1.299 2019/05/22 18:37:08 brouard
235: Summary: Cleaned 0.99r19
236:
1.299 brouard 237: Revision 1.298 2019/05/22 18:19:56 brouard
238: *** empty log message ***
239:
1.298 brouard 240: Revision 1.297 2019/05/22 17:56:10 brouard
241: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
242:
1.297 brouard 243: Revision 1.296 2019/05/20 13:03:18 brouard
244: Summary: Projection syntax simplified
245:
246:
247: We can now start projections, forward or backward, from the mean date
248: of inteviews up to or down to a number of years of projection:
249: prevforecast=1 yearsfproj=15.3 mobil_average=0
250: or
251: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
252: or
253: prevbackcast=1 yearsbproj=12.3 mobil_average=1
254: or
255: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
256:
1.296 brouard 257: Revision 1.295 2019/05/18 09:52:50 brouard
258: Summary: doxygen tex bug
259:
1.295 brouard 260: Revision 1.294 2019/05/16 14:54:33 brouard
261: Summary: There was some wrong lines added
262:
1.294 brouard 263: Revision 1.293 2019/05/09 15:17:34 brouard
264: *** empty log message ***
265:
1.293 brouard 266: Revision 1.292 2019/05/09 14:17:20 brouard
267: Summary: Some updates
268:
1.292 brouard 269: Revision 1.291 2019/05/09 13:44:18 brouard
270: Summary: Before ncovmax
271:
1.291 brouard 272: Revision 1.290 2019/05/09 13:39:37 brouard
273: Summary: 0.99r18 unlimited number of individuals
274:
275: The number n which was limited to 20,000 cases is now unlimited, from firstobs to lastobs. If the number is too for the virtual memory, probably an error will occur.
276:
1.290 brouard 277: Revision 1.289 2018/12/13 09:16:26 brouard
278: Summary: Bug for young ages (<-30) will be in r17
279:
1.289 brouard 280: Revision 1.288 2018/05/02 20:58:27 brouard
281: Summary: Some bugs fixed
282:
1.288 brouard 283: Revision 1.287 2018/05/01 17:57:25 brouard
284: Summary: Bug fixed by providing frequencies only for non missing covariates
285:
1.287 brouard 286: Revision 1.286 2018/04/27 14:27:04 brouard
287: Summary: some minor bugs
288:
1.286 brouard 289: Revision 1.285 2018/04/21 21:02:16 brouard
290: Summary: Some bugs fixed, valgrind tested
291:
1.285 brouard 292: Revision 1.284 2018/04/20 05:22:13 brouard
293: Summary: Computing mean and stdeviation of fixed quantitative variables
294:
1.284 brouard 295: Revision 1.283 2018/04/19 14:49:16 brouard
296: Summary: Some minor bugs fixed
297:
1.283 brouard 298: Revision 1.282 2018/02/27 22:50:02 brouard
299: *** empty log message ***
300:
1.282 brouard 301: Revision 1.281 2018/02/27 19:25:23 brouard
302: Summary: Adding second argument for quitting
303:
1.281 brouard 304: Revision 1.280 2018/02/21 07:58:13 brouard
305: Summary: 0.99r15
306:
307: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
308:
1.280 brouard 309: Revision 1.279 2017/07/20 13:35:01 brouard
310: Summary: temporary working
311:
1.279 brouard 312: Revision 1.278 2017/07/19 14:09:02 brouard
313: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
314:
1.278 brouard 315: Revision 1.277 2017/07/17 08:53:49 brouard
316: Summary: BOM files can be read now
317:
1.277 brouard 318: Revision 1.276 2017/06/30 15:48:31 brouard
319: Summary: Graphs improvements
320:
1.276 brouard 321: Revision 1.275 2017/06/30 13:39:33 brouard
322: Summary: Saito's color
323:
1.275 brouard 324: Revision 1.274 2017/06/29 09:47:08 brouard
325: Summary: Version 0.99r14
326:
1.274 brouard 327: Revision 1.273 2017/06/27 11:06:02 brouard
328: Summary: More documentation on projections
329:
1.273 brouard 330: Revision 1.272 2017/06/27 10:22:40 brouard
331: Summary: Color of backprojection changed from 6 to 5(yellow)
332:
1.272 brouard 333: Revision 1.271 2017/06/27 10:17:50 brouard
334: Summary: Some bug with rint
335:
1.271 brouard 336: Revision 1.270 2017/05/24 05:45:29 brouard
337: *** empty log message ***
338:
1.270 brouard 339: Revision 1.269 2017/05/23 08:39:25 brouard
340: Summary: Code into subroutine, cleanings
341:
1.269 brouard 342: Revision 1.268 2017/05/18 20:09:32 brouard
343: Summary: backprojection and confidence intervals of backprevalence
344:
1.268 brouard 345: Revision 1.267 2017/05/13 10:25:05 brouard
346: Summary: temporary save for backprojection
347:
1.267 brouard 348: Revision 1.266 2017/05/13 07:26:12 brouard
349: Summary: Version 0.99r13 (improvements and bugs fixed)
350:
1.266 brouard 351: Revision 1.265 2017/04/26 16:22:11 brouard
352: Summary: imach 0.99r13 Some bugs fixed
353:
1.265 brouard 354: Revision 1.264 2017/04/26 06:01:29 brouard
355: Summary: Labels in graphs
356:
1.264 brouard 357: Revision 1.263 2017/04/24 15:23:15 brouard
358: Summary: to save
359:
1.263 brouard 360: Revision 1.262 2017/04/18 16:48:12 brouard
361: *** empty log message ***
362:
1.262 brouard 363: Revision 1.261 2017/04/05 10:14:09 brouard
364: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
365:
1.261 brouard 366: Revision 1.260 2017/04/04 17:46:59 brouard
367: Summary: Gnuplot indexations fixed (humm)
368:
1.260 brouard 369: Revision 1.259 2017/04/04 13:01:16 brouard
370: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
371:
1.259 brouard 372: Revision 1.258 2017/04/03 10:17:47 brouard
373: Summary: Version 0.99r12
374:
375: Some cleanings, conformed with updated documentation.
376:
1.258 brouard 377: Revision 1.257 2017/03/29 16:53:30 brouard
378: Summary: Temp
379:
1.257 brouard 380: Revision 1.256 2017/03/27 05:50:23 brouard
381: Summary: Temporary
382:
1.256 brouard 383: Revision 1.255 2017/03/08 16:02:28 brouard
384: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
385:
1.255 brouard 386: Revision 1.254 2017/03/08 07:13:00 brouard
387: Summary: Fixing data parameter line
388:
1.254 brouard 389: Revision 1.253 2016/12/15 11:59:41 brouard
390: Summary: 0.99 in progress
391:
1.253 brouard 392: Revision 1.252 2016/09/15 21:15:37 brouard
393: *** empty log message ***
394:
1.252 brouard 395: Revision 1.251 2016/09/15 15:01:13 brouard
396: Summary: not working
397:
1.251 brouard 398: Revision 1.250 2016/09/08 16:07:27 brouard
399: Summary: continue
400:
1.250 brouard 401: Revision 1.249 2016/09/07 17:14:18 brouard
402: Summary: Starting values from frequencies
403:
1.249 brouard 404: Revision 1.248 2016/09/07 14:10:18 brouard
405: *** empty log message ***
406:
1.248 brouard 407: Revision 1.247 2016/09/02 11:11:21 brouard
408: *** empty log message ***
409:
1.247 brouard 410: Revision 1.246 2016/09/02 08:49:22 brouard
411: *** empty log message ***
412:
1.246 brouard 413: Revision 1.245 2016/09/02 07:25:01 brouard
414: *** empty log message ***
415:
1.245 brouard 416: Revision 1.244 2016/09/02 07:17:34 brouard
417: *** empty log message ***
418:
1.244 brouard 419: Revision 1.243 2016/09/02 06:45:35 brouard
420: *** empty log message ***
421:
1.243 brouard 422: Revision 1.242 2016/08/30 15:01:20 brouard
423: Summary: Fixing a lots
424:
1.242 brouard 425: Revision 1.241 2016/08/29 17:17:25 brouard
426: Summary: gnuplot problem in Back projection to fix
427:
1.241 brouard 428: Revision 1.240 2016/08/29 07:53:18 brouard
429: Summary: Better
430:
1.240 brouard 431: Revision 1.239 2016/08/26 15:51:03 brouard
432: Summary: Improvement in Powell output in order to copy and paste
433:
434: Author:
435:
1.239 brouard 436: Revision 1.238 2016/08/26 14:23:35 brouard
437: Summary: Starting tests of 0.99
438:
1.238 brouard 439: Revision 1.237 2016/08/26 09:20:19 brouard
440: Summary: to valgrind
441:
1.237 brouard 442: Revision 1.236 2016/08/25 10:50:18 brouard
443: *** empty log message ***
444:
1.236 brouard 445: Revision 1.235 2016/08/25 06:59:23 brouard
446: *** empty log message ***
447:
1.235 brouard 448: Revision 1.234 2016/08/23 16:51:20 brouard
449: *** empty log message ***
450:
1.234 brouard 451: Revision 1.233 2016/08/23 07:40:50 brouard
452: Summary: not working
453:
1.233 brouard 454: Revision 1.232 2016/08/22 14:20:21 brouard
455: Summary: not working
456:
1.232 brouard 457: Revision 1.231 2016/08/22 07:17:15 brouard
458: Summary: not working
459:
1.231 brouard 460: Revision 1.230 2016/08/22 06:55:53 brouard
461: Summary: Not working
462:
1.230 brouard 463: Revision 1.229 2016/07/23 09:45:53 brouard
464: Summary: Completing for func too
465:
1.229 brouard 466: Revision 1.228 2016/07/22 17:45:30 brouard
467: Summary: Fixing some arrays, still debugging
468:
1.227 brouard 469: Revision 1.226 2016/07/12 18:42:34 brouard
470: Summary: temp
471:
1.226 brouard 472: Revision 1.225 2016/07/12 08:40:03 brouard
473: Summary: saving but not running
474:
1.225 brouard 475: Revision 1.224 2016/07/01 13:16:01 brouard
476: Summary: Fixes
477:
1.224 brouard 478: Revision 1.223 2016/02/19 09:23:35 brouard
479: Summary: temporary
480:
1.223 brouard 481: Revision 1.222 2016/02/17 08:14:50 brouard
482: Summary: Probably last 0.98 stable version 0.98r6
483:
1.222 brouard 484: Revision 1.221 2016/02/15 23:35:36 brouard
485: Summary: minor bug
486:
1.220 brouard 487: Revision 1.219 2016/02/15 00:48:12 brouard
488: *** empty log message ***
489:
1.219 brouard 490: Revision 1.218 2016/02/12 11:29:23 brouard
491: Summary: 0.99 Back projections
492:
1.218 brouard 493: Revision 1.217 2015/12/23 17:18:31 brouard
494: Summary: Experimental backcast
495:
1.217 brouard 496: Revision 1.216 2015/12/18 17:32:11 brouard
497: Summary: 0.98r4 Warning and status=-2
498:
499: Version 0.98r4 is now:
500: - displaying an error when status is -1, date of interview unknown and date of death known;
501: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
502: Older changes concerning s=-2, dating from 2005 have been supersed.
503:
1.216 brouard 504: Revision 1.215 2015/12/16 08:52:24 brouard
505: Summary: 0.98r4 working
506:
1.215 brouard 507: Revision 1.214 2015/12/16 06:57:54 brouard
508: Summary: temporary not working
509:
1.214 brouard 510: Revision 1.213 2015/12/11 18:22:17 brouard
511: Summary: 0.98r4
512:
1.213 brouard 513: Revision 1.212 2015/11/21 12:47:24 brouard
514: Summary: minor typo
515:
1.212 brouard 516: Revision 1.211 2015/11/21 12:41:11 brouard
517: Summary: 0.98r3 with some graph of projected cross-sectional
518:
519: Author: Nicolas Brouard
520:
1.211 brouard 521: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 522: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 523: Summary: Adding ftolpl parameter
524: Author: N Brouard
525:
526: We had difficulties to get smoothed confidence intervals. It was due
527: to the period prevalence which wasn't computed accurately. The inner
528: parameter ftolpl is now an outer parameter of the .imach parameter
529: file after estepm. If ftolpl is small 1.e-4 and estepm too,
530: computation are long.
531:
1.209 brouard 532: Revision 1.208 2015/11/17 14:31:57 brouard
533: Summary: temporary
534:
1.208 brouard 535: Revision 1.207 2015/10/27 17:36:57 brouard
536: *** empty log message ***
537:
1.207 brouard 538: Revision 1.206 2015/10/24 07:14:11 brouard
539: *** empty log message ***
540:
1.206 brouard 541: Revision 1.205 2015/10/23 15:50:53 brouard
542: Summary: 0.98r3 some clarification for graphs on likelihood contributions
543:
1.205 brouard 544: Revision 1.204 2015/10/01 16:20:26 brouard
545: Summary: Some new graphs of contribution to likelihood
546:
1.204 brouard 547: Revision 1.203 2015/09/30 17:45:14 brouard
548: Summary: looking at better estimation of the hessian
549:
550: Also a better criteria for convergence to the period prevalence And
551: therefore adding the number of years needed to converge. (The
552: prevalence in any alive state shold sum to one
553:
1.203 brouard 554: Revision 1.202 2015/09/22 19:45:16 brouard
555: Summary: Adding some overall graph on contribution to likelihood. Might change
556:
1.202 brouard 557: Revision 1.201 2015/09/15 17:34:58 brouard
558: Summary: 0.98r0
559:
560: - Some new graphs like suvival functions
561: - Some bugs fixed like model=1+age+V2.
562:
1.201 brouard 563: Revision 1.200 2015/09/09 16:53:55 brouard
564: Summary: Big bug thanks to Flavia
565:
566: Even model=1+age+V2. did not work anymore
567:
1.200 brouard 568: Revision 1.199 2015/09/07 14:09:23 brouard
569: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
570:
1.199 brouard 571: Revision 1.198 2015/09/03 07:14:39 brouard
572: Summary: 0.98q5 Flavia
573:
1.198 brouard 574: Revision 1.197 2015/09/01 18:24:39 brouard
575: *** empty log message ***
576:
1.197 brouard 577: Revision 1.196 2015/08/18 23:17:52 brouard
578: Summary: 0.98q5
579:
1.196 brouard 580: Revision 1.195 2015/08/18 16:28:39 brouard
581: Summary: Adding a hack for testing purpose
582:
583: After reading the title, ftol and model lines, if the comment line has
584: a q, starting with #q, the answer at the end of the run is quit. It
585: permits to run test files in batch with ctest. The former workaround was
586: $ echo q | imach foo.imach
587:
1.195 brouard 588: Revision 1.194 2015/08/18 13:32:00 brouard
589: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
590:
1.194 brouard 591: Revision 1.193 2015/08/04 07:17:42 brouard
592: Summary: 0.98q4
593:
1.193 brouard 594: Revision 1.192 2015/07/16 16:49:02 brouard
595: Summary: Fixing some outputs
596:
1.192 brouard 597: Revision 1.191 2015/07/14 10:00:33 brouard
598: Summary: Some fixes
599:
1.191 brouard 600: Revision 1.190 2015/05/05 08:51:13 brouard
601: Summary: Adding digits in output parameters (7 digits instead of 6)
602:
603: Fix 1+age+.
604:
1.190 brouard 605: Revision 1.189 2015/04/30 14:45:16 brouard
606: Summary: 0.98q2
607:
1.189 brouard 608: Revision 1.188 2015/04/30 08:27:53 brouard
609: *** empty log message ***
610:
1.188 brouard 611: Revision 1.187 2015/04/29 09:11:15 brouard
612: *** empty log message ***
613:
1.187 brouard 614: Revision 1.186 2015/04/23 12:01:52 brouard
615: Summary: V1*age is working now, version 0.98q1
616:
617: Some codes had been disabled in order to simplify and Vn*age was
618: working in the optimization phase, ie, giving correct MLE parameters,
619: but, as usual, outputs were not correct and program core dumped.
620:
1.186 brouard 621: Revision 1.185 2015/03/11 13:26:42 brouard
622: Summary: Inclusion of compile and links command line for Intel Compiler
623:
1.185 brouard 624: Revision 1.184 2015/03/11 11:52:39 brouard
625: Summary: Back from Windows 8. Intel Compiler
626:
1.184 brouard 627: Revision 1.183 2015/03/10 20:34:32 brouard
628: Summary: 0.98q0, trying with directest, mnbrak fixed
629:
630: We use directest instead of original Powell test; probably no
631: incidence on the results, but better justifications;
632: We fixed Numerical Recipes mnbrak routine which was wrong and gave
633: wrong results.
634:
1.183 brouard 635: Revision 1.182 2015/02/12 08:19:57 brouard
636: Summary: Trying to keep directest which seems simpler and more general
637: Author: Nicolas Brouard
638:
1.182 brouard 639: Revision 1.181 2015/02/11 23:22:24 brouard
640: Summary: Comments on Powell added
641:
642: Author:
643:
1.181 brouard 644: Revision 1.180 2015/02/11 17:33:45 brouard
645: Summary: Finishing move from main to function (hpijx and prevalence_limit)
646:
1.180 brouard 647: Revision 1.179 2015/01/04 09:57:06 brouard
648: Summary: back to OS/X
649:
1.179 brouard 650: Revision 1.178 2015/01/04 09:35:48 brouard
651: *** empty log message ***
652:
1.178 brouard 653: Revision 1.177 2015/01/03 18:40:56 brouard
654: Summary: Still testing ilc32 on OSX
655:
1.177 brouard 656: Revision 1.176 2015/01/03 16:45:04 brouard
657: *** empty log message ***
658:
1.176 brouard 659: Revision 1.175 2015/01/03 16:33:42 brouard
660: *** empty log message ***
661:
1.175 brouard 662: Revision 1.174 2015/01/03 16:15:49 brouard
663: Summary: Still in cross-compilation
664:
1.174 brouard 665: Revision 1.173 2015/01/03 12:06:26 brouard
666: Summary: trying to detect cross-compilation
667:
1.173 brouard 668: Revision 1.172 2014/12/27 12:07:47 brouard
669: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
670:
1.172 brouard 671: Revision 1.171 2014/12/23 13:26:59 brouard
672: Summary: Back from Visual C
673:
674: Still problem with utsname.h on Windows
675:
1.171 brouard 676: Revision 1.170 2014/12/23 11:17:12 brouard
677: Summary: Cleaning some \%% back to %%
678:
679: The escape was mandatory for a specific compiler (which one?), but too many warnings.
680:
1.170 brouard 681: Revision 1.169 2014/12/22 23:08:31 brouard
682: Summary: 0.98p
683:
684: Outputs some informations on compiler used, OS etc. Testing on different platforms.
685:
1.169 brouard 686: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 687: Summary: update
1.169 brouard 688:
1.168 brouard 689: Revision 1.167 2014/12/22 13:50:56 brouard
690: Summary: Testing uname and compiler version and if compiled 32 or 64
691:
692: Testing on Linux 64
693:
1.167 brouard 694: Revision 1.166 2014/12/22 11:40:47 brouard
695: *** empty log message ***
696:
1.166 brouard 697: Revision 1.165 2014/12/16 11:20:36 brouard
698: Summary: After compiling on Visual C
699:
700: * imach.c (Module): Merging 1.61 to 1.162
701:
1.165 brouard 702: Revision 1.164 2014/12/16 10:52:11 brouard
703: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
704:
705: * imach.c (Module): Merging 1.61 to 1.162
706:
1.164 brouard 707: Revision 1.163 2014/12/16 10:30:11 brouard
708: * imach.c (Module): Merging 1.61 to 1.162
709:
1.163 brouard 710: Revision 1.162 2014/09/25 11:43:39 brouard
711: Summary: temporary backup 0.99!
712:
1.162 brouard 713: Revision 1.1 2014/09/16 11:06:58 brouard
714: Summary: With some code (wrong) for nlopt
715:
716: Author:
717:
718: Revision 1.161 2014/09/15 20:41:41 brouard
719: Summary: Problem with macro SQR on Intel compiler
720:
1.161 brouard 721: Revision 1.160 2014/09/02 09:24:05 brouard
722: *** empty log message ***
723:
1.160 brouard 724: Revision 1.159 2014/09/01 10:34:10 brouard
725: Summary: WIN32
726: Author: Brouard
727:
1.159 brouard 728: Revision 1.158 2014/08/27 17:11:51 brouard
729: *** empty log message ***
730:
1.158 brouard 731: Revision 1.157 2014/08/27 16:26:55 brouard
732: Summary: Preparing windows Visual studio version
733: Author: Brouard
734:
735: In order to compile on Visual studio, time.h is now correct and time_t
736: and tm struct should be used. difftime should be used but sometimes I
737: just make the differences in raw time format (time(&now).
738: Trying to suppress #ifdef LINUX
739: Add xdg-open for __linux in order to open default browser.
740:
1.157 brouard 741: Revision 1.156 2014/08/25 20:10:10 brouard
742: *** empty log message ***
743:
1.156 brouard 744: Revision 1.155 2014/08/25 18:32:34 brouard
745: Summary: New compile, minor changes
746: Author: Brouard
747:
1.155 brouard 748: Revision 1.154 2014/06/20 17:32:08 brouard
749: Summary: Outputs now all graphs of convergence to period prevalence
750:
1.154 brouard 751: Revision 1.153 2014/06/20 16:45:46 brouard
752: Summary: If 3 live state, convergence to period prevalence on same graph
753: Author: Brouard
754:
1.153 brouard 755: Revision 1.152 2014/06/18 17:54:09 brouard
756: Summary: open browser, use gnuplot on same dir than imach if not found in the path
757:
1.152 brouard 758: Revision 1.151 2014/06/18 16:43:30 brouard
759: *** empty log message ***
760:
1.151 brouard 761: Revision 1.150 2014/06/18 16:42:35 brouard
762: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
763: Author: brouard
764:
1.150 brouard 765: Revision 1.149 2014/06/18 15:51:14 brouard
766: Summary: Some fixes in parameter files errors
767: Author: Nicolas Brouard
768:
1.149 brouard 769: Revision 1.148 2014/06/17 17:38:48 brouard
770: Summary: Nothing new
771: Author: Brouard
772:
773: Just a new packaging for OS/X version 0.98nS
774:
1.148 brouard 775: Revision 1.147 2014/06/16 10:33:11 brouard
776: *** empty log message ***
777:
1.147 brouard 778: Revision 1.146 2014/06/16 10:20:28 brouard
779: Summary: Merge
780: Author: Brouard
781:
782: Merge, before building revised version.
783:
1.146 brouard 784: Revision 1.145 2014/06/10 21:23:15 brouard
785: Summary: Debugging with valgrind
786: Author: Nicolas Brouard
787:
788: Lot of changes in order to output the results with some covariates
789: After the Edimburgh REVES conference 2014, it seems mandatory to
790: improve the code.
791: No more memory valgrind error but a lot has to be done in order to
792: continue the work of splitting the code into subroutines.
793: Also, decodemodel has been improved. Tricode is still not
794: optimal. nbcode should be improved. Documentation has been added in
795: the source code.
796:
1.144 brouard 797: Revision 1.143 2014/01/26 09:45:38 brouard
798: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
799:
800: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
801: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
802:
1.143 brouard 803: Revision 1.142 2014/01/26 03:57:36 brouard
804: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
805:
806: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
807:
1.142 brouard 808: Revision 1.141 2014/01/26 02:42:01 brouard
809: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
810:
1.141 brouard 811: Revision 1.140 2011/09/02 10:37:54 brouard
812: Summary: times.h is ok with mingw32 now.
813:
1.140 brouard 814: Revision 1.139 2010/06/14 07:50:17 brouard
815: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
816: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
817:
1.139 brouard 818: Revision 1.138 2010/04/30 18:19:40 brouard
819: *** empty log message ***
820:
1.138 brouard 821: Revision 1.137 2010/04/29 18:11:38 brouard
822: (Module): Checking covariates for more complex models
823: than V1+V2. A lot of change to be done. Unstable.
824:
1.137 brouard 825: Revision 1.136 2010/04/26 20:30:53 brouard
826: (Module): merging some libgsl code. Fixing computation
827: of likelione (using inter/intrapolation if mle = 0) in order to
828: get same likelihood as if mle=1.
829: Some cleaning of code and comments added.
830:
1.136 brouard 831: Revision 1.135 2009/10/29 15:33:14 brouard
832: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
833:
1.135 brouard 834: Revision 1.134 2009/10/29 13:18:53 brouard
835: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
836:
1.134 brouard 837: Revision 1.133 2009/07/06 10:21:25 brouard
838: just nforces
839:
1.133 brouard 840: Revision 1.132 2009/07/06 08:22:05 brouard
841: Many tings
842:
1.132 brouard 843: Revision 1.131 2009/06/20 16:22:47 brouard
844: Some dimensions resccaled
845:
1.131 brouard 846: Revision 1.130 2009/05/26 06:44:34 brouard
847: (Module): Max Covariate is now set to 20 instead of 8. A
848: lot of cleaning with variables initialized to 0. Trying to make
849: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
850:
1.130 brouard 851: Revision 1.129 2007/08/31 13:49:27 lievre
852: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
853:
1.129 lievre 854: Revision 1.128 2006/06/30 13:02:05 brouard
855: (Module): Clarifications on computing e.j
856:
1.128 brouard 857: Revision 1.127 2006/04/28 18:11:50 brouard
858: (Module): Yes the sum of survivors was wrong since
859: imach-114 because nhstepm was no more computed in the age
860: loop. Now we define nhstepma in the age loop.
861: (Module): In order to speed up (in case of numerous covariates) we
862: compute health expectancies (without variances) in a first step
863: and then all the health expectancies with variances or standard
864: deviation (needs data from the Hessian matrices) which slows the
865: computation.
866: In the future we should be able to stop the program is only health
867: expectancies and graph are needed without standard deviations.
868:
1.127 brouard 869: Revision 1.126 2006/04/28 17:23:28 brouard
870: (Module): Yes the sum of survivors was wrong since
871: imach-114 because nhstepm was no more computed in the age
872: loop. Now we define nhstepma in the age loop.
873: Version 0.98h
874:
1.126 brouard 875: Revision 1.125 2006/04/04 15:20:31 lievre
876: Errors in calculation of health expectancies. Age was not initialized.
877: Forecasting file added.
878:
879: Revision 1.124 2006/03/22 17:13:53 lievre
880: Parameters are printed with %lf instead of %f (more numbers after the comma).
881: The log-likelihood is printed in the log file
882:
883: Revision 1.123 2006/03/20 10:52:43 brouard
884: * imach.c (Module): <title> changed, corresponds to .htm file
885: name. <head> headers where missing.
886:
887: * imach.c (Module): Weights can have a decimal point as for
888: English (a comma might work with a correct LC_NUMERIC environment,
889: otherwise the weight is truncated).
890: Modification of warning when the covariates values are not 0 or
891: 1.
892: Version 0.98g
893:
894: Revision 1.122 2006/03/20 09:45:41 brouard
895: (Module): Weights can have a decimal point as for
896: English (a comma might work with a correct LC_NUMERIC environment,
897: otherwise the weight is truncated).
898: Modification of warning when the covariates values are not 0 or
899: 1.
900: Version 0.98g
901:
902: Revision 1.121 2006/03/16 17:45:01 lievre
903: * imach.c (Module): Comments concerning covariates added
904:
905: * imach.c (Module): refinements in the computation of lli if
906: status=-2 in order to have more reliable computation if stepm is
907: not 1 month. Version 0.98f
908:
909: Revision 1.120 2006/03/16 15:10:38 lievre
910: (Module): refinements in the computation of lli if
911: status=-2 in order to have more reliable computation if stepm is
912: not 1 month. Version 0.98f
913:
914: Revision 1.119 2006/03/15 17:42:26 brouard
915: (Module): Bug if status = -2, the loglikelihood was
916: computed as likelihood omitting the logarithm. Version O.98e
917:
918: Revision 1.118 2006/03/14 18:20:07 brouard
919: (Module): varevsij Comments added explaining the second
920: table of variances if popbased=1 .
921: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
922: (Module): Function pstamp added
923: (Module): Version 0.98d
924:
925: Revision 1.117 2006/03/14 17:16:22 brouard
926: (Module): varevsij Comments added explaining the second
927: table of variances if popbased=1 .
928: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
929: (Module): Function pstamp added
930: (Module): Version 0.98d
931:
932: Revision 1.116 2006/03/06 10:29:27 brouard
933: (Module): Variance-covariance wrong links and
934: varian-covariance of ej. is needed (Saito).
935:
936: Revision 1.115 2006/02/27 12:17:45 brouard
937: (Module): One freematrix added in mlikeli! 0.98c
938:
939: Revision 1.114 2006/02/26 12:57:58 brouard
940: (Module): Some improvements in processing parameter
941: filename with strsep.
942:
943: Revision 1.113 2006/02/24 14:20:24 brouard
944: (Module): Memory leaks checks with valgrind and:
945: datafile was not closed, some imatrix were not freed and on matrix
946: allocation too.
947:
948: Revision 1.112 2006/01/30 09:55:26 brouard
949: (Module): Back to gnuplot.exe instead of wgnuplot.exe
950:
951: Revision 1.111 2006/01/25 20:38:18 brouard
952: (Module): Lots of cleaning and bugs added (Gompertz)
953: (Module): Comments can be added in data file. Missing date values
954: can be a simple dot '.'.
955:
956: Revision 1.110 2006/01/25 00:51:50 brouard
957: (Module): Lots of cleaning and bugs added (Gompertz)
958:
959: Revision 1.109 2006/01/24 19:37:15 brouard
960: (Module): Comments (lines starting with a #) are allowed in data.
961:
962: Revision 1.108 2006/01/19 18:05:42 lievre
963: Gnuplot problem appeared...
964: To be fixed
965:
966: Revision 1.107 2006/01/19 16:20:37 brouard
967: Test existence of gnuplot in imach path
968:
969: Revision 1.106 2006/01/19 13:24:36 brouard
970: Some cleaning and links added in html output
971:
972: Revision 1.105 2006/01/05 20:23:19 lievre
973: *** empty log message ***
974:
975: Revision 1.104 2005/09/30 16:11:43 lievre
976: (Module): sump fixed, loop imx fixed, and simplifications.
977: (Module): If the status is missing at the last wave but we know
978: that the person is alive, then we can code his/her status as -2
979: (instead of missing=-1 in earlier versions) and his/her
980: contributions to the likelihood is 1 - Prob of dying from last
981: health status (= 1-p13= p11+p12 in the easiest case of somebody in
982: the healthy state at last known wave). Version is 0.98
983:
984: Revision 1.103 2005/09/30 15:54:49 lievre
985: (Module): sump fixed, loop imx fixed, and simplifications.
986:
987: Revision 1.102 2004/09/15 17:31:30 brouard
988: Add the possibility to read data file including tab characters.
989:
990: Revision 1.101 2004/09/15 10:38:38 brouard
991: Fix on curr_time
992:
993: Revision 1.100 2004/07/12 18:29:06 brouard
994: Add version for Mac OS X. Just define UNIX in Makefile
995:
996: Revision 1.99 2004/06/05 08:57:40 brouard
997: *** empty log message ***
998:
999: Revision 1.98 2004/05/16 15:05:56 brouard
1000: New version 0.97 . First attempt to estimate force of mortality
1001: directly from the data i.e. without the need of knowing the health
1002: state at each age, but using a Gompertz model: log u =a + b*age .
1003: This is the basic analysis of mortality and should be done before any
1004: other analysis, in order to test if the mortality estimated from the
1005: cross-longitudinal survey is different from the mortality estimated
1006: from other sources like vital statistic data.
1007:
1008: The same imach parameter file can be used but the option for mle should be -3.
1009:
1.324 brouard 1010: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 1011: former routines in order to include the new code within the former code.
1012:
1013: The output is very simple: only an estimate of the intercept and of
1014: the slope with 95% confident intervals.
1015:
1016: Current limitations:
1017: A) Even if you enter covariates, i.e. with the
1018: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
1019: B) There is no computation of Life Expectancy nor Life Table.
1020:
1021: Revision 1.97 2004/02/20 13:25:42 lievre
1022: Version 0.96d. Population forecasting command line is (temporarily)
1023: suppressed.
1024:
1025: Revision 1.96 2003/07/15 15:38:55 brouard
1026: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
1027: rewritten within the same printf. Workaround: many printfs.
1028:
1029: Revision 1.95 2003/07/08 07:54:34 brouard
1030: * imach.c (Repository):
1031: (Repository): Using imachwizard code to output a more meaningful covariance
1032: matrix (cov(a12,c31) instead of numbers.
1033:
1034: Revision 1.94 2003/06/27 13:00:02 brouard
1035: Just cleaning
1036:
1037: Revision 1.93 2003/06/25 16:33:55 brouard
1038: (Module): On windows (cygwin) function asctime_r doesn't
1039: exist so I changed back to asctime which exists.
1040: (Module): Version 0.96b
1041:
1042: Revision 1.92 2003/06/25 16:30:45 brouard
1043: (Module): On windows (cygwin) function asctime_r doesn't
1044: exist so I changed back to asctime which exists.
1045:
1046: Revision 1.91 2003/06/25 15:30:29 brouard
1047: * imach.c (Repository): Duplicated warning errors corrected.
1048: (Repository): Elapsed time after each iteration is now output. It
1049: helps to forecast when convergence will be reached. Elapsed time
1050: is stamped in powell. We created a new html file for the graphs
1051: concerning matrix of covariance. It has extension -cov.htm.
1052:
1053: Revision 1.90 2003/06/24 12:34:15 brouard
1054: (Module): Some bugs corrected for windows. Also, when
1055: mle=-1 a template is output in file "or"mypar.txt with the design
1056: of the covariance matrix to be input.
1057:
1058: Revision 1.89 2003/06/24 12:30:52 brouard
1059: (Module): Some bugs corrected for windows. Also, when
1060: mle=-1 a template is output in file "or"mypar.txt with the design
1061: of the covariance matrix to be input.
1062:
1063: Revision 1.88 2003/06/23 17:54:56 brouard
1064: * imach.c (Repository): Create a sub-directory where all the secondary files are. Only imach, htm, gp and r(imach) are on the main directory. Correct time and other things.
1065:
1066: Revision 1.87 2003/06/18 12:26:01 brouard
1067: Version 0.96
1068:
1069: Revision 1.86 2003/06/17 20:04:08 brouard
1070: (Module): Change position of html and gnuplot routines and added
1071: routine fileappend.
1072:
1073: Revision 1.85 2003/06/17 13:12:43 brouard
1074: * imach.c (Repository): Check when date of death was earlier that
1075: current date of interview. It may happen when the death was just
1076: prior to the death. In this case, dh was negative and likelihood
1077: was wrong (infinity). We still send an "Error" but patch by
1078: assuming that the date of death was just one stepm after the
1079: interview.
1080: (Repository): Because some people have very long ID (first column)
1081: we changed int to long in num[] and we added a new lvector for
1082: memory allocation. But we also truncated to 8 characters (left
1083: truncation)
1084: (Repository): No more line truncation errors.
1085:
1086: Revision 1.84 2003/06/13 21:44:43 brouard
1087: * imach.c (Repository): Replace "freqsummary" at a correct
1088: place. It differs from routine "prevalence" which may be called
1089: many times. Probs is memory consuming and must be used with
1090: parcimony.
1091: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1092:
1093: Revision 1.83 2003/06/10 13:39:11 lievre
1094: *** empty log message ***
1095:
1096: Revision 1.82 2003/06/05 15:57:20 brouard
1097: Add log in imach.c and fullversion number is now printed.
1098:
1099: */
1100: /*
1101: Interpolated Markov Chain
1102:
1103: Short summary of the programme:
1104:
1.227 brouard 1105: This program computes Healthy Life Expectancies or State-specific
1106: (if states aren't health statuses) Expectancies from
1107: cross-longitudinal data. Cross-longitudinal data consist in:
1108:
1109: -1- a first survey ("cross") where individuals from different ages
1110: are interviewed on their health status or degree of disability (in
1111: the case of a health survey which is our main interest)
1112:
1113: -2- at least a second wave of interviews ("longitudinal") which
1114: measure each change (if any) in individual health status. Health
1115: expectancies are computed from the time spent in each health state
1116: according to a model. More health states you consider, more time is
1117: necessary to reach the Maximum Likelihood of the parameters involved
1118: in the model. The simplest model is the multinomial logistic model
1119: where pij is the probability to be observed in state j at the second
1120: wave conditional to be observed in state i at the first
1121: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1122: etc , where 'age' is age and 'sex' is a covariate. If you want to
1123: have a more complex model than "constant and age", you should modify
1124: the program where the markup *Covariates have to be included here
1125: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1126: convergence.
1127:
1128: The advantage of this computer programme, compared to a simple
1129: multinomial logistic model, is clear when the delay between waves is not
1130: identical for each individual. Also, if a individual missed an
1131: intermediate interview, the information is lost, but taken into
1132: account using an interpolation or extrapolation.
1133:
1134: hPijx is the probability to be observed in state i at age x+h
1135: conditional to the observed state i at age x. The delay 'h' can be
1136: split into an exact number (nh*stepm) of unobserved intermediate
1137: states. This elementary transition (by month, quarter,
1138: semester or year) is modelled as a multinomial logistic. The hPx
1139: matrix is simply the matrix product of nh*stepm elementary matrices
1140: and the contribution of each individual to the likelihood is simply
1141: hPijx.
1142:
1143: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1144: of the life expectancies. It also computes the period (stable) prevalence.
1145:
1146: Back prevalence and projections:
1.227 brouard 1147:
1148: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1149: double agemaxpar, double ftolpl, int *ncvyearp, double
1150: dateprev1,double dateprev2, int firstpass, int lastpass, int
1151: mobilavproj)
1152:
1153: Computes the back prevalence limit for any combination of
1154: covariate values k at any age between ageminpar and agemaxpar and
1155: returns it in **bprlim. In the loops,
1156:
1157: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1158: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1159:
1160: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1161: Computes for any combination of covariates k and any age between bage and fage
1162: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1163: oldm=oldms;savm=savms;
1.227 brouard 1164:
1.267 brouard 1165: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1166: Computes the transition matrix starting at age 'age' over
1167: 'nhstepm*hstepm*stepm' months (i.e. until
1168: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1169: nhstepm*hstepm matrices.
1170:
1171: Returns p3mat[i][j][h] after calling
1172: p3mat[i][j][h]=matprod2(newm,
1173: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1174: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1175: oldm);
1.226 brouard 1176:
1177: Important routines
1178:
1179: - func (or funcone), computes logit (pij) distinguishing
1180: o fixed variables (single or product dummies or quantitative);
1181: o varying variables by:
1182: (1) wave (single, product dummies, quantitative),
1183: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1184: % fixed dummy (treated) or quantitative (not done because time-consuming);
1185: % varying dummy (not done) or quantitative (not done);
1186: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1187: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1188: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1189: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1190: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1191:
1.226 brouard 1192:
1193:
1.324 brouard 1194: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1195: Institut national d'études démographiques, Paris.
1.126 brouard 1196: This software have been partly granted by Euro-REVES, a concerted action
1197: from the European Union.
1198: It is copyrighted identically to a GNU software product, ie programme and
1199: software can be distributed freely for non commercial use. Latest version
1200: can be accessed at http://euroreves.ined.fr/imach .
1201:
1202: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1203: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1204:
1205: **********************************************************************/
1206: /*
1207: main
1208: read parameterfile
1209: read datafile
1210: concatwav
1211: freqsummary
1212: if (mle >= 1)
1213: mlikeli
1214: print results files
1215: if mle==1
1216: computes hessian
1217: read end of parameter file: agemin, agemax, bage, fage, estepm
1218: begin-prev-date,...
1219: open gnuplot file
1220: open html file
1.145 brouard 1221: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1222: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1223: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1224: freexexit2 possible for memory heap.
1225:
1226: h Pij x | pij_nom ficrestpij
1227: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1228: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1229: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1230:
1231: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1232: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1233: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1234: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1235: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1236:
1.126 brouard 1237: forecasting if prevfcast==1 prevforecast call prevalence()
1238: health expectancies
1239: Variance-covariance of DFLE
1240: prevalence()
1241: movingaverage()
1242: varevsij()
1243: if popbased==1 varevsij(,popbased)
1244: total life expectancies
1245: Variance of period (stable) prevalence
1246: end
1247: */
1248:
1.187 brouard 1249: /* #define DEBUG */
1250: /* #define DEBUGBRENT */
1.203 brouard 1251: /* #define DEBUGLINMIN */
1252: /* #define DEBUGHESS */
1253: #define DEBUGHESSIJ
1.224 brouard 1254: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1255: #define POWELL /* Instead of NLOPT */
1.224 brouard 1256: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1257: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1258: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1259: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1260:
1261: #include <math.h>
1262: #include <stdio.h>
1263: #include <stdlib.h>
1264: #include <string.h>
1.226 brouard 1265: #include <ctype.h>
1.159 brouard 1266:
1267: #ifdef _WIN32
1268: #include <io.h>
1.172 brouard 1269: #include <windows.h>
1270: #include <tchar.h>
1.159 brouard 1271: #else
1.126 brouard 1272: #include <unistd.h>
1.159 brouard 1273: #endif
1.126 brouard 1274:
1275: #include <limits.h>
1276: #include <sys/types.h>
1.171 brouard 1277:
1278: #if defined(__GNUC__)
1279: #include <sys/utsname.h> /* Doesn't work on Windows */
1280: #endif
1281:
1.126 brouard 1282: #include <sys/stat.h>
1283: #include <errno.h>
1.159 brouard 1284: /* extern int errno; */
1.126 brouard 1285:
1.157 brouard 1286: /* #ifdef LINUX */
1287: /* #include <time.h> */
1288: /* #include "timeval.h" */
1289: /* #else */
1290: /* #include <sys/time.h> */
1291: /* #endif */
1292:
1.126 brouard 1293: #include <time.h>
1294:
1.136 brouard 1295: #ifdef GSL
1296: #include <gsl/gsl_errno.h>
1297: #include <gsl/gsl_multimin.h>
1298: #endif
1299:
1.167 brouard 1300:
1.162 brouard 1301: #ifdef NLOPT
1302: #include <nlopt.h>
1303: typedef struct {
1304: double (* function)(double [] );
1305: } myfunc_data ;
1306: #endif
1307:
1.126 brouard 1308: /* #include <libintl.h> */
1309: /* #define _(String) gettext (String) */
1310:
1.251 brouard 1311: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1312:
1313: #define GNUPLOTPROGRAM "gnuplot"
1.343 brouard 1314: #define GNUPLOTVERSION 5.1
1315: double gnuplotversion=GNUPLOTVERSION;
1.126 brouard 1316: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1317: #define FILENAMELENGTH 256
1.126 brouard 1318:
1319: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1320: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1321:
1.144 brouard 1322: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1323: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1324:
1325: #define NINTERVMAX 8
1.144 brouard 1326: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1327: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1328: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1329: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1330: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1331: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1332: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1333: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1334: /* #define AGESUP 130 */
1.288 brouard 1335: /* #define AGESUP 150 */
1336: #define AGESUP 200
1.268 brouard 1337: #define AGEINF 0
1.218 brouard 1338: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1339: #define AGEBASE 40
1.194 brouard 1340: #define AGEOVERFLOW 1.e20
1.164 brouard 1341: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1342: #ifdef _WIN32
1343: #define DIRSEPARATOR '\\'
1344: #define CHARSEPARATOR "\\"
1345: #define ODIRSEPARATOR '/'
1346: #else
1.126 brouard 1347: #define DIRSEPARATOR '/'
1348: #define CHARSEPARATOR "/"
1349: #define ODIRSEPARATOR '\\'
1350: #endif
1351:
1.346 ! brouard 1352: /* $Id: imach.c,v 1.345 2022/09/16 13:40:11 brouard Exp $ */
1.126 brouard 1353: /* $State: Exp $ */
1.196 brouard 1354: #include "version.h"
1355: char version[]=__IMACH_VERSION__;
1.337 brouard 1356: char copyright[]="September 2022,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2020, Nihon University 2021-202, INED 2000-2022";
1.346 ! brouard 1357: char fullversion[]="$Revision: 1.345 $ $Date: 2022/09/16 13:40:11 $";
1.126 brouard 1358: char strstart[80];
1359: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1360: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.342 brouard 1361: int debugILK=0; /* debugILK is set by a #d in a comment line */
1.187 brouard 1362: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1363: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1364: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 brouard 1365: int cptcovn=0; /**< cptcovn decodemodel: number of covariates k of the models excluding age*products =6 and age*age but including products */
1.330 brouard 1366: int cptcovt=0; /**< cptcovt: total number of covariates of the model (2) nbocc(+)+1 = 8 excepting constant and age and age*age */
1.335 brouard 1367: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1368: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1369: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1370: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 brouard 1371: int cptcoveff=0; /* Total number of single dummy covariates (fixed or time varying) to vary for printing results (2**cptcoveff combinations of dummies)(computed in tricode as cptcov) */
1.233 brouard 1372: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1373: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.339 brouard 1374: int ncovvt=0; /* Total number of effective (wave) varying covariates (dummy or quantitative or products [without age]) in the model */
1.232 brouard 1375: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1376: int nsd=0; /**< Total number of single dummy variables (output) */
1377: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1378: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1379: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1380: int ntveff=0; /**< ntveff number of effective time varying variables */
1381: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1382: int cptcov=0; /* Working variable */
1.334 brouard 1383: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1384: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1385: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1386: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1387: int nlstate=2; /* Number of live states */
1388: int ndeath=1; /* Number of dead states */
1.130 brouard 1389: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.339 brouard 1390: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable*/
1391: int ncovcolt=0; /* ncovcolt=ncovcol+nqv+ntv+nqtv; total of covariates in the data, not in the model equation*/
1.126 brouard 1392: int popbased=0;
1393:
1394: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1395: int maxwav=0; /* Maxim number of waves */
1396: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1397: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1398: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1399: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1400: int mle=1, weightopt=0;
1.126 brouard 1401: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1402: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1403: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1404: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1405: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1406: int selected(int kvar); /* Is covariate kvar selected for printing results */
1407:
1.130 brouard 1408: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1409: double **matprod2(); /* test */
1.126 brouard 1410: double **oldm, **newm, **savm; /* Working pointers to matrices */
1411: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1412: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1413:
1.136 brouard 1414: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1415: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1416: FILE *ficlog, *ficrespow;
1.130 brouard 1417: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1418: double fretone; /* Only one call to likelihood */
1.130 brouard 1419: long ipmx=0; /* Number of contributions */
1.126 brouard 1420: double sw; /* Sum of weights */
1421: char filerespow[FILENAMELENGTH];
1422: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1423: FILE *ficresilk;
1424: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1425: FILE *ficresprobmorprev;
1426: FILE *fichtm, *fichtmcov; /* Html File */
1427: FILE *ficreseij;
1428: char filerese[FILENAMELENGTH];
1429: FILE *ficresstdeij;
1430: char fileresstde[FILENAMELENGTH];
1431: FILE *ficrescveij;
1432: char filerescve[FILENAMELENGTH];
1433: FILE *ficresvij;
1434: char fileresv[FILENAMELENGTH];
1.269 brouard 1435:
1.126 brouard 1436: char title[MAXLINE];
1.234 brouard 1437: char model[MAXLINE]; /**< The model line */
1.217 brouard 1438: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1439: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1440: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1441: char command[FILENAMELENGTH];
1442: int outcmd=0;
1443:
1.217 brouard 1444: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1445: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1446: char filelog[FILENAMELENGTH]; /* Log file */
1447: char filerest[FILENAMELENGTH];
1448: char fileregp[FILENAMELENGTH];
1449: char popfile[FILENAMELENGTH];
1450:
1451: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1452:
1.157 brouard 1453: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1454: /* struct timezone tzp; */
1455: /* extern int gettimeofday(); */
1456: struct tm tml, *gmtime(), *localtime();
1457:
1458: extern time_t time();
1459:
1460: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1461: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1462: struct tm tm;
1463:
1.126 brouard 1464: char strcurr[80], strfor[80];
1465:
1466: char *endptr;
1467: long lval;
1468: double dval;
1469:
1470: #define NR_END 1
1471: #define FREE_ARG char*
1472: #define FTOL 1.0e-10
1473:
1474: #define NRANSI
1.240 brouard 1475: #define ITMAX 200
1476: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1477:
1478: #define TOL 2.0e-4
1479:
1480: #define CGOLD 0.3819660
1481: #define ZEPS 1.0e-10
1482: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1483:
1484: #define GOLD 1.618034
1485: #define GLIMIT 100.0
1486: #define TINY 1.0e-20
1487:
1488: static double maxarg1,maxarg2;
1489: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1490: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1491:
1492: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1493: #define rint(a) floor(a+0.5)
1.166 brouard 1494: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1495: #define mytinydouble 1.0e-16
1.166 brouard 1496: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1497: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1498: /* static double dsqrarg; */
1499: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1500: static double sqrarg;
1501: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1502: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1503: int agegomp= AGEGOMP;
1504:
1505: int imx;
1506: int stepm=1;
1507: /* Stepm, step in month: minimum step interpolation*/
1508:
1509: int estepm;
1510: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1511:
1512: int m,nb;
1513: long *num;
1.197 brouard 1514: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1515: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1516: covariate for which somebody answered excluding
1517: undefined. Usually 2: 0 and 1. */
1518: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1519: covariate for which somebody answered including
1520: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1521: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1522: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1523: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1524: double **precov; /* New global variable to store for each resultline, values of model covariates given by the resultlines (in order to speed up) */
1.126 brouard 1525: double *ageexmed,*agecens;
1526: double dateintmean=0;
1.296 brouard 1527: double anprojd, mprojd, jprojd; /* For eventual projections */
1528: double anprojf, mprojf, jprojf;
1.126 brouard 1529:
1.296 brouard 1530: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1531: double anbackf, mbackf, jbackf;
1532: double jintmean,mintmean,aintmean;
1.126 brouard 1533: double *weight;
1534: int **s; /* Status */
1.141 brouard 1535: double *agedc;
1.145 brouard 1536: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1537: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1538: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1539: double **coqvar; /* Fixed quantitative covariate nqv */
1.341 brouard 1540: double ***cotvar; /* Time varying covariate start at ncovcol + nqv + (1 to ntv) */
1.225 brouard 1541: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1542: double idx;
1543: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1544: /* Some documentation */
1545: /* Design original data
1546: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1547: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1548: * ntv=3 nqtv=1
1.330 brouard 1549: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1550: * For time varying covariate, quanti or dummies
1551: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1.341 brouard 1552: * cotvar[wav][ncovcol+nqv+ iv(1 to nqtv)][i]= [(1 to nqtv)][i]=(V12) quanti
1.319 brouard 1553: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1554: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1555: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1556: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1557: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1558: * k= 1 2 3 4 5 6 7 8 9 10 11
1559: */
1560: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1561: /* ncovcol=1(Males=0 Females=1) nqv=1(raedyrs) ntv=2(withoutiadl=0 withiadl=1, witoutadl=0 withoutadl=1) nqtv=1(bmi) nlstate=3 ndeath=1
1562: # States 1=Coresidence, 2 Living alone, 3 Institution
1563: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1564: */
1.343 brouard 1565: /* V5+V4+ V3+V4*V3 +V5*age+V2 +V1*V2+V1*age+V1 */
1566: /* kmodel 1 2 3 4 5 6 7 8 9 */
1.319 brouard 1567: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1568: /* fixed or varying), 1 for age product, 2 for*/
1569: /* product */
1570: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1571: /*(single or product without age), 2 dummy*/
1572: /* with age product, 3 quant with age product*/
1573: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1574: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 brouard 1575: /*TnsdVar[Tvar] 1 2 3 */
1.337 brouard 1576: /*Tvaraff[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.319 brouard 1577: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.338 brouard 1578: /*TvarsDind[nsd] 2 3 9 */ /* position K of single dummy cova */
1.319 brouard 1579: /* nsq 1 2 */ /* Counting single quantit tv */
1580: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1581: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1582: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1583: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1584: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1585: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1586: /* Tvardk[4][1]=4;Tvardk[4][2]=3;Tvardk[7][1]=1;Tvardk[7][2]=2 */ /* Variables of a prod at position in the model equation*/
1.319 brouard 1587: /* TvarF TvarF[1]=Tvar[6]=2, TvarF[2]=Tvar[7]=7, TvarF[3]=Tvar[9]=1 ID of fixed covariates or product V2, V1*V2, V1 */
1.320 brouard 1588: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1589: /* Type */
1590: /* V 1 2 3 4 5 */
1591: /* F F V V V */
1592: /* D Q D D Q */
1593: /* */
1594: int *TvarsD;
1.330 brouard 1595: int *TnsdVar;
1.234 brouard 1596: int *TvarsDind;
1597: int *TvarsQ;
1598: int *TvarsQind;
1599:
1.318 brouard 1600: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1601: int nresult=0;
1.258 brouard 1602: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1603: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1604: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1605: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1606: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1607: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1608: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1609: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1610: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1611: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1612: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1613:
1614: /* ncovcol=1(Males=0 Females=1) nqv=1(raedyrs) ntv=2(withoutiadl=0 withiadl=1, witoutadl=0 withoutadl=1) nqtv=1(bmi) nlstate=3 ndeath=1
1615: # States 1=Coresidence, 2 Living alone, 3 Institution
1616: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1617: */
1.234 brouard 1618: /* int *TDvar; /\**< TDvar[1]=4, TDvarF[2]=3, TDvar[3]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
1.232 brouard 1619: int *TvarF; /**< TvarF[1]=Tvar[6]=2, TvarF[2]=Tvar[7]=7, TvarF[3]=Tvar[9]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1620: int *TvarFind; /**< TvarFind[1]=6, TvarFind[2]=7, Tvarind[3]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1621: int *TvarV; /**< TvarV[1]=Tvar[1]=5, TvarV[2]=Tvar[2]=4 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1622: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1623: int *TvarA; /**< TvarA[1]=Tvar[5]=5, TvarA[2]=Tvar[8]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1624: int *TvarAind; /**< TvarindA[1]=5, TvarAind[2]=8 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 1625: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1626: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1627: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1628: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1629: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1630: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1631: int *TvarVQ; /* TvarVQ[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying quantitative variable */
1632: int *TvarVQind; /* TvarVQind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying quantitative variable */
1.339 brouard 1633: int *TvarVV; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1634: int *TvarVVind; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1635: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
1636: /* model V1+V3+age*V1+age*V3+V1*V3 */
1637: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
1638: /* TvarVV={3,1,3}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1639: /* TvarVVind={2,5,5}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1.230 brouard 1640: int *Tvarsel; /**< Selected covariates for output */
1641: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1642: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1643: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1644: int *Dummy; /** Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product */
1.238 brouard 1645: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1646: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1647: int *Tage;
1.227 brouard 1648: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1649: int *Tmodelind; /** Tmodelind[Tvaraff[3]]=9 for V1 position,Tvaraff[1]@9={4, 3, 1, 0, 0, 0, 0, 0, 0}, model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.230 brouard 1650: int *TmodelInvind; /** Tmodelind[Tvaraff[3]]=9 for V1 position,Tvaraff[1]@9={4, 3, 1, 0, 0, 0, 0, 0, 0}, model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1651: int *TmodelInvQind; /** Tmodelqind[1]=1 for V5(quantitative varying) position,Tvaraff[1]@9={4, 3, 1, 0, 0, 0, 0, 0, 0}, model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.145 brouard 1652: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1653: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1654: int **Tvard;
1.330 brouard 1655: int **Tvardk;
1.227 brouard 1656: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1657: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1658: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1659: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1660: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1661: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1662: double *lsurv, *lpop, *tpop;
1663:
1.231 brouard 1664: #define FD 1; /* Fixed dummy covariate */
1665: #define FQ 2; /* Fixed quantitative covariate */
1666: #define FP 3; /* Fixed product covariate */
1667: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1668: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1669: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1670: #define VD 10; /* Varying dummy covariate */
1671: #define VQ 11; /* Varying quantitative covariate */
1672: #define VP 12; /* Varying product covariate */
1673: #define VPDD 13; /* Varying product dummy*dummy covariate */
1674: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1675: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1676: #define APFD 16; /* Age product * fixed dummy covariate */
1677: #define APFQ 17; /* Age product * fixed quantitative covariate */
1678: #define APVD 18; /* Age product * varying dummy covariate */
1679: #define APVQ 19; /* Age product * varying quantitative covariate */
1680:
1681: #define FTYPE 1; /* Fixed covariate */
1682: #define VTYPE 2; /* Varying covariate (loop in wave) */
1683: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1684:
1685: struct kmodel{
1686: int maintype; /* main type */
1687: int subtype; /* subtype */
1688: };
1689: struct kmodel modell[NCOVMAX];
1690:
1.143 brouard 1691: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1692: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1693:
1694: /**************** split *************************/
1695: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1696: {
1697: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1698: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1699: */
1700: char *ss; /* pointer */
1.186 brouard 1701: int l1=0, l2=0; /* length counters */
1.126 brouard 1702:
1703: l1 = strlen(path ); /* length of path */
1704: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1705: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1706: if ( ss == NULL ) { /* no directory, so determine current directory */
1707: strcpy( name, path ); /* we got the fullname name because no directory */
1708: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1709: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1710: /* get current working directory */
1711: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1712: #ifdef WIN32
1713: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1714: #else
1715: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1716: #endif
1.126 brouard 1717: return( GLOCK_ERROR_GETCWD );
1718: }
1719: /* got dirc from getcwd*/
1720: printf(" DIRC = %s \n",dirc);
1.205 brouard 1721: } else { /* strip directory from path */
1.126 brouard 1722: ss++; /* after this, the filename */
1723: l2 = strlen( ss ); /* length of filename */
1724: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1725: strcpy( name, ss ); /* save file name */
1726: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1727: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1728: printf(" DIRC2 = %s \n",dirc);
1729: }
1730: /* We add a separator at the end of dirc if not exists */
1731: l1 = strlen( dirc ); /* length of directory */
1732: if( dirc[l1-1] != DIRSEPARATOR ){
1733: dirc[l1] = DIRSEPARATOR;
1734: dirc[l1+1] = 0;
1735: printf(" DIRC3 = %s \n",dirc);
1736: }
1737: ss = strrchr( name, '.' ); /* find last / */
1738: if (ss >0){
1739: ss++;
1740: strcpy(ext,ss); /* save extension */
1741: l1= strlen( name);
1742: l2= strlen(ss)+1;
1743: strncpy( finame, name, l1-l2);
1744: finame[l1-l2]= 0;
1745: }
1746:
1747: return( 0 ); /* we're done */
1748: }
1749:
1750:
1751: /******************************************/
1752:
1753: void replace_back_to_slash(char *s, char*t)
1754: {
1755: int i;
1756: int lg=0;
1757: i=0;
1758: lg=strlen(t);
1759: for(i=0; i<= lg; i++) {
1760: (s[i] = t[i]);
1761: if (t[i]== '\\') s[i]='/';
1762: }
1763: }
1764:
1.132 brouard 1765: char *trimbb(char *out, char *in)
1.137 brouard 1766: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1767: char *s;
1768: s=out;
1769: while (*in != '\0'){
1.137 brouard 1770: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1771: in++;
1772: }
1773: *out++ = *in++;
1774: }
1775: *out='\0';
1776: return s;
1777: }
1778:
1.187 brouard 1779: /* char *substrchaine(char *out, char *in, char *chain) */
1780: /* { */
1781: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1782: /* char *s, *t; */
1783: /* t=in;s=out; */
1784: /* while ((*in != *chain) && (*in != '\0')){ */
1785: /* *out++ = *in++; */
1786: /* } */
1787:
1788: /* /\* *in matches *chain *\/ */
1789: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1790: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1791: /* } */
1792: /* in--; chain--; */
1793: /* while ( (*in != '\0')){ */
1794: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1795: /* *out++ = *in++; */
1796: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1797: /* } */
1798: /* *out='\0'; */
1799: /* out=s; */
1800: /* return out; */
1801: /* } */
1802: char *substrchaine(char *out, char *in, char *chain)
1803: {
1804: /* Substract chain 'chain' from 'in', return and output 'out' */
1805: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1806:
1807: char *strloc;
1808:
1809: strcpy (out, in);
1810: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1811: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1812: if(strloc != NULL){
1813: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1814: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1815: /* strcpy (strloc, strloc +strlen(chain));*/
1816: }
1817: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1818: return out;
1819: }
1820:
1821:
1.145 brouard 1822: char *cutl(char *blocc, char *alocc, char *in, char occ)
1823: {
1.187 brouard 1824: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1825: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1826: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1827: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1828: */
1.160 brouard 1829: char *s, *t;
1.145 brouard 1830: t=in;s=in;
1831: while ((*in != occ) && (*in != '\0')){
1832: *alocc++ = *in++;
1833: }
1834: if( *in == occ){
1835: *(alocc)='\0';
1836: s=++in;
1837: }
1838:
1839: if (s == t) {/* occ not found */
1840: *(alocc-(in-s))='\0';
1841: in=s;
1842: }
1843: while ( *in != '\0'){
1844: *blocc++ = *in++;
1845: }
1846:
1847: *blocc='\0';
1848: return t;
1849: }
1.137 brouard 1850: char *cutv(char *blocc, char *alocc, char *in, char occ)
1851: {
1.187 brouard 1852: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1853: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1854: gives blocc="abcdef2ghi" and alocc="j".
1855: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1856: */
1857: char *s, *t;
1858: t=in;s=in;
1859: while (*in != '\0'){
1860: while( *in == occ){
1861: *blocc++ = *in++;
1862: s=in;
1863: }
1864: *blocc++ = *in++;
1865: }
1866: if (s == t) /* occ not found */
1867: *(blocc-(in-s))='\0';
1868: else
1869: *(blocc-(in-s)-1)='\0';
1870: in=s;
1871: while ( *in != '\0'){
1872: *alocc++ = *in++;
1873: }
1874:
1875: *alocc='\0';
1876: return s;
1877: }
1878:
1.126 brouard 1879: int nbocc(char *s, char occ)
1880: {
1881: int i,j=0;
1882: int lg=20;
1883: i=0;
1884: lg=strlen(s);
1885: for(i=0; i<= lg; i++) {
1.234 brouard 1886: if (s[i] == occ ) j++;
1.126 brouard 1887: }
1888: return j;
1889: }
1890:
1.137 brouard 1891: /* void cutv(char *u,char *v, char*t, char occ) */
1892: /* { */
1893: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1894: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1895: /* gives u="abcdef2ghi" and v="j" *\/ */
1896: /* int i,lg,j,p=0; */
1897: /* i=0; */
1898: /* lg=strlen(t); */
1899: /* for(j=0; j<=lg-1; j++) { */
1900: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1901: /* } */
1.126 brouard 1902:
1.137 brouard 1903: /* for(j=0; j<p; j++) { */
1904: /* (u[j] = t[j]); */
1905: /* } */
1906: /* u[p]='\0'; */
1.126 brouard 1907:
1.137 brouard 1908: /* for(j=0; j<= lg; j++) { */
1909: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1910: /* } */
1911: /* } */
1.126 brouard 1912:
1.160 brouard 1913: #ifdef _WIN32
1914: char * strsep(char **pp, const char *delim)
1915: {
1916: char *p, *q;
1917:
1918: if ((p = *pp) == NULL)
1919: return 0;
1920: if ((q = strpbrk (p, delim)) != NULL)
1921: {
1922: *pp = q + 1;
1923: *q = '\0';
1924: }
1925: else
1926: *pp = 0;
1927: return p;
1928: }
1929: #endif
1930:
1.126 brouard 1931: /********************** nrerror ********************/
1932:
1933: void nrerror(char error_text[])
1934: {
1935: fprintf(stderr,"ERREUR ...\n");
1936: fprintf(stderr,"%s\n",error_text);
1937: exit(EXIT_FAILURE);
1938: }
1939: /*********************** vector *******************/
1940: double *vector(int nl, int nh)
1941: {
1942: double *v;
1943: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1944: if (!v) nrerror("allocation failure in vector");
1945: return v-nl+NR_END;
1946: }
1947:
1948: /************************ free vector ******************/
1949: void free_vector(double*v, int nl, int nh)
1950: {
1951: free((FREE_ARG)(v+nl-NR_END));
1952: }
1953:
1954: /************************ivector *******************************/
1955: int *ivector(long nl,long nh)
1956: {
1957: int *v;
1958: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1959: if (!v) nrerror("allocation failure in ivector");
1960: return v-nl+NR_END;
1961: }
1962:
1963: /******************free ivector **************************/
1964: void free_ivector(int *v, long nl, long nh)
1965: {
1966: free((FREE_ARG)(v+nl-NR_END));
1967: }
1968:
1969: /************************lvector *******************************/
1970: long *lvector(long nl,long nh)
1971: {
1972: long *v;
1973: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1974: if (!v) nrerror("allocation failure in ivector");
1975: return v-nl+NR_END;
1976: }
1977:
1978: /******************free lvector **************************/
1979: void free_lvector(long *v, long nl, long nh)
1980: {
1981: free((FREE_ARG)(v+nl-NR_END));
1982: }
1983:
1984: /******************* imatrix *******************************/
1985: int **imatrix(long nrl, long nrh, long ncl, long nch)
1986: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1987: {
1988: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1989: int **m;
1990:
1991: /* allocate pointers to rows */
1992: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1993: if (!m) nrerror("allocation failure 1 in matrix()");
1994: m += NR_END;
1995: m -= nrl;
1996:
1997:
1998: /* allocate rows and set pointers to them */
1999: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
2000: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2001: m[nrl] += NR_END;
2002: m[nrl] -= ncl;
2003:
2004: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
2005:
2006: /* return pointer to array of pointers to rows */
2007: return m;
2008: }
2009:
2010: /****************** free_imatrix *************************/
2011: void free_imatrix(m,nrl,nrh,ncl,nch)
2012: int **m;
2013: long nch,ncl,nrh,nrl;
2014: /* free an int matrix allocated by imatrix() */
2015: {
2016: free((FREE_ARG) (m[nrl]+ncl-NR_END));
2017: free((FREE_ARG) (m+nrl-NR_END));
2018: }
2019:
2020: /******************* matrix *******************************/
2021: double **matrix(long nrl, long nrh, long ncl, long nch)
2022: {
2023: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
2024: double **m;
2025:
2026: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2027: if (!m) nrerror("allocation failure 1 in matrix()");
2028: m += NR_END;
2029: m -= nrl;
2030:
2031: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2032: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2033: m[nrl] += NR_END;
2034: m[nrl] -= ncl;
2035:
2036: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2037: return m;
1.145 brouard 2038: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
2039: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
2040: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 2041: */
2042: }
2043:
2044: /*************************free matrix ************************/
2045: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
2046: {
2047: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2048: free((FREE_ARG)(m+nrl-NR_END));
2049: }
2050:
2051: /******************* ma3x *******************************/
2052: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
2053: {
2054: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
2055: double ***m;
2056:
2057: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2058: if (!m) nrerror("allocation failure 1 in matrix()");
2059: m += NR_END;
2060: m -= nrl;
2061:
2062: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2063: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2064: m[nrl] += NR_END;
2065: m[nrl] -= ncl;
2066:
2067: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2068:
2069: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
2070: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
2071: m[nrl][ncl] += NR_END;
2072: m[nrl][ncl] -= nll;
2073: for (j=ncl+1; j<=nch; j++)
2074: m[nrl][j]=m[nrl][j-1]+nlay;
2075:
2076: for (i=nrl+1; i<=nrh; i++) {
2077: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2078: for (j=ncl+1; j<=nch; j++)
2079: m[i][j]=m[i][j-1]+nlay;
2080: }
2081: return m;
2082: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2083: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2084: */
2085: }
2086:
2087: /*************************free ma3x ************************/
2088: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2089: {
2090: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2091: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2092: free((FREE_ARG)(m+nrl-NR_END));
2093: }
2094:
2095: /*************** function subdirf ***********/
2096: char *subdirf(char fileres[])
2097: {
2098: /* Caution optionfilefiname is hidden */
2099: strcpy(tmpout,optionfilefiname);
2100: strcat(tmpout,"/"); /* Add to the right */
2101: strcat(tmpout,fileres);
2102: return tmpout;
2103: }
2104:
2105: /*************** function subdirf2 ***********/
2106: char *subdirf2(char fileres[], char *preop)
2107: {
1.314 brouard 2108: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2109: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2110: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2111: /* Caution optionfilefiname is hidden */
2112: strcpy(tmpout,optionfilefiname);
2113: strcat(tmpout,"/");
2114: strcat(tmpout,preop);
2115: strcat(tmpout,fileres);
2116: return tmpout;
2117: }
2118:
2119: /*************** function subdirf3 ***********/
2120: char *subdirf3(char fileres[], char *preop, char *preop2)
2121: {
2122:
2123: /* Caution optionfilefiname is hidden */
2124: strcpy(tmpout,optionfilefiname);
2125: strcat(tmpout,"/");
2126: strcat(tmpout,preop);
2127: strcat(tmpout,preop2);
2128: strcat(tmpout,fileres);
2129: return tmpout;
2130: }
1.213 brouard 2131:
2132: /*************** function subdirfext ***********/
2133: char *subdirfext(char fileres[], char *preop, char *postop)
2134: {
2135:
2136: strcpy(tmpout,preop);
2137: strcat(tmpout,fileres);
2138: strcat(tmpout,postop);
2139: return tmpout;
2140: }
1.126 brouard 2141:
1.213 brouard 2142: /*************** function subdirfext3 ***********/
2143: char *subdirfext3(char fileres[], char *preop, char *postop)
2144: {
2145:
2146: /* Caution optionfilefiname is hidden */
2147: strcpy(tmpout,optionfilefiname);
2148: strcat(tmpout,"/");
2149: strcat(tmpout,preop);
2150: strcat(tmpout,fileres);
2151: strcat(tmpout,postop);
2152: return tmpout;
2153: }
2154:
1.162 brouard 2155: char *asc_diff_time(long time_sec, char ascdiff[])
2156: {
2157: long sec_left, days, hours, minutes;
2158: days = (time_sec) / (60*60*24);
2159: sec_left = (time_sec) % (60*60*24);
2160: hours = (sec_left) / (60*60) ;
2161: sec_left = (sec_left) %(60*60);
2162: minutes = (sec_left) /60;
2163: sec_left = (sec_left) % (60);
2164: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2165: return ascdiff;
2166: }
2167:
1.126 brouard 2168: /***************** f1dim *************************/
2169: extern int ncom;
2170: extern double *pcom,*xicom;
2171: extern double (*nrfunc)(double []);
2172:
2173: double f1dim(double x)
2174: {
2175: int j;
2176: double f;
2177: double *xt;
2178:
2179: xt=vector(1,ncom);
2180: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2181: f=(*nrfunc)(xt);
2182: free_vector(xt,1,ncom);
2183: return f;
2184: }
2185:
2186: /*****************brent *************************/
2187: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2188: {
2189: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2190: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2191: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2192: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2193: * returned function value.
2194: */
1.126 brouard 2195: int iter;
2196: double a,b,d,etemp;
1.159 brouard 2197: double fu=0,fv,fw,fx;
1.164 brouard 2198: double ftemp=0.;
1.126 brouard 2199: double p,q,r,tol1,tol2,u,v,w,x,xm;
2200: double e=0.0;
2201:
2202: a=(ax < cx ? ax : cx);
2203: b=(ax > cx ? ax : cx);
2204: x=w=v=bx;
2205: fw=fv=fx=(*f)(x);
2206: for (iter=1;iter<=ITMAX;iter++) {
2207: xm=0.5*(a+b);
2208: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2209: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2210: printf(".");fflush(stdout);
2211: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2212: #ifdef DEBUGBRENT
1.126 brouard 2213: printf("br %d,x=%.10e xm=%.10e b=%.10e a=%.10e tol=%.10e tol1=%.10e tol2=%.10e x-xm=%.10e fx=%.12e fu=%.12e,fw=%.12e,ftemp=%.12e,ftol=%.12e\n",iter,x,xm,b,a,tol,tol1,tol2,(x-xm),fx,fu,fw,ftemp,ftol);
2214: fprintf(ficlog,"br %d,x=%.10e xm=%.10e b=%.10e a=%.10e tol=%.10e tol1=%.10e tol2=%.10e x-xm=%.10e fx=%.12e fu=%.12e,fw=%.12e,ftemp=%.12e,ftol=%.12e\n",iter,x,xm,b,a,tol,tol1,tol2,(x-xm),fx,fu,fw,ftemp,ftol);
2215: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2216: #endif
2217: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2218: *xmin=x;
2219: return fx;
2220: }
2221: ftemp=fu;
2222: if (fabs(e) > tol1) {
2223: r=(x-w)*(fx-fv);
2224: q=(x-v)*(fx-fw);
2225: p=(x-v)*q-(x-w)*r;
2226: q=2.0*(q-r);
2227: if (q > 0.0) p = -p;
2228: q=fabs(q);
2229: etemp=e;
2230: e=d;
2231: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2232: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2233: else {
1.224 brouard 2234: d=p/q;
2235: u=x+d;
2236: if (u-a < tol2 || b-u < tol2)
2237: d=SIGN(tol1,xm-x);
1.126 brouard 2238: }
2239: } else {
2240: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2241: }
2242: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2243: fu=(*f)(u);
2244: if (fu <= fx) {
2245: if (u >= x) a=x; else b=x;
2246: SHFT(v,w,x,u)
1.183 brouard 2247: SHFT(fv,fw,fx,fu)
2248: } else {
2249: if (u < x) a=u; else b=u;
2250: if (fu <= fw || w == x) {
1.224 brouard 2251: v=w;
2252: w=u;
2253: fv=fw;
2254: fw=fu;
1.183 brouard 2255: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2256: v=u;
2257: fv=fu;
1.183 brouard 2258: }
2259: }
1.126 brouard 2260: }
2261: nrerror("Too many iterations in brent");
2262: *xmin=x;
2263: return fx;
2264: }
2265:
2266: /****************** mnbrak ***********************/
2267:
2268: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2269: double (*func)(double))
1.183 brouard 2270: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2271: the downhill direction (defined by the function as evaluated at the initial points) and returns
2272: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2273: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2274: */
1.126 brouard 2275: double ulim,u,r,q, dum;
2276: double fu;
1.187 brouard 2277:
2278: double scale=10.;
2279: int iterscale=0;
2280:
2281: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2282: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2283:
2284:
2285: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2286: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2287: /* *bx = *ax - (*ax - *bx)/scale; */
2288: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2289: /* } */
2290:
1.126 brouard 2291: if (*fb > *fa) {
2292: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2293: SHFT(dum,*fb,*fa,dum)
2294: }
1.126 brouard 2295: *cx=(*bx)+GOLD*(*bx-*ax);
2296: *fc=(*func)(*cx);
1.183 brouard 2297: #ifdef DEBUG
1.224 brouard 2298: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2299: fprintf(ficlog,"mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
1.183 brouard 2300: #endif
1.224 brouard 2301: while (*fb > *fc) { /* Declining a,b,c with fa> fb > fc. If fc=inf it exits and if flat fb=fc it exits too.*/
1.126 brouard 2302: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2303: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2304: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2305: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2306: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2307: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2308: fu=(*func)(u);
1.163 brouard 2309: #ifdef DEBUG
2310: /* f(x)=A(x-u)**2+f(u) */
2311: double A, fparabu;
2312: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2313: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2314: printf("\nmnbrak (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf), (*u=%.12f, fu=%.12lf, fparabu=%.12f, q=%lf < %lf=r)\n",*ax,*fa,*bx,*fb,*cx,*fc,u,fu, fparabu,q,r);
2315: fprintf(ficlog,"\nmnbrak (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf), (*u=%.12f, fu=%.12lf, fparabu=%.12f, q=%lf < %lf=r)\n",*ax,*fa,*bx,*fb,*cx,*fc,u,fu, fparabu,q,r);
1.183 brouard 2316: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2317: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2318: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2319: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2320: #endif
1.184 brouard 2321: #ifdef MNBRAKORIGINAL
1.183 brouard 2322: #else
1.191 brouard 2323: /* if (fu > *fc) { */
2324: /* #ifdef DEBUG */
2325: /* printf("mnbrak4 fu > fc \n"); */
2326: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2327: /* #endif */
2328: /* /\* SHFT(u,*cx,*cx,u) /\\* ie a=c, c=u and u=c; in that case, next SHFT(a,b,c,u) will give a=b=b, b=c=u, c=u=c and *\\/ *\/ */
2329: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2330: /* dum=u; /\* Shifting c and u *\/ */
2331: /* u = *cx; */
2332: /* *cx = dum; */
2333: /* dum = fu; */
2334: /* fu = *fc; */
2335: /* *fc =dum; */
2336: /* } else { /\* end *\/ */
2337: /* #ifdef DEBUG */
2338: /* printf("mnbrak3 fu < fc \n"); */
2339: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2340: /* #endif */
2341: /* dum=u; /\* Shifting c and u *\/ */
2342: /* u = *cx; */
2343: /* *cx = dum; */
2344: /* dum = fu; */
2345: /* fu = *fc; */
2346: /* *fc =dum; */
2347: /* } */
1.224 brouard 2348: #ifdef DEBUGMNBRAK
2349: double A, fparabu;
2350: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2351: fparabu= *fa - A*(*ax-u)*(*ax-u);
2352: printf("\nmnbrak35 ax=%lf fa=%lf bx=%lf fb=%lf, u=%lf fp=%lf fu=%lf < or >= fc=%lf cx=%lf, q=%lf < %lf=r \n",*ax, *fa, *bx,*fb,u,fparabu,fu,*fc,*cx,q,r);
2353: fprintf(ficlog,"\nmnbrak35 ax=%lf fa=%lf bx=%lf fb=%lf, u=%lf fp=%lf fu=%lf < or >= fc=%lf cx=%lf, q=%lf < %lf=r \n",*ax, *fa, *bx,*fb,u,fparabu,fu,*fc,*cx,q,r);
1.183 brouard 2354: #endif
1.191 brouard 2355: dum=u; /* Shifting c and u */
2356: u = *cx;
2357: *cx = dum;
2358: dum = fu;
2359: fu = *fc;
2360: *fc =dum;
1.183 brouard 2361: #endif
1.162 brouard 2362: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2363: #ifdef DEBUG
1.224 brouard 2364: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2365: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2366: #endif
1.126 brouard 2367: fu=(*func)(u);
2368: if (fu < *fc) {
1.183 brouard 2369: #ifdef DEBUG
1.224 brouard 2370: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2371: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2372: #endif
2373: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2374: SHFT(*fb,*fc,fu,(*func)(u))
2375: #ifdef DEBUG
2376: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2377: #endif
2378: }
1.162 brouard 2379: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2380: #ifdef DEBUG
1.224 brouard 2381: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2382: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2383: #endif
1.126 brouard 2384: u=ulim;
2385: fu=(*func)(u);
1.183 brouard 2386: } else { /* u could be left to b (if r > q parabola has a maximum) */
2387: #ifdef DEBUG
1.224 brouard 2388: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2389: fprintf(ficlog,"\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
1.183 brouard 2390: #endif
1.126 brouard 2391: u=(*cx)+GOLD*(*cx-*bx);
2392: fu=(*func)(u);
1.224 brouard 2393: #ifdef DEBUG
2394: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2395: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2396: #endif
1.183 brouard 2397: } /* end tests */
1.126 brouard 2398: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2399: SHFT(*fa,*fb,*fc,fu)
2400: #ifdef DEBUG
1.224 brouard 2401: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2402: fprintf(ficlog, "\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
1.183 brouard 2403: #endif
2404: } /* end while; ie return (a, b, c, fa, fb, fc) such that a < b < c with f(a) > f(b) and fb < f(c) */
1.126 brouard 2405: }
2406:
2407: /*************** linmin ************************/
1.162 brouard 2408: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2409: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2410: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2411: the value of func at the returned location p . This is actually all accomplished by calling the
2412: routines mnbrak and brent .*/
1.126 brouard 2413: int ncom;
2414: double *pcom,*xicom;
2415: double (*nrfunc)(double []);
2416:
1.224 brouard 2417: #ifdef LINMINORIGINAL
1.126 brouard 2418: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2419: #else
2420: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2421: #endif
1.126 brouard 2422: {
2423: double brent(double ax, double bx, double cx,
2424: double (*f)(double), double tol, double *xmin);
2425: double f1dim(double x);
2426: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2427: double *fc, double (*func)(double));
2428: int j;
2429: double xx,xmin,bx,ax;
2430: double fx,fb,fa;
1.187 brouard 2431:
1.203 brouard 2432: #ifdef LINMINORIGINAL
2433: #else
2434: double scale=10., axs, xxs; /* Scale added for infinity */
2435: #endif
2436:
1.126 brouard 2437: ncom=n;
2438: pcom=vector(1,n);
2439: xicom=vector(1,n);
2440: nrfunc=func;
2441: for (j=1;j<=n;j++) {
2442: pcom[j]=p[j];
1.202 brouard 2443: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2444: }
1.187 brouard 2445:
1.203 brouard 2446: #ifdef LINMINORIGINAL
2447: xx=1.;
2448: #else
2449: axs=0.0;
2450: xxs=1.;
2451: do{
2452: xx= xxs;
2453: #endif
1.187 brouard 2454: ax=0.;
2455: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2456: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2457: /* xt[x,j]=pcom[j]+x*xicom[j] f(ax) = f(xt(a,j=1,n)) = f(p(j) + 0 * xi(j)) and f(xx) = f(xt(x, j=1,n)) = f(p(j) + 1 * xi(j)) */
2458: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2459: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2460: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2461: /* Find a bracket a,x,b in direction n=xi ie xicom, order may change. Scale is [0:xxs*xi[j]] et non plus [0:xi[j]]*/
1.203 brouard 2462: #ifdef LINMINORIGINAL
2463: #else
2464: if (fx != fx){
1.224 brouard 2465: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2466: printf("|");
2467: fprintf(ficlog,"|");
1.203 brouard 2468: #ifdef DEBUGLINMIN
1.224 brouard 2469: printf("\nLinmin NAN : input [axs=%lf:xxs=%lf], mnbrak outputs fx=%lf <(fb=%lf and fa=%lf) with xx=%lf in [ax=%lf:bx=%lf] \n", axs, xxs, fx,fb, fa, xx, ax, bx);
1.203 brouard 2470: #endif
2471: }
1.224 brouard 2472: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2473: #endif
2474:
1.191 brouard 2475: #ifdef DEBUGLINMIN
2476: printf("\nLinmin after mnbrak: ax=%12.7f xx=%12.7f bx=%12.7f fa=%12.2f fx=%12.2f fb=%12.2f\n", ax,xx,bx,fa,fx,fb);
1.202 brouard 2477: fprintf(ficlog,"\nLinmin after mnbrak: ax=%12.7f xx=%12.7f bx=%12.7f fa=%12.2f fx=%12.2f fb=%12.2f\n", ax,xx,bx,fa,fx,fb);
1.191 brouard 2478: #endif
1.224 brouard 2479: #ifdef LINMINORIGINAL
2480: #else
1.317 brouard 2481: if(fb == fx){ /* Flat function in the direction */
2482: xmin=xx;
1.224 brouard 2483: *flat=1;
1.317 brouard 2484: }else{
1.224 brouard 2485: *flat=0;
2486: #endif
2487: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2488: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2489: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2490: /* fmin = f(p[j] + xmin * xi[j]) */
2491: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2492: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2493: #ifdef DEBUG
1.224 brouard 2494: printf("retour brent from bracket (a=%lf fa=%lf, xx=%lf fx=%lf, b=%lf fb=%lf): fret=%lf xmin=%lf\n",ax,fa,xx,fx,bx,fb,*fret,xmin);
2495: fprintf(ficlog,"retour brent from bracket (a=%lf fa=%lf, xx=%lf fx=%lf, b=%lf fb=%lf): fret=%lf xmin=%lf\n",ax,fa,xx,fx,bx,fb,*fret,xmin);
2496: #endif
2497: #ifdef LINMINORIGINAL
2498: #else
2499: }
1.126 brouard 2500: #endif
1.191 brouard 2501: #ifdef DEBUGLINMIN
2502: printf("linmin end ");
1.202 brouard 2503: fprintf(ficlog,"linmin end ");
1.191 brouard 2504: #endif
1.126 brouard 2505: for (j=1;j<=n;j++) {
1.203 brouard 2506: #ifdef LINMINORIGINAL
2507: xi[j] *= xmin;
2508: #else
2509: #ifdef DEBUGLINMIN
2510: if(xxs <1.0)
2511: printf(" before xi[%d]=%12.8f", j,xi[j]);
2512: #endif
2513: xi[j] *= xmin*xxs; /* xi rescaled by xmin and number of loops: if xmin=-1.237 and xi=(1,0,...,0) xi=(-1.237,0,...,0) */
2514: #ifdef DEBUGLINMIN
2515: if(xxs <1.0)
2516: printf(" after xi[%d]=%12.8f, xmin=%12.8f, ax=%12.8f, xx=%12.8f, bx=%12.8f, xxs=%12.8f", j,xi[j], xmin, ax, xx, bx,xxs );
2517: #endif
2518: #endif
1.187 brouard 2519: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2520: }
1.191 brouard 2521: #ifdef DEBUGLINMIN
1.203 brouard 2522: printf("\n");
1.191 brouard 2523: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2524: fprintf(ficlog,"Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.191 brouard 2525: for (j=1;j<=n;j++) {
1.202 brouard 2526: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2527: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2528: if(j % ncovmodel == 0){
1.191 brouard 2529: printf("\n");
1.202 brouard 2530: fprintf(ficlog,"\n");
2531: }
1.191 brouard 2532: }
1.203 brouard 2533: #else
1.191 brouard 2534: #endif
1.126 brouard 2535: free_vector(xicom,1,n);
2536: free_vector(pcom,1,n);
2537: }
2538:
2539:
2540: /*************** powell ************************/
1.162 brouard 2541: /*
1.317 brouard 2542: Minimization of a function func of n variables. Input consists in an initial starting point
2543: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2544: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2545: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2546: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2547: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2548: */
1.224 brouard 2549: #ifdef LINMINORIGINAL
2550: #else
2551: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2552: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2553: #endif
1.126 brouard 2554: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2555: double (*func)(double []))
2556: {
1.224 brouard 2557: #ifdef LINMINORIGINAL
2558: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2559: double (*func)(double []));
1.224 brouard 2560: #else
1.241 brouard 2561: void linmin(double p[], double xi[], int n, double *fret,
2562: double (*func)(double []),int *flat);
1.224 brouard 2563: #endif
1.239 brouard 2564: int i,ibig,j,jk,k;
1.126 brouard 2565: double del,t,*pt,*ptt,*xit;
1.181 brouard 2566: double directest;
1.126 brouard 2567: double fp,fptt;
2568: double *xits;
2569: int niterf, itmp;
2570:
2571: pt=vector(1,n);
2572: ptt=vector(1,n);
2573: xit=vector(1,n);
2574: xits=vector(1,n);
2575: *fret=(*func)(p);
2576: for (j=1;j<=n;j++) pt[j]=p[j];
1.338 brouard 2577: rcurr_time = time(NULL);
2578: fp=(*fret); /* Initialisation */
1.126 brouard 2579: for (*iter=1;;++(*iter)) {
2580: ibig=0;
2581: del=0.0;
1.157 brouard 2582: rlast_time=rcurr_time;
2583: /* (void) gettimeofday(&curr_time,&tzp); */
2584: rcurr_time = time(NULL);
2585: curr_time = *localtime(&rcurr_time);
1.337 brouard 2586: /* printf("\nPowell iter=%d -2*LL=%.12f gain=%.12f=%.3g %ld sec. %ld sec.",*iter,*fret, fp-*fret,fp-*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout); */
2587: /* fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f gain=%.12f=%.3g %ld sec. %ld sec.",*iter,*fret, fp-*fret,fp-*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog); */
2588: printf("\nPowell iter=%d -2*LL=%.12f gain=%.3lg %ld sec. %ld sec.",*iter,*fret,fp-*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2589: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f gain=%.3lg %ld sec. %ld sec.",*iter,*fret,fp-*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
1.157 brouard 2590: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2591: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2592: for (i=1;i<=n;i++) {
1.126 brouard 2593: fprintf(ficrespow," %.12lf", p[i]);
2594: }
1.239 brouard 2595: fprintf(ficrespow,"\n");fflush(ficrespow);
2596: printf("\n#model= 1 + age ");
2597: fprintf(ficlog,"\n#model= 1 + age ");
2598: if(nagesqr==1){
1.241 brouard 2599: printf(" + age*age ");
2600: fprintf(ficlog," + age*age ");
1.239 brouard 2601: }
2602: for(j=1;j <=ncovmodel-2;j++){
2603: if(Typevar[j]==0) {
2604: printf(" + V%d ",Tvar[j]);
2605: fprintf(ficlog," + V%d ",Tvar[j]);
2606: }else if(Typevar[j]==1) {
2607: printf(" + V%d*age ",Tvar[j]);
2608: fprintf(ficlog," + V%d*age ",Tvar[j]);
2609: }else if(Typevar[j]==2) {
2610: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2611: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2612: }
2613: }
1.126 brouard 2614: printf("\n");
1.239 brouard 2615: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2616: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2617: fprintf(ficlog,"\n");
1.239 brouard 2618: for(i=1,jk=1; i <=nlstate; i++){
2619: for(k=1; k <=(nlstate+ndeath); k++){
2620: if (k != i) {
2621: printf("%d%d ",i,k);
2622: fprintf(ficlog,"%d%d ",i,k);
2623: for(j=1; j <=ncovmodel; j++){
2624: printf("%12.7f ",p[jk]);
2625: fprintf(ficlog,"%12.7f ",p[jk]);
2626: jk++;
2627: }
2628: printf("\n");
2629: fprintf(ficlog,"\n");
2630: }
2631: }
2632: }
1.241 brouard 2633: if(*iter <=3 && *iter >1){
1.157 brouard 2634: tml = *localtime(&rcurr_time);
2635: strcpy(strcurr,asctime(&tml));
2636: rforecast_time=rcurr_time;
1.126 brouard 2637: itmp = strlen(strcurr);
2638: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2639: strcurr[itmp-1]='\0';
1.162 brouard 2640: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2641: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2642: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2643: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2644: forecast_time = *localtime(&rforecast_time);
2645: strcpy(strfor,asctime(&forecast_time));
2646: itmp = strlen(strfor);
2647: if(strfor[itmp-1]=='\n')
2648: strfor[itmp-1]='\0';
2649: printf(" - if your program needs %d iterations to converge, convergence will be \n reached in %s i.e.\n on %s (current time is %s);\n",niterf, asc_diff_time(rforecast_time-rcurr_time,tmpout),strfor,strcurr);
2650: fprintf(ficlog," - if your program needs %d iterations to converge, convergence will be \n reached in %s i.e.\n on %s (current time is %s);\n",niterf, asc_diff_time(rforecast_time-rcurr_time,tmpout),strfor,strcurr);
1.126 brouard 2651: }
2652: }
1.187 brouard 2653: for (i=1;i<=n;i++) { /* For each direction i */
2654: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2655: fptt=(*fret);
2656: #ifdef DEBUG
1.203 brouard 2657: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2658: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2659: #endif
1.203 brouard 2660: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2661: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2662: #ifdef LINMINORIGINAL
1.188 brouard 2663: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2664: #else
2665: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2666: flatdir[i]=flat; /* Function is vanishing in that direction i */
2667: #endif
2668: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2669: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2670: /* because that direction will be replaced unless the gain del is small */
2671: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2672: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2673: /* with the new direction. */
2674: del=fabs(fptt-(*fret));
2675: ibig=i;
1.126 brouard 2676: }
2677: #ifdef DEBUG
2678: printf("%d %.12e",i,(*fret));
2679: fprintf(ficlog,"%d %.12e",i,(*fret));
2680: for (j=1;j<=n;j++) {
1.224 brouard 2681: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2682: printf(" x(%d)=%.12e",j,xit[j]);
2683: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2684: }
2685: for(j=1;j<=n;j++) {
1.225 brouard 2686: printf(" p(%d)=%.12e",j,p[j]);
2687: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2688: }
2689: printf("\n");
2690: fprintf(ficlog,"\n");
2691: #endif
1.187 brouard 2692: } /* end loop on each direction i */
2693: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2694: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2695: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2696: for(j=1;j<=n;j++) {
2697: if(flatdir[j] >0){
2698: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2699: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2700: }
1.319 brouard 2701: /* printf("\n"); */
2702: /* fprintf(ficlog,"\n"); */
2703: }
1.243 brouard 2704: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2705: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2706: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2707: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2708: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2709: /* decreased of more than 3.84 */
2710: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2711: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2712: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2713:
1.188 brouard 2714: /* Starting the program with initial values given by a former maximization will simply change */
2715: /* the scales of the directions and the directions, because the are reset to canonical directions */
2716: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2717: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2718: #ifdef DEBUG
2719: int k[2],l;
2720: k[0]=1;
2721: k[1]=-1;
2722: printf("Max: %.12e",(*func)(p));
2723: fprintf(ficlog,"Max: %.12e",(*func)(p));
2724: for (j=1;j<=n;j++) {
2725: printf(" %.12e",p[j]);
2726: fprintf(ficlog," %.12e",p[j]);
2727: }
2728: printf("\n");
2729: fprintf(ficlog,"\n");
2730: for(l=0;l<=1;l++) {
2731: for (j=1;j<=n;j++) {
2732: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2733: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2734: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2735: }
2736: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2737: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2738: }
2739: #endif
2740:
2741: free_vector(xit,1,n);
2742: free_vector(xits,1,n);
2743: free_vector(ptt,1,n);
2744: free_vector(pt,1,n);
2745: return;
1.192 brouard 2746: } /* enough precision */
1.240 brouard 2747: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2748: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2749: ptt[j]=2.0*p[j]-pt[j];
2750: xit[j]=p[j]-pt[j];
2751: pt[j]=p[j];
2752: }
1.181 brouard 2753: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2754: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2755: if (*iter <=4) {
1.225 brouard 2756: #else
2757: #endif
1.224 brouard 2758: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2759: #else
1.161 brouard 2760: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2761: #endif
1.162 brouard 2762: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2763: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2764: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2765: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2766: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2767: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2768: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2769: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2770: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2771: /* Even if f3 <f1, directest can be negative and t >0 */
2772: /* mu² and del² are equal when f3=f1 */
2773: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2774: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2775: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2776: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2777: #ifdef NRCORIGINAL
2778: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2779: #else
2780: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del); /* Intel compiler doesn't work on one line; bug reported */
1.161 brouard 2781: t= t- del*SQR(fp-fptt);
1.183 brouard 2782: #endif
1.202 brouard 2783: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2784: #ifdef DEBUG
1.181 brouard 2785: printf("t1= %.12lf, t2= %.12lf, t=%.12lf directest=%.12lf\n", 2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del),del*SQR(fp-fptt),t,directest);
2786: fprintf(ficlog,"t1= %.12lf, t2= %.12lf, t=%.12lf directest=%.12lf\n", 2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del),del*SQR(fp-fptt),t,directest);
1.161 brouard 2787: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2788: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2789: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2790: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2791: printf("tt= %.12lf, t=%.12lf\n",2.0*(fp-2.0*(*fret)+fptt)*(fp-(*fret)-del)*(fp-(*fret)-del)-del*(fp-fptt)*(fp-fptt),t);
2792: fprintf(ficlog, "tt= %.12lf, t=%.12lf\n",2.0*(fp-2.0*(*fret)+fptt)*(fp-(*fret)-del)*(fp-(*fret)-del)-del*(fp-fptt)*(fp-fptt),t);
2793: #endif
1.183 brouard 2794: #ifdef POWELLORIGINAL
2795: if (t < 0.0) { /* Then we use it for new direction */
2796: #else
1.182 brouard 2797: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2798: printf("directest= %.12lf (if <0 we include P0 Pn as new direction), t= %.12lf, f1= %.12lf,f2= %.12lf,f3= %.12lf, del= %.12lf\n",directest, t, fp,(*fret),fptt,del);
1.192 brouard 2799: printf("f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
1.224 brouard 2800: fprintf(ficlog,"directest= %.12lf (if directest<0 or t<0 we include P0 Pn as new direction), t= %.12lf, f1= %.12lf,f2= %.12lf,f3= %.12lf, del= %.12lf\n",directest, t, fp,(*fret),fptt, del);
1.192 brouard 2801: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2802: }
1.181 brouard 2803: if (directest < 0.0) { /* Then we use it for new direction */
2804: #endif
1.191 brouard 2805: #ifdef DEBUGLINMIN
1.234 brouard 2806: printf("Before linmin in direction P%d-P0\n",n);
2807: for (j=1;j<=n;j++) {
2808: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2809: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2810: if(j % ncovmodel == 0){
2811: printf("\n");
2812: fprintf(ficlog,"\n");
2813: }
2814: }
1.224 brouard 2815: #endif
2816: #ifdef LINMINORIGINAL
1.234 brouard 2817: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2818: #else
1.234 brouard 2819: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2820: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2821: #endif
1.234 brouard 2822:
1.191 brouard 2823: #ifdef DEBUGLINMIN
1.234 brouard 2824: for (j=1;j<=n;j++) {
2825: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2826: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2827: if(j % ncovmodel == 0){
2828: printf("\n");
2829: fprintf(ficlog,"\n");
2830: }
2831: }
1.224 brouard 2832: #endif
1.234 brouard 2833: for (j=1;j<=n;j++) {
2834: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2835: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2836: }
1.224 brouard 2837: #ifdef LINMINORIGINAL
2838: #else
1.234 brouard 2839: for (j=1, flatd=0;j<=n;j++) {
2840: if(flatdir[j]>0)
2841: flatd++;
2842: }
2843: if(flatd >0){
1.255 brouard 2844: printf("%d flat directions: ",flatd);
2845: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2846: for (j=1;j<=n;j++) {
2847: if(flatdir[j]>0){
2848: printf("%d ",j);
2849: fprintf(ficlog,"%d ",j);
2850: }
2851: }
2852: printf("\n");
2853: fprintf(ficlog,"\n");
1.319 brouard 2854: #ifdef FLATSUP
2855: free_vector(xit,1,n);
2856: free_vector(xits,1,n);
2857: free_vector(ptt,1,n);
2858: free_vector(pt,1,n);
2859: return;
2860: #endif
1.234 brouard 2861: }
1.191 brouard 2862: #endif
1.234 brouard 2863: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2864: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2865:
1.126 brouard 2866: #ifdef DEBUG
1.234 brouard 2867: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2868: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2869: for(j=1;j<=n;j++){
2870: printf(" %lf",xit[j]);
2871: fprintf(ficlog," %lf",xit[j]);
2872: }
2873: printf("\n");
2874: fprintf(ficlog,"\n");
1.126 brouard 2875: #endif
1.192 brouard 2876: } /* end of t or directest negative */
1.224 brouard 2877: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2878: #else
1.234 brouard 2879: } /* end if (fptt < fp) */
1.192 brouard 2880: #endif
1.225 brouard 2881: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2882: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2883: #else
1.224 brouard 2884: #endif
1.234 brouard 2885: } /* loop iteration */
1.126 brouard 2886: }
1.234 brouard 2887:
1.126 brouard 2888: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2889:
1.235 brouard 2890: double **prevalim(double **prlim, int nlstate, double x[], double age, double **oldm, double **savm, double ftolpl, int *ncvyear, int ij, int nres)
1.234 brouard 2891: {
1.338 brouard 2892: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij . Nicely done
1.279 brouard 2893: * (and selected quantitative values in nres)
2894: * by left multiplying the unit
2895: * matrix by transitions matrix until convergence is reached with precision ftolpl
2896: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2897: * Wx is row vector: population in state 1, population in state 2, population dead
2898: * or prevalence in state 1, prevalence in state 2, 0
2899: * newm is the matrix after multiplications, its rows are identical at a factor.
2900: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2901: * Output is prlim.
2902: * Initial matrix pimij
2903: */
1.206 brouard 2904: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2905: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2906: /* 0, 0 , 1} */
2907: /*
2908: * and after some iteration: */
2909: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2910: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2911: /* 0, 0 , 1} */
2912: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2913: /* {0.51571254859325999, 0.4842874514067399, */
2914: /* 0.51326036147820708, 0.48673963852179264} */
2915: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2916:
1.332 brouard 2917: int i, ii,j,k, k1;
1.209 brouard 2918: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2919: /* double **matprod2(); */ /* test */
1.218 brouard 2920: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2921: double **newm;
1.209 brouard 2922: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2923: int ncvloop=0;
1.288 brouard 2924: int first=0;
1.169 brouard 2925:
1.209 brouard 2926: min=vector(1,nlstate);
2927: max=vector(1,nlstate);
2928: meandiff=vector(1,nlstate);
2929:
1.218 brouard 2930: /* Starting with matrix unity */
1.126 brouard 2931: for (ii=1;ii<=nlstate+ndeath;ii++)
2932: for (j=1;j<=nlstate+ndeath;j++){
2933: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2934: }
1.169 brouard 2935:
2936: cov[1]=1.;
2937:
2938: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2939: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2940: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2941: ncvloop++;
1.126 brouard 2942: newm=savm;
2943: /* Covariates have to be included here again */
1.138 brouard 2944: cov[2]=agefin;
1.319 brouard 2945: if(nagesqr==1){
2946: cov[3]= agefin*agefin;
2947: }
1.332 brouard 2948: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
2949: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
2950: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
2951: if(Typevar[k1]==1){ /* A product with age */
2952: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
2953: }else{
2954: cov[2+nagesqr+k1]=precov[nres][k1];
2955: }
2956: }/* End of loop on model equation */
2957:
2958: /* Start of old code (replaced by a loop on position in the model equation */
2959: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
2960: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
2961: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
2962: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
2963: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
2964: /* * k 1 2 3 4 5 6 7 8 */
2965: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
2966: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
2967: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
2968: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
2969: /* *nsd=3 (1) (2) (3) */
2970: /* *TvarsD[nsd] [1]=2 1 3 */
2971: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
2972: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
2973: /* *Tage[] [1]=1 [2]=2 [3]=3 */
2974: /* *Tvard[] [1][1]=1 [2][1]=1 */
2975: /* * [1][2]=3 [2][2]=2 */
2976: /* *Tprod[](=k) [1]=1 [2]=8 */
2977: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
2978: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
2979: /* *TvarsDpType */
2980: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
2981: /* * nsd=1 (1) (2) */
2982: /* *TvarsD[nsd] 3 2 */
2983: /* *TnsdVar (3)=1 (2)=2 */
2984: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
2985: /* *Tage[] [1]=2 [2]= 3 */
2986: /* *\/ */
2987: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
2988: /* /\* printf("prevalim Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,Tvar[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); *\/ */
2989: /* } */
2990: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
2991: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
2992: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
2993: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
2994: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
2995: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2996: /* /\* printf("prevalim Quantitative k=%d TvarsQind[%d]=%d, TvarsQ[%d]=V%d,Tqresult[%d][%d]=%f\n",k,k,TvarsQind[k],k,TvarsQ[k],nres,k,Tqresult[nres][k]); *\/ */
2997: /* } */
2998: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
2999: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
3000: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3001: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
3002: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
3003: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3004: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
3005: /* } */
3006: /* /\* printf("prevalim Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); *\/ */
3007: /* } */
3008: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3009: /* /\* printf("prevalim Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2]); *\/ */
3010: /* if(Dummy[Tvard[k][1]]==0){ */
3011: /* if(Dummy[Tvard[k][2]]==0){ */
3012: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3013: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3014: /* }else{ */
3015: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3016: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
3017: /* } */
3018: /* }else{ */
3019: /* if(Dummy[Tvard[k][2]]==0){ */
3020: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3021: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
3022: /* }else{ */
3023: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3024: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
3025: /* } */
3026: /* } */
3027: /* } /\* End product without age *\/ */
3028: /* ENd of old code */
1.138 brouard 3029: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3030: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3031: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 3032: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3033: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 3034: /* age and covariate values of ij are in 'cov' */
1.142 brouard 3035: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 3036:
1.126 brouard 3037: savm=oldm;
3038: oldm=newm;
1.209 brouard 3039:
3040: for(j=1; j<=nlstate; j++){
3041: max[j]=0.;
3042: min[j]=1.;
3043: }
3044: for(i=1;i<=nlstate;i++){
3045: sumnew=0;
3046: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
3047: for(j=1; j<=nlstate; j++){
3048: prlim[i][j]= newm[i][j]/(1-sumnew);
3049: max[j]=FMAX(max[j],prlim[i][j]);
3050: min[j]=FMIN(min[j],prlim[i][j]);
3051: }
3052: }
3053:
1.126 brouard 3054: maxmax=0.;
1.209 brouard 3055: for(j=1; j<=nlstate; j++){
3056: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
3057: maxmax=FMAX(maxmax,meandiff[j]);
3058: /* printf(" age= %d meandiff[%d]=%f, agefin=%d max[%d]=%f min[%d]=%f maxmax=%f\n", (int)age, j, meandiff[j],(int)agefin, j, max[j], j, min[j],maxmax); */
1.169 brouard 3059: } /* j loop */
1.203 brouard 3060: *ncvyear= (int)age- (int)agefin;
1.208 brouard 3061: /* printf("maxmax=%lf maxmin=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, maxmin, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.126 brouard 3062: if(maxmax < ftolpl){
1.209 brouard 3063: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
3064: free_vector(min,1,nlstate);
3065: free_vector(max,1,nlstate);
3066: free_vector(meandiff,1,nlstate);
1.126 brouard 3067: return prlim;
3068: }
1.288 brouard 3069: } /* agefin loop */
1.208 brouard 3070: /* After some age loop it doesn't converge */
1.288 brouard 3071: if(!first){
3072: first=1;
3073: printf("Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d). Others in log file only...\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
1.317 brouard 3074: fprintf(ficlog, "Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d).\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
3075: }else if (first >=1 && first <10){
3076: fprintf(ficlog, "Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d).\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
3077: first++;
3078: }else if (first ==10){
3079: fprintf(ficlog, "Warning: the stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.d years and %d loops. Try to lower 'ftolpl'. Youngest age to start was %d=(%d-%d).\n", (int)age, maxmax, ftolpl, *ncvyear, ncvloop, (int)(agefin+stepm/YEARM), (int)(age-stepm/YEARM), (int)delaymax);
3080: printf("Warning: the stable prevalence dit not converge. This warning came too often, IMaCh will stop notifying, even in its log file. Look at the graphs to appreciate the non convergence.\n");
3081: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
3082: first++;
1.288 brouard 3083: }
3084:
1.209 brouard 3085: /* Try to lower 'ftol', for example from 1.e-8 to 6.e-9.\n", ftolpl, (int)age, (int)delaymax, (int)agefin, ncvloop, (int)age-(int)agefin); */
3086: free_vector(min,1,nlstate);
3087: free_vector(max,1,nlstate);
3088: free_vector(meandiff,1,nlstate);
1.208 brouard 3089:
1.169 brouard 3090: return prlim; /* should not reach here */
1.126 brouard 3091: }
3092:
1.217 brouard 3093:
3094: /**** Back Prevalence limit (stable or period prevalence) ****************/
3095:
1.218 brouard 3096: /* double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ageminpar, double agemaxpar, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, double ftolpl, int *ncvyear, int ij) */
3097: /* double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, double ftolpl, int *ncvyear, int ij) */
1.242 brouard 3098: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 3099: {
1.264 brouard 3100: /* Computes the prevalence limit in each live state at age x and for covariate combination ij (<=2**cptcoveff) by left multiplying the unit
1.217 brouard 3101: matrix by transitions matrix until convergence is reached with precision ftolpl */
3102: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
3103: /* Wx is row vector: population in state 1, population in state 2, population dead */
3104: /* or prevalence in state 1, prevalence in state 2, 0 */
3105: /* newm is the matrix after multiplications, its rows are identical at a factor */
3106: /* Initial matrix pimij */
3107: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
3108: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
3109: /* 0, 0 , 1} */
3110: /*
3111: * and after some iteration: */
3112: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
3113: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
3114: /* 0, 0 , 1} */
3115: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
3116: /* {0.51571254859325999, 0.4842874514067399, */
3117: /* 0.51326036147820708, 0.48673963852179264} */
3118: /* If we start from prlim again, prlim tends to a constant matrix */
3119:
1.332 brouard 3120: int i, ii,j,k, k1;
1.247 brouard 3121: int first=0;
1.217 brouard 3122: double *min, *max, *meandiff, maxmax,sumnew=0.;
3123: /* double **matprod2(); */ /* test */
3124: double **out, cov[NCOVMAX+1], **bmij();
3125: double **newm;
1.218 brouard 3126: double **dnewm, **doldm, **dsavm; /* for use */
3127: double **oldm, **savm; /* for use */
3128:
1.217 brouard 3129: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
3130: int ncvloop=0;
3131:
3132: min=vector(1,nlstate);
3133: max=vector(1,nlstate);
3134: meandiff=vector(1,nlstate);
3135:
1.266 brouard 3136: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
3137: oldm=oldms; savm=savms;
3138:
3139: /* Starting with matrix unity */
3140: for (ii=1;ii<=nlstate+ndeath;ii++)
3141: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 3142: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3143: }
3144:
3145: cov[1]=1.;
3146:
3147: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3148: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 3149: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 3150: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
3151: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 3152: ncvloop++;
1.218 brouard 3153: newm=savm; /* oldm should be kept from previous iteration or unity at start */
3154: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 3155: /* Covariates have to be included here again */
3156: cov[2]=agefin;
1.319 brouard 3157: if(nagesqr==1){
1.217 brouard 3158: cov[3]= agefin*agefin;;
1.319 brouard 3159: }
1.332 brouard 3160: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3161: if(Typevar[k1]==1){ /* A product with age */
3162: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 3163: }else{
1.332 brouard 3164: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 3165: }
1.332 brouard 3166: }/* End of loop on model equation */
3167:
3168: /* Old code */
3169:
3170: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
3171: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
3172: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
3173: /* /\* printf("bprevalim Dummy agefin=%.0f combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov[%d]=%lf codtabm(%d,Tvar[%d])=%d \n",agefin,ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],2+nagesqr+TvarsDind[k],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); *\/ */
3174: /* } */
3175: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
3176: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3177: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3178: /* /\* /\\* printf("prevalim ij=%d k=%d Tvar[%d]=%d nbcode=%d cov=%lf codtabm(%d,Tvar[%d])=%d \n",ij,k, k, Tvar[k],nbcode[Tvar[k]][codtabm(ij,Tvar[k])],cov[2+k], ij, k, codtabm(ij,Tvar[k])]); *\\/ *\/ */
3179: /* /\* } *\/ */
3180: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3181: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3182: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3183: /* /\* printf("prevalim Quantitative k=%d TvarsQind[%d]=%d, TvarsQ[%d]=V%d,Tqresult[%d][%d]=%f\n",k,k,TvarsQind[k],k,TvarsQ[k],nres,k,Tqresult[nres][k]); *\/ */
3184: /* } */
3185: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
3186: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
3187: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
3188: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3189: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3190: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
3191: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3192: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3193: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3194: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3195: /* } */
3196: /* /\* printf("prevalim Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); *\/ */
3197: /* } */
3198: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3199: /* /\* printf("prevalim Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2]); *\/ */
3200: /* if(Dummy[Tvard[k][1]]==0){ */
3201: /* if(Dummy[Tvard[k][2]]==0){ */
3202: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3203: /* }else{ */
3204: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3205: /* } */
3206: /* }else{ */
3207: /* if(Dummy[Tvard[k][2]]==0){ */
3208: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3209: /* }else{ */
3210: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3211: /* } */
3212: /* } */
3213: /* } */
1.217 brouard 3214:
3215: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3216: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3217: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3218: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3219: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3220: /* ij should be linked to the correct index of cov */
3221: /* age and covariate values ij are in 'cov', but we need to pass
3222: * ij for the observed prevalence at age and status and covariate
3223: * number: prevacurrent[(int)agefin][ii][ij]
3224: */
3225: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, ageminpar, agemaxpar, dnewm, doldm, dsavm,ij)); /\* Bug Valgrind *\/ */
3226: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij)); /\* Bug Valgrind *\/ */
3227: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij)); /* Bug Valgrind */
1.268 brouard 3228: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3229: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3230: /* for(i=1; i<=nlstate+ndeath; i++) { */
3231: /* printf("%d newm= ",i); */
3232: /* for(j=1;j<=nlstate+ndeath;j++) { */
3233: /* printf("%f ",newm[i][j]); */
3234: /* } */
3235: /* printf("oldm * "); */
3236: /* for(j=1;j<=nlstate+ndeath;j++) { */
3237: /* printf("%f ",oldm[i][j]); */
3238: /* } */
1.268 brouard 3239: /* printf(" bmmij "); */
1.266 brouard 3240: /* for(j=1;j<=nlstate+ndeath;j++) { */
3241: /* printf("%f ",pmmij[i][j]); */
3242: /* } */
3243: /* printf("\n"); */
3244: /* } */
3245: /* } */
1.217 brouard 3246: savm=oldm;
3247: oldm=newm;
1.266 brouard 3248:
1.217 brouard 3249: for(j=1; j<=nlstate; j++){
3250: max[j]=0.;
3251: min[j]=1.;
3252: }
3253: for(j=1; j<=nlstate; j++){
3254: for(i=1;i<=nlstate;i++){
1.234 brouard 3255: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3256: bprlim[i][j]= newm[i][j];
3257: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3258: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3259: }
3260: }
1.218 brouard 3261:
1.217 brouard 3262: maxmax=0.;
3263: for(i=1; i<=nlstate; i++){
1.318 brouard 3264: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3265: maxmax=FMAX(maxmax,meandiff[i]);
3266: /* printf("Back age= %d meandiff[%d]=%f, agefin=%d max[%d]=%f min[%d]=%f maxmax=%f\n", (int)age, i, meandiff[i],(int)agefin, i, max[i], i, min[i],maxmax); */
1.268 brouard 3267: } /* i loop */
1.217 brouard 3268: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3269: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3270: if(maxmax < ftolpl){
1.220 brouard 3271: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3272: free_vector(min,1,nlstate);
3273: free_vector(max,1,nlstate);
3274: free_vector(meandiff,1,nlstate);
3275: return bprlim;
3276: }
1.288 brouard 3277: } /* agefin loop */
1.217 brouard 3278: /* After some age loop it doesn't converge */
1.288 brouard 3279: if(!first){
1.247 brouard 3280: first=1;
3281: printf("Warning: the back stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. Others in log file only...\n\
3282: Oldest age to start was %d-%d=%d, ncvloop=%d, ncvyear=%d\n", (int)age, maxmax, ftolpl, delaymax, (int)age, (int)delaymax, (int)agefin, ncvloop, *ncvyear);
3283: }
3284: fprintf(ficlog,"Warning: the back stable prevalence at age %d did not converge with the required precision (%g > ftolpl=%g) within %.0f years. Try to lower 'ftolpl'. \n\
1.217 brouard 3285: Oldest age to start was %d-%d=%d, ncvloop=%d, ncvyear=%d\n", (int)age, maxmax, ftolpl, delaymax, (int)age, (int)delaymax, (int)agefin, ncvloop, *ncvyear);
3286: /* Try to lower 'ftol', for example from 1.e-8 to 6.e-9.\n", ftolpl, (int)age, (int)delaymax, (int)agefin, ncvloop, (int)age-(int)agefin); */
3287: free_vector(min,1,nlstate);
3288: free_vector(max,1,nlstate);
3289: free_vector(meandiff,1,nlstate);
3290:
3291: return bprlim; /* should not reach here */
3292: }
3293:
1.126 brouard 3294: /*************** transition probabilities ***************/
3295:
3296: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3297: {
1.138 brouard 3298: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3299: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3300: model to the ncovmodel covariates (including constant and age).
3301: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3302: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3303: ncth covariate in the global vector x is given by the formula:
3304: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3305: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3306: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3307: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3308: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3309: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3310: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3311: */
3312: double s1, lnpijopii;
1.126 brouard 3313: /*double t34;*/
1.164 brouard 3314: int i,j, nc, ii, jj;
1.126 brouard 3315:
1.223 brouard 3316: for(i=1; i<= nlstate; i++){
3317: for(j=1; j<i;j++){
3318: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3319: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3320: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3321: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3322: }
3323: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3324: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3325: }
3326: for(j=i+1; j<=nlstate+ndeath;j++){
3327: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3328: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3329: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3330: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3331: }
3332: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3333: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3334: }
3335: }
1.218 brouard 3336:
1.223 brouard 3337: for(i=1; i<= nlstate; i++){
3338: s1=0;
3339: for(j=1; j<i; j++){
1.339 brouard 3340: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3341: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3342: }
3343: for(j=i+1; j<=nlstate+ndeath; j++){
1.339 brouard 3344: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3345: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3346: }
3347: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3348: ps[i][i]=1./(s1+1.);
3349: /* Computing other pijs */
3350: for(j=1; j<i; j++)
1.325 brouard 3351: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3352: for(j=i+1; j<=nlstate+ndeath; j++)
3353: ps[i][j]= exp(ps[i][j])*ps[i][i];
3354: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3355: } /* end i */
1.218 brouard 3356:
1.223 brouard 3357: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3358: for(jj=1; jj<= nlstate+ndeath; jj++){
3359: ps[ii][jj]=0;
3360: ps[ii][ii]=1;
3361: }
3362: }
1.294 brouard 3363:
3364:
1.223 brouard 3365: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3366: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3367: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3368: /* } */
3369: /* printf("\n "); */
3370: /* } */
3371: /* printf("\n ");printf("%lf ",cov[2]);*/
3372: /*
3373: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3374: goto end;*/
1.266 brouard 3375: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3376: }
3377:
1.218 brouard 3378: /*************** backward transition probabilities ***************/
3379:
3380: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ageminpar, double agemaxpar, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3381: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3382: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3383: {
1.302 brouard 3384: /* Computes the backward probability at age agefin, cov[2], and covariate combination 'ij'. In fact cov is already filled and x too.
1.266 brouard 3385: * Call to pmij(cov and x), call to cross prevalence, sums and inverses, left multiply, and returns in **ps as well as **bmij.
1.222 brouard 3386: */
1.218 brouard 3387: int i, ii, j,k;
1.222 brouard 3388:
3389: double **out, **pmij();
3390: double sumnew=0.;
1.218 brouard 3391: double agefin;
1.292 brouard 3392: double k3=0.; /* constant of the w_x diagonal matrix (in order for B to sum to 1 even for death state) */
1.222 brouard 3393: double **dnewm, **dsavm, **doldm;
3394: double **bbmij;
3395:
1.218 brouard 3396: doldm=ddoldms; /* global pointers */
1.222 brouard 3397: dnewm=ddnewms;
3398: dsavm=ddsavms;
1.318 brouard 3399:
3400: /* Debug */
3401: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3402: agefin=cov[2];
1.268 brouard 3403: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3404: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3405: the observed prevalence (with this covariate ij) at beginning of transition */
3406: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3407:
3408: /* P_x */
1.325 brouard 3409: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3410: /* outputs pmmij which is a stochastic matrix in row */
3411:
3412: /* Diag(w_x) */
1.292 brouard 3413: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3414: sumnew=0.;
1.269 brouard 3415: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3416: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3417: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3418: sumnew+=prevacurrent[(int)agefin][ii][ij];
3419: }
3420: if(sumnew >0.01){ /* At least some value in the prevalence */
3421: for (ii=1;ii<=nlstate+ndeath;ii++){
3422: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3423: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3424: }
3425: }else{
3426: for (ii=1;ii<=nlstate+ndeath;ii++){
3427: for (j=1;j<=nlstate+ndeath;j++)
3428: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3429: }
3430: /* if(sumnew <0.9){ */
3431: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3432: /* } */
3433: }
3434: k3=0.0; /* We put the last diagonal to 0 */
3435: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3436: doldm[ii][ii]= k3;
3437: }
3438: /* End doldm, At the end doldm is diag[(w_i)] */
3439:
1.292 brouard 3440: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3441: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3442:
1.292 brouard 3443: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3444: /* w1 p11 + w2 p21 only on live states N1./N..*N11/N1. + N2./N..*N21/N2.=(N11+N21)/N..=N.1/N.. */
1.222 brouard 3445: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3446: sumnew=0.;
1.222 brouard 3447: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3448: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3449: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3450: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3451: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3452: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3453: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3454: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3455: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3456: /* }else */
1.268 brouard 3457: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3458: } /*End ii */
3459: } /* End j, At the end dsavm is diag[1/(w_1p1i+w_2 p2i)] for ALL states even if the sum is only for live states */
3460:
1.292 brouard 3461: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3462: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3463: /* end bmij */
1.266 brouard 3464: return ps; /*pointer is unchanged */
1.218 brouard 3465: }
1.217 brouard 3466: /*************** transition probabilities ***************/
3467:
1.218 brouard 3468: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3469: {
3470: /* According to parameters values stored in x and the covariate's values stored in cov,
3471: computes the probability to be observed in state j being in state i by appying the
3472: model to the ncovmodel covariates (including constant and age).
3473: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3474: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3475: ncth covariate in the global vector x is given by the formula:
3476: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3477: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3478: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3479: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3480: Outputs ps[i][j] the probability to be observed in j being in j according to
3481: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3482: */
3483: double s1, lnpijopii;
3484: /*double t34;*/
3485: int i,j, nc, ii, jj;
3486:
1.234 brouard 3487: for(i=1; i<= nlstate; i++){
3488: for(j=1; j<i;j++){
3489: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3490: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3491: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3492: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3493: }
3494: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3495: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3496: }
3497: for(j=i+1; j<=nlstate+ndeath;j++){
3498: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3499: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3500: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3501: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3502: }
3503: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3504: }
3505: }
3506:
3507: for(i=1; i<= nlstate; i++){
3508: s1=0;
3509: for(j=1; j<i; j++){
3510: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3511: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3512: }
3513: for(j=i+1; j<=nlstate+ndeath; j++){
3514: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3515: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3516: }
3517: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3518: ps[i][i]=1./(s1+1.);
3519: /* Computing other pijs */
3520: for(j=1; j<i; j++)
3521: ps[i][j]= exp(ps[i][j])*ps[i][i];
3522: for(j=i+1; j<=nlstate+ndeath; j++)
3523: ps[i][j]= exp(ps[i][j])*ps[i][i];
3524: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3525: } /* end i */
3526:
3527: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3528: for(jj=1; jj<= nlstate+ndeath; jj++){
3529: ps[ii][jj]=0;
3530: ps[ii][ii]=1;
3531: }
3532: }
1.296 brouard 3533: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3534: for(jj=1; jj<= nlstate+ndeath; jj++){
3535: s1=0.;
3536: for(ii=1; ii<= nlstate+ndeath; ii++){
3537: s1+=ps[ii][jj];
3538: }
3539: for(ii=1; ii<= nlstate; ii++){
3540: ps[ii][jj]=ps[ii][jj]/s1;
3541: }
3542: }
3543: /* Transposition */
3544: for(jj=1; jj<= nlstate+ndeath; jj++){
3545: for(ii=jj; ii<= nlstate+ndeath; ii++){
3546: s1=ps[ii][jj];
3547: ps[ii][jj]=ps[jj][ii];
3548: ps[jj][ii]=s1;
3549: }
3550: }
3551: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3552: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3553: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3554: /* } */
3555: /* printf("\n "); */
3556: /* } */
3557: /* printf("\n ");printf("%lf ",cov[2]);*/
3558: /*
3559: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3560: goto end;*/
3561: return ps;
1.217 brouard 3562: }
3563:
3564:
1.126 brouard 3565: /**************** Product of 2 matrices ******************/
3566:
1.145 brouard 3567: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3568: {
3569: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3570: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3571: /* in, b, out are matrice of pointers which should have been initialized
3572: before: only the contents of out is modified. The function returns
3573: a pointer to pointers identical to out */
1.145 brouard 3574: int i, j, k;
1.126 brouard 3575: for(i=nrl; i<= nrh; i++)
1.145 brouard 3576: for(k=ncolol; k<=ncoloh; k++){
3577: out[i][k]=0.;
3578: for(j=ncl; j<=nch; j++)
3579: out[i][k] +=in[i][j]*b[j][k];
3580: }
1.126 brouard 3581: return out;
3582: }
3583:
3584:
3585: /************* Higher Matrix Product ***************/
3586:
1.235 brouard 3587: double ***hpxij(double ***po, int nhstepm, double age, int hstepm, double *x, int nlstate, int stepm, double **oldm, double **savm, int ij, int nres )
1.126 brouard 3588: {
1.336 brouard 3589: /* Already optimized with precov.
3590: Computes the transition matrix starting at age 'age' and dummies values in each resultline (loop on ij to find the corresponding combination) to over
1.126 brouard 3591: 'nhstepm*hstepm*stepm' months (i.e. until
3592: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3593: nhstepm*hstepm matrices.
3594: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3595: (typically every 2 years instead of every month which is too big
3596: for the memory).
3597: Model is determined by parameters x and covariates have to be
3598: included manually here.
3599:
3600: */
3601:
1.330 brouard 3602: int i, j, d, h, k, k1;
1.131 brouard 3603: double **out, cov[NCOVMAX+1];
1.126 brouard 3604: double **newm;
1.187 brouard 3605: double agexact;
1.214 brouard 3606: double agebegin, ageend;
1.126 brouard 3607:
3608: /* Hstepm could be zero and should return the unit matrix */
3609: for (i=1;i<=nlstate+ndeath;i++)
3610: for (j=1;j<=nlstate+ndeath;j++){
3611: oldm[i][j]=(i==j ? 1.0 : 0.0);
3612: po[i][j][0]=(i==j ? 1.0 : 0.0);
3613: }
3614: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3615: for(h=1; h <=nhstepm; h++){
3616: for(d=1; d <=hstepm; d++){
3617: newm=savm;
3618: /* Covariates have to be included here again */
3619: cov[1]=1.;
1.214 brouard 3620: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3621: cov[2]=agexact;
1.319 brouard 3622: if(nagesqr==1){
1.227 brouard 3623: cov[3]= agexact*agexact;
1.319 brouard 3624: }
1.330 brouard 3625: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
3626: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
3627: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.332 brouard 3628: if(Typevar[k1]==1){ /* A product with age */
3629: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
3630: }else{
3631: cov[2+nagesqr+k1]=precov[nres][k1];
3632: }
3633: }/* End of loop on model equation */
3634: /* Old code */
3635: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
3636: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
3637: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
3638: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
3639: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
3640: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3641: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3642: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
3643: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
3644: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
3645: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
3646: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
3647: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3648: /* /\* printf("hpxij Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,TnsdVar[TvarsD[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,TnsdVar[TvarsD[k]])); *\/ */
3649: /* printf("hpxij Dummy combi=%d k1=%d Tvar[%d]=V%d cov[2+%d+%d]=%lf resultmodel[nres][%d]=%d nres/nresult=%d/%d \n",ij,k1,k1, Tvar[k1],nagesqr,k1,cov[2+nagesqr+k1],k1,resultmodel[nres][k1],nres,nresult); */
3650: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3651: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
3652: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
3653: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
3654: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
3655: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
3656: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3657: /* printf("hPxij Quantitative k1=%d resultmodel[nres][%d]=%d,Tqresult[%d][%d]=%f\n",k1,k1,resultmodel[nres][k1],nres,resultmodel[nres][k1],Tqresult[nres][resultmodel[nres][k1]]); */
3658: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3659: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
3660: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
3661: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
3662: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
3663: /* printf("DhPxij Dummy with age k1=%d Tvar[%d]=%d TinvDoQresult[nres=%d][%d]=%.f age=%.2f,cov[2+%d+%d]=%.3f\n",k1,k1,Tvar[k1],nres,TinvDoQresult[nres][Tvar[k1]],cov[2],nagesqr,k1,cov[2+nagesqr+k1]); */
3664: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3665:
3666: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
3667: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
3668: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
3669: /* /\* *\/ */
1.330 brouard 3670: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3671: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3672: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 3673: /* /\*cptcovage=2 1 2 *\/ */
3674: /* /\*Tage[k]= 5 8 *\/ */
3675: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
3676: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3677: /* printf("QhPxij Quant with age k1=%d resultmodel[nres][%d]=%d,Tqresult[%d][%d]=%f\n",k1,k1,resultmodel[nres][k1],nres,resultmodel[nres][k1],Tqresult[nres][resultmodel[nres][k1]]); */
3678: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3679: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
3680: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
3681: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
3682: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
3683: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
3684: /* /\* printf("hPxij Age combi=%d k=%d cptcovage=%d Tage[%d]=%d Tvar[Tage[%d]]=V%d nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[Tvar[Tage[k]]]])]=%d nres=%d\n",ij,k,cptcovage,k,Tage[k],k,Tvar[Tage[k]], nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[Tvar[Tage[k]]])],nres); *\/ */
3685: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
3686: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
3687: /* /\* } *\/ */
3688: /* /\* printf("hPxij Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); *\/ */
3689: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
3690: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
3691: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
3692: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
3693: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
3694: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
3695: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
3696: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
3697: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 3698:
1.332 brouard 3699: /* /\* printf("hPxij Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]=%d nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]=%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2],nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])],nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]); *\/ */
3700: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3701: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
3702: /* printf("hPxij Prod ij=%d k1=%d cov[2+%d+%d]=%.5f Tvard[%d][1]=V%d * Tvard[%d][2]=V%d ; TinvDoQresult[nres][Tvardk[k1][1]]=%.4f * TinvDoQresult[nres][Tvardk[k1][1]]=%.4f\n",ij,k1,nagesqr,k1,cov[2+nagesqr+k1],k1,Tvardk[k1][1], k1,Tvardk[k1][2], TinvDoQresult[nres][Tvardk[k1][1]], TinvDoQresult[nres][Tvardk[k1][2]]); */
3703: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3704:
3705: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
3706: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
3707: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3708: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
3709: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])]; *\/ */
3710: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
3711: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
3712: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
3713: /* /\* } *\/ */
3714: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
3715: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
3716: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
3717: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3718: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
3719: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
3720: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3721: /* /\* } *\/ */
3722: /* /\* }/\\*end of products quantitative *\\/ *\/ */
3723: /* }/\*end of products *\/ */
3724: /* } /\* End of loop on model equation *\/ */
1.235 brouard 3725: /* for (k=1; k<=cptcovn;k++) */
3726: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3727: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3728: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3729: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3730: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3731:
3732:
1.126 brouard 3733: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3734: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3735: /* right multiplication of oldm by the current matrix */
1.126 brouard 3736: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3737: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3738: /* if((int)age == 70){ */
3739: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3740: /* for(i=1; i<=nlstate+ndeath; i++) { */
3741: /* printf("%d pmmij ",i); */
3742: /* for(j=1;j<=nlstate+ndeath;j++) { */
3743: /* printf("%f ",pmmij[i][j]); */
3744: /* } */
3745: /* printf(" oldm "); */
3746: /* for(j=1;j<=nlstate+ndeath;j++) { */
3747: /* printf("%f ",oldm[i][j]); */
3748: /* } */
3749: /* printf("\n"); */
3750: /* } */
3751: /* } */
1.126 brouard 3752: savm=oldm;
3753: oldm=newm;
3754: }
3755: for(i=1; i<=nlstate+ndeath; i++)
3756: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3757: po[i][j][h]=newm[i][j];
3758: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3759: }
1.128 brouard 3760: /*printf("h=%d ",h);*/
1.126 brouard 3761: } /* end h */
1.267 brouard 3762: /* printf("\n H=%d \n",h); */
1.126 brouard 3763: return po;
3764: }
3765:
1.217 brouard 3766: /************* Higher Back Matrix Product ***************/
1.218 brouard 3767: /* double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, double **oldm, double **savm, double **dnewm, double **doldm, double **dsavm, int ij ) */
1.267 brouard 3768: double ***hbxij(double ***po, int nhstepm, double age, int hstepm, double *x, double ***prevacurrent, int nlstate, int stepm, int ij, int nres )
1.217 brouard 3769: {
1.332 brouard 3770: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
3771: computes the transition matrix starting at age 'age' over
1.217 brouard 3772: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3773: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3774: nhstepm*hstepm matrices.
3775: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3776: (typically every 2 years instead of every month which is too big
1.217 brouard 3777: for the memory).
1.218 brouard 3778: Model is determined by parameters x and covariates have to be
1.266 brouard 3779: included manually here. Then we use a call to bmij(x and cov)
3780: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3781: */
1.217 brouard 3782:
1.332 brouard 3783: int i, j, d, h, k, k1;
1.266 brouard 3784: double **out, cov[NCOVMAX+1], **bmij();
3785: double **newm, ***newmm;
1.217 brouard 3786: double agexact;
3787: double agebegin, ageend;
1.222 brouard 3788: double **oldm, **savm;
1.217 brouard 3789:
1.266 brouard 3790: newmm=po; /* To be saved */
3791: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3792: /* Hstepm could be zero and should return the unit matrix */
3793: for (i=1;i<=nlstate+ndeath;i++)
3794: for (j=1;j<=nlstate+ndeath;j++){
3795: oldm[i][j]=(i==j ? 1.0 : 0.0);
3796: po[i][j][0]=(i==j ? 1.0 : 0.0);
3797: }
3798: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3799: for(h=1; h <=nhstepm; h++){
3800: for(d=1; d <=hstepm; d++){
3801: newm=savm;
3802: /* Covariates have to be included here again */
3803: cov[1]=1.;
1.271 brouard 3804: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3805: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3806: /* Debug */
3807: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3808: cov[2]=agexact;
1.332 brouard 3809: if(nagesqr==1){
1.222 brouard 3810: cov[3]= agexact*agexact;
1.332 brouard 3811: }
3812: /** New code */
3813: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3814: if(Typevar[k1]==1){ /* A product with age */
3815: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 3816: }else{
1.332 brouard 3817: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 3818: }
1.332 brouard 3819: }/* End of loop on model equation */
3820: /** End of new code */
3821: /** This was old code */
3822: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
3823: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3824: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3825: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
3826: /* /\* printf("hbxij Dummy agexact=%.0f combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov[%d]=%lf codtabm(%d,Tvar[%d])=%d \n",agexact,ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],2+nagesqr+TvarsDind[k],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); *\/ */
3827: /* } */
3828: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3829: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3830: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3831: /* /\* printf("hPxij Quantitative k=%d TvarsQind[%d]=%d, TvarsQ[%d]=V%d,Tqresult[%d][%d]=%f\n",k,k,TvarsQind[k],k,TvarsQ[k],nres,k,Tqresult[nres][k]); *\/ */
3832: /* } */
3833: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
3834: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
3835: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3836: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3837: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3838: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3839: /* } */
3840: /* /\* printf("hBxij Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); *\/ */
3841: /* } */
3842: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
3843: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3844: /* if(Dummy[Tvard[k][1]]==0){ */
3845: /* if(Dummy[Tvard[k][2]]==0){ */
3846: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
3847: /* }else{ */
3848: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3849: /* } */
3850: /* }else{ */
3851: /* if(Dummy[Tvard[k][2]]==0){ */
3852: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3853: /* }else{ */
3854: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3855: /* } */
3856: /* } */
3857: /* } */
3858: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
3859: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
3860: /** End of old code */
3861:
1.218 brouard 3862: /* Careful transposed matrix */
1.266 brouard 3863: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3864: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3865: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3866: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3867: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3868: /* if((int)age == 70){ */
3869: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3870: /* for(i=1; i<=nlstate+ndeath; i++) { */
3871: /* printf("%d pmmij ",i); */
3872: /* for(j=1;j<=nlstate+ndeath;j++) { */
3873: /* printf("%f ",pmmij[i][j]); */
3874: /* } */
3875: /* printf(" oldm "); */
3876: /* for(j=1;j<=nlstate+ndeath;j++) { */
3877: /* printf("%f ",oldm[i][j]); */
3878: /* } */
3879: /* printf("\n"); */
3880: /* } */
3881: /* } */
3882: savm=oldm;
3883: oldm=newm;
3884: }
3885: for(i=1; i<=nlstate+ndeath; i++)
3886: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3887: po[i][j][h]=newm[i][j];
1.268 brouard 3888: /* if(h==nhstepm) */
3889: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3890: }
1.268 brouard 3891: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3892: } /* end h */
1.268 brouard 3893: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3894: return po;
3895: }
3896:
3897:
1.162 brouard 3898: #ifdef NLOPT
3899: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3900: double fret;
3901: double *xt;
3902: int j;
3903: myfunc_data *d2 = (myfunc_data *) pd;
3904: /* xt = (p1-1); */
3905: xt=vector(1,n);
3906: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3907:
3908: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3909: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3910: printf("Function = %.12lf ",fret);
3911: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3912: printf("\n");
3913: free_vector(xt,1,n);
3914: return fret;
3915: }
3916: #endif
1.126 brouard 3917:
3918: /*************** log-likelihood *************/
3919: double func( double *x)
3920: {
1.336 brouard 3921: int i, ii, j, k, mi, d, kk, kf=0;
1.226 brouard 3922: int ioffset=0;
1.339 brouard 3923: int ipos=0,iposold=0,ncovv=0;
3924:
1.340 brouard 3925: double cotvarv, cotvarvold;
1.226 brouard 3926: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3927: double **out;
3928: double lli; /* Individual log likelihood */
3929: int s1, s2;
1.228 brouard 3930: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
1.336 brouard 3931:
1.226 brouard 3932: double bbh, survp;
3933: double agexact;
1.336 brouard 3934: double agebegin, ageend;
1.226 brouard 3935: /*extern weight */
3936: /* We are differentiating ll according to initial status */
3937: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3938: /*for(i=1;i<imx;i++)
3939: printf(" %d\n",s[4][i]);
3940: */
1.162 brouard 3941:
1.226 brouard 3942: ++countcallfunc;
1.162 brouard 3943:
1.226 brouard 3944: cov[1]=1.;
1.126 brouard 3945:
1.226 brouard 3946: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3947: ioffset=0;
1.226 brouard 3948: if(mle==1){
3949: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3950: /* Computes the values of the ncovmodel covariates of the model
3951: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3952: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3953: to be observed in j being in i according to the model.
3954: */
1.243 brouard 3955: ioffset=2+nagesqr ;
1.233 brouard 3956: /* Fixed */
1.345 brouard 3957: for (kf=1; kf<=ncovf;kf++){ /* For each fixed covariate dummy or quant or prod */
1.319 brouard 3958: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3959: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3960: /* TvarF[1]=Tvar[6]=2, TvarF[2]=Tvar[7]=7, TvarF[3]=Tvar[9]=1 ID of fixed covariates or product V2, V1*V2, V1 */
1.320 brouard 3961: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.336 brouard 3962: cov[ioffset+TvarFind[kf]]=covar[Tvar[TvarFind[kf]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (TvarFind[1]=6)*/
1.319 brouard 3963: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3964: }
1.226 brouard 3965: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3966: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3967: has been calculated etc */
3968: /* For an individual i, wav[i] gives the number of effective waves */
3969: /* We compute the contribution to Likelihood of each effective transition
3970: mw[mi][i] is real wave of the mi th effectve wave */
3971: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3972: s2=s[mw[mi+1][i]][i];
1.341 brouard 3973: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i] because now is moved after nvocol+nqv
1.226 brouard 3974: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3975: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3976: */
1.336 brouard 3977: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
3978: /* Wave varying (but not age varying) */
1.339 brouard 3979: /* for(k=1; k <= ncovv ; k++){ /\* Varying covariates in the model (single and product but no age )"V5+V4+V3+V4*V3+V5*age+V1*age+V1" +TvarVind 1,2,3,4(V4*V3) Tvar[1]@7{5, 4, 3, 6, 5, 1, 1 ; 6 because the created covar is after V5 and is 6, minus 1+1, 3,2,1,4 positions in cotvar*\/ */
3980: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? *\/ */
3981: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
3982: /* } */
1.340 brouard 3983: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age )*/
3984: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
3985: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
1.345 brouard 3986: if(FixedV[itv]!=0){ /* Not a fixed covariate */
1.341 brouard 3987: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* cotvar[wav][ncovcol+nqv+iv][i] */
1.340 brouard 3988: }else{ /* fixed covariate */
1.345 brouard 3989: cotvarv=covar[itv][i]; /* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model */
1.340 brouard 3990: }
1.339 brouard 3991: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 3992: cotvarvold=cotvarv;
3993: }else{ /* A second product */
3994: cotvarv=cotvarv*cotvarvold;
1.339 brouard 3995: }
3996: iposold=ipos;
1.340 brouard 3997: cov[ioffset+ipos]=cotvarv;
1.234 brouard 3998: }
1.339 brouard 3999: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
4000: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4001: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4002: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4003: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4004: /* printf(" i=%d,mi=%d,itv=%d,TmodelInvind[itv]=%d,cotvar[mw[mi][i]][TmodelInvind[itv]][i]=%f\n", i, mi, itv, TmodelInvind[itv],cotvar[mw[mi][i]][TmodelInvind[itv]][i]); */
4005: /* } */
4006: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
4007: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4008: /* /\* printf(" i=%d,mi=%d,iqtv=%d,TmodelInvQind[iqtv]=%d,cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]=%f\n", i, mi, iqtv, TmodelInvQind[iqtv],cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]); *\/ */
4009: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
4010: /* } */
4011: /* for products of time varying to be done */
1.234 brouard 4012: for (ii=1;ii<=nlstate+ndeath;ii++)
4013: for (j=1;j<=nlstate+ndeath;j++){
4014: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4015: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4016: }
1.336 brouard 4017:
4018: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4019: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
1.234 brouard 4020: for(d=0; d<dh[mi][i]; d++){
4021: newm=savm;
4022: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4023: cov[2]=agexact;
4024: if(nagesqr==1)
4025: cov[3]= agexact*agexact; /* Should be changed here */
4026: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 4027: if(!FixedV[Tvar[Tage[kk]]])
4028: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4029: else
1.341 brouard 4030: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.234 brouard 4031: }
4032: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4033: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4034: savm=oldm;
4035: oldm=newm;
4036: } /* end mult */
4037:
4038: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4039: /* But now since version 0.9 we anticipate for bias at large stepm.
4040: * If stepm is larger than one month (smallest stepm) and if the exact delay
4041: * (in months) between two waves is not a multiple of stepm, we rounded to
4042: * the nearest (and in case of equal distance, to the lowest) interval but now
4043: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4044: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4045: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 4046: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4047: * -stepm/2 to stepm/2 .
4048: * For stepm=1 the results are the same as for previous versions of Imach.
4049: * For stepm > 1 the results are less biased than in previous versions.
4050: */
1.234 brouard 4051: s1=s[mw[mi][i]][i];
4052: s2=s[mw[mi+1][i]][i];
4053: bbh=(double)bh[mi][i]/(double)stepm;
4054: /* bias bh is positive if real duration
4055: * is higher than the multiple of stepm and negative otherwise.
4056: */
4057: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
4058: if( s2 > nlstate){
4059: /* i.e. if s2 is a death state and if the date of death is known
4060: then the contribution to the likelihood is the probability to
4061: die between last step unit time and current step unit time,
4062: which is also equal to probability to die before dh
4063: minus probability to die before dh-stepm .
4064: In version up to 0.92 likelihood was computed
4065: as if date of death was unknown. Death was treated as any other
4066: health state: the date of the interview describes the actual state
4067: and not the date of a change in health state. The former idea was
4068: to consider that at each interview the state was recorded
4069: (healthy, disable or death) and IMaCh was corrected; but when we
4070: introduced the exact date of death then we should have modified
4071: the contribution of an exact death to the likelihood. This new
4072: contribution is smaller and very dependent of the step unit
4073: stepm. It is no more the probability to die between last interview
4074: and month of death but the probability to survive from last
4075: interview up to one month before death multiplied by the
4076: probability to die within a month. Thanks to Chris
4077: Jackson for correcting this bug. Former versions increased
4078: mortality artificially. The bad side is that we add another loop
4079: which slows down the processing. The difference can be up to 10%
4080: lower mortality.
4081: */
4082: /* If, at the beginning of the maximization mostly, the
4083: cumulative probability or probability to be dead is
4084: constant (ie = 1) over time d, the difference is equal to
4085: 0. out[s1][3] = savm[s1][3]: probability, being at state
4086: s1 at precedent wave, to be dead a month before current
4087: wave is equal to probability, being at state s1 at
4088: precedent wave, to be dead at mont of the current
4089: wave. Then the observed probability (that this person died)
4090: is null according to current estimated parameter. In fact,
4091: it should be very low but not zero otherwise the log go to
4092: infinity.
4093: */
1.183 brouard 4094: /* #ifdef INFINITYORIGINAL */
4095: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4096: /* #else */
4097: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
4098: /* lli=log(mytinydouble); */
4099: /* else */
4100: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4101: /* #endif */
1.226 brouard 4102: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4103:
1.226 brouard 4104: } else if ( s2==-1 ) { /* alive */
4105: for (j=1,survp=0. ; j<=nlstate; j++)
4106: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4107: /*survp += out[s1][j]; */
4108: lli= log(survp);
4109: }
1.336 brouard 4110: /* else if (s2==-4) { */
4111: /* for (j=3,survp=0. ; j<=nlstate; j++) */
4112: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4113: /* lli= log(survp); */
4114: /* } */
4115: /* else if (s2==-5) { */
4116: /* for (j=1,survp=0. ; j<=2; j++) */
4117: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4118: /* lli= log(survp); */
4119: /* } */
1.226 brouard 4120: else{
4121: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4122: /* lli= (savm[s1][s2]>(double)1.e-8 ?log((1.+bbh)*out[s1][s2]- bbh*(savm[s1][s2])):log((1.+bbh)*out[s1][s2]));*/ /* linear interpolation */
4123: }
4124: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
4125: /*if(lli ==000.0)*/
1.340 brouard 4126: /* printf("num[i], i=%d, bbh= %f lli=%f savm=%f out=%f %d\n",bbh,lli,savm[s1][s2], out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]],i); */
1.226 brouard 4127: ipmx +=1;
4128: sw += weight[i];
4129: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4130: /* if (lli < log(mytinydouble)){ */
4131: /* printf("Close to inf lli = %.10lf < %.10lf i= %d mi= %d, s[%d][i]=%d s1=%d s2=%d\n", lli,log(mytinydouble), i, mi,mw[mi][i], s[mw[mi][i]][i], s1,s2); */
4132: /* fprintf(ficlog,"Close to inf lli = %.10lf i= %d mi= %d, s[mw[mi][i]][i]=%d\n", lli, i, mi,s[mw[mi][i]][i]); */
4133: /* } */
4134: } /* end of wave */
4135: } /* end of individual */
4136: } else if(mle==2){
4137: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 4138: ioffset=2+nagesqr ;
4139: for (k=1; k<=ncovf;k++)
4140: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 4141: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 4142: for(k=1; k <= ncovv ; k++){
1.341 brouard 4143: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.319 brouard 4144: }
1.226 brouard 4145: for (ii=1;ii<=nlstate+ndeath;ii++)
4146: for (j=1;j<=nlstate+ndeath;j++){
4147: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4148: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4149: }
4150: for(d=0; d<=dh[mi][i]; d++){
4151: newm=savm;
4152: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4153: cov[2]=agexact;
4154: if(nagesqr==1)
4155: cov[3]= agexact*agexact;
4156: for (kk=1; kk<=cptcovage;kk++) {
4157: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4158: }
4159: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4160: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4161: savm=oldm;
4162: oldm=newm;
4163: } /* end mult */
4164:
4165: s1=s[mw[mi][i]][i];
4166: s2=s[mw[mi+1][i]][i];
4167: bbh=(double)bh[mi][i]/(double)stepm;
4168: lli= (savm[s1][s2]>(double)1.e-8 ?log((1.+bbh)*out[s1][s2]- bbh*(savm[s1][s2])):log((1.+bbh)*out[s1][s2])); /* linear interpolation */
4169: ipmx +=1;
4170: sw += weight[i];
4171: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4172: } /* end of wave */
4173: } /* end of individual */
4174: } else if(mle==3){ /* exponential inter-extrapolation */
4175: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4176: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4177: for(mi=1; mi<= wav[i]-1; mi++){
4178: for (ii=1;ii<=nlstate+ndeath;ii++)
4179: for (j=1;j<=nlstate+ndeath;j++){
4180: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4181: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4182: }
4183: for(d=0; d<dh[mi][i]; d++){
4184: newm=savm;
4185: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4186: cov[2]=agexact;
4187: if(nagesqr==1)
4188: cov[3]= agexact*agexact;
4189: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4190: if(!FixedV[Tvar[Tage[kk]]])
4191: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4192: else
1.341 brouard 4193: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.226 brouard 4194: }
4195: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4196: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4197: savm=oldm;
4198: oldm=newm;
4199: } /* end mult */
4200:
4201: s1=s[mw[mi][i]][i];
4202: s2=s[mw[mi+1][i]][i];
4203: bbh=(double)bh[mi][i]/(double)stepm;
4204: lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2])); /* exponential inter-extrapolation */
4205: ipmx +=1;
4206: sw += weight[i];
4207: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4208: } /* end of wave */
4209: } /* end of individual */
4210: }else if (mle==4){ /* ml=4 no inter-extrapolation */
4211: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4212: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4213: for(mi=1; mi<= wav[i]-1; mi++){
4214: for (ii=1;ii<=nlstate+ndeath;ii++)
4215: for (j=1;j<=nlstate+ndeath;j++){
4216: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4217: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4218: }
4219: for(d=0; d<dh[mi][i]; d++){
4220: newm=savm;
4221: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4222: cov[2]=agexact;
4223: if(nagesqr==1)
4224: cov[3]= agexact*agexact;
4225: for (kk=1; kk<=cptcovage;kk++) {
4226: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4227: }
1.126 brouard 4228:
1.226 brouard 4229: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4230: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4231: savm=oldm;
4232: oldm=newm;
4233: } /* end mult */
4234:
4235: s1=s[mw[mi][i]][i];
4236: s2=s[mw[mi+1][i]][i];
4237: if( s2 > nlstate){
4238: lli=log(out[s1][s2] - savm[s1][s2]);
4239: } else if ( s2==-1 ) { /* alive */
4240: for (j=1,survp=0. ; j<=nlstate; j++)
4241: survp += out[s1][j];
4242: lli= log(survp);
4243: }else{
4244: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4245: }
4246: ipmx +=1;
4247: sw += weight[i];
4248: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.343 brouard 4249: /* printf("num[i]=%09ld, i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",num[i],i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.226 brouard 4250: } /* end of wave */
4251: } /* end of individual */
4252: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4253: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4254: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4255: for(mi=1; mi<= wav[i]-1; mi++){
4256: for (ii=1;ii<=nlstate+ndeath;ii++)
4257: for (j=1;j<=nlstate+ndeath;j++){
4258: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4259: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4260: }
4261: for(d=0; d<dh[mi][i]; d++){
4262: newm=savm;
4263: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4264: cov[2]=agexact;
4265: if(nagesqr==1)
4266: cov[3]= agexact*agexact;
4267: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4268: if(!FixedV[Tvar[Tage[kk]]])
4269: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4270: else
1.341 brouard 4271: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.226 brouard 4272: }
1.126 brouard 4273:
1.226 brouard 4274: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4275: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4276: savm=oldm;
4277: oldm=newm;
4278: } /* end mult */
4279:
4280: s1=s[mw[mi][i]][i];
4281: s2=s[mw[mi+1][i]][i];
4282: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4283: ipmx +=1;
4284: sw += weight[i];
4285: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4286: /*printf("i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],out[s1][s2],savm[s1][s2]);*/
4287: } /* end of wave */
4288: } /* end of individual */
4289: } /* End of if */
4290: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4291: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4292: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4293: return -l;
1.126 brouard 4294: }
4295:
4296: /*************** log-likelihood *************/
4297: double funcone( double *x)
4298: {
1.228 brouard 4299: /* Same as func but slower because of a lot of printf and if */
1.335 brouard 4300: int i, ii, j, k, mi, d, kk, kf=0;
1.228 brouard 4301: int ioffset=0;
1.339 brouard 4302: int ipos=0,iposold=0,ncovv=0;
4303:
1.340 brouard 4304: double cotvarv, cotvarvold;
1.131 brouard 4305: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4306: double **out;
4307: double lli; /* Individual log likelihood */
4308: double llt;
4309: int s1, s2;
1.228 brouard 4310: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4311:
1.126 brouard 4312: double bbh, survp;
1.187 brouard 4313: double agexact;
1.214 brouard 4314: double agebegin, ageend;
1.126 brouard 4315: /*extern weight */
4316: /* We are differentiating ll according to initial status */
4317: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4318: /*for(i=1;i<imx;i++)
4319: printf(" %d\n",s[4][i]);
4320: */
4321: cov[1]=1.;
4322:
4323: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4324: ioffset=0;
4325: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.336 brouard 4326: /* Computes the values of the ncovmodel covariates of the model
4327: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
4328: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
4329: to be observed in j being in i according to the model.
4330: */
1.243 brouard 4331: /* ioffset=2+nagesqr+cptcovage; */
4332: ioffset=2+nagesqr;
1.232 brouard 4333: /* Fixed */
1.224 brouard 4334: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4335: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.335 brouard 4336: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
1.339 brouard 4337: /* printf("Debug3 TvarFind[%d]=%d",kf, TvarFind[kf]); */
4338: /* printf(" Tvar[TvarFind[kf]]=%d", Tvar[TvarFind[kf]]); */
4339: /* printf(" i=%d covar[Tvar[TvarFind[kf]]][i]=%f\n",i,covar[Tvar[TvarFind[kf]]][i]); */
1.335 brouard 4340: cov[ioffset+TvarFind[kf]]=covar[Tvar[TvarFind[kf]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (k=6)*/
1.232 brouard 4341: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4342: /* cov[2+6]=covar[Tvar[6]][i]; */
4343: /* cov[2+6]=covar[2][i]; V2 */
4344: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4345: /* cov[2+7]=covar[Tvar[7]][i]; */
4346: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4347: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4348: /* cov[2+9]=covar[Tvar[9]][i]; */
4349: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4350: }
1.336 brouard 4351: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
4352: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
4353: has been calculated etc */
4354: /* For an individual i, wav[i] gives the number of effective waves */
4355: /* We compute the contribution to Likelihood of each effective transition
4356: mw[mi][i] is real wave of the mi th effectve wave */
4357: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
4358: s2=s[mw[mi+1][i]][i];
1.341 brouard 4359: And the iv th varying covariate in the DATA is the cotvar[mw[mi+1][i]][ncovcol+nqv+iv][i]
1.336 brouard 4360: */
4361: /* This part may be useless now because everythin should be in covar */
1.232 brouard 4362: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4363: /* cov[++ioffset]=coqvar[TvarFQ[k]][i];/\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V2 and V1*V2 is fixed (k=6 and 7?)*\/ */
4364: /* } */
1.231 brouard 4365: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4366: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4367: /* } */
1.225 brouard 4368:
1.233 brouard 4369:
4370: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.339 brouard 4371: /* Wave varying (but not age varying) *//* V1+V3+age*V1+age*V3+V1*V3 with V4 tv and V5 tvq k= 1 to 5 and extra at V(5+1)=6 for V1*V3 */
4372: /* for(k=1; k <= ncovv ; k++){ /\* Varying covariates (single and product but no age )*\/ */
4373: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; *\/ */
4374: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
4375: /* } */
4376:
4377: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
4378: /* model V1+V3+age*V1+age*V3+V1*V3 */
4379: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
4380: /* TvarVV[1]=V3 (first time varying in the model equation, TvarVV[2]=V1 (in V1*V3) TvarVV[3]=3(V3) */
4381: /* We need the position of the time varying or product in the model */
4382: /* TvarVVind={2,5,5}, for V3 at position 2 and then the product V1*V3 is decomposed into V1 and V3 but at same position 5 */
4383: /* TvarVV gives the variable name */
1.340 brouard 4384: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
4385: * k= 1 2 3 4 5 6 7 8 9
4386: * varying 1 2 3 4 5
4387: * ncovv 1 2 3 4 5 6 7 8
1.343 brouard 4388: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
1.340 brouard 4389: * TvarVVind 2 3 7 7 8 8 9 9
4390: * TvarFind[k] 1 0 0 0 0 0 0 0 0
4391: */
1.345 brouard 4392: /* Other model ncovcol=5 nqv=0 ntv=3 nqtv=0 nlstate=3
1.346 ! brouard 4393: * V2 V3 V4 are fixed V6 V7 are timevarying so V8 and V5 are not in the model and product column will start at 9 Tvar[4]=6
1.345 brouard 4394: * FixedV[ncovcol+qv+ntv+nqtv] V5
4395: * V1 V2 V3 V4 V5 V6 V7 V8
4396: * 0 0 0 0 0 1 1 1
4397: * model= V2 + V3 + V4 + V6 + V7 + V6*V2 + V7*V2 + V6*V3 + V7*V3 + V6*V4 + V7*V4
4398: * kmodel 1 2 3 4 5 6 7 8 9 10 11
4399: * ncovf 1 2 3
4400: * ncovvt=14 1 2 3 4 5 6 7 8 9 10 11 12 13 14
4401: * TvarVV[1]@14 = itv {6, 7, 6, 2, 7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
4402: * TvarVVind[1]@14= {4, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11}
4403: * TvarFind[1]@14= {1, 2, 3, 0 <repeats 12 times>}
4404: * Tvar[1]@20= {2, 3, 4, 6, 7, 9, 10, 11, 12, 13, 14}
4405: * TvarFind[itv] 0 0 0
4406: * FixedV[itv] 1 1 1 0 1 0 1 0 1 0 0
4407: * Tvar[TvarFind[ncovf]]=[1]=2 [2]=3 [4]=4
4408: * Tvar[TvarFind[itv]] [0]=? ?ncovv 1 à ncovvt]
4409: * Not a fixed cotvar[mw][itv][i] 6 7 6 2 7, 2, 6, 3, 7, 3, 6, 4, 7, 4}
4410: * fixed covar[itv] [6] [7] [6][2]
4411: */
4412:
1.340 brouard 4413: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
1.345 brouard 4414: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate, exploding product */
1.340 brouard 4415: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
1.345 brouard 4416: /* if(TvarFind[itv]==0){ /\* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv *\/ */
4417: if(FixedV[itv]!=0){ /* Not a fixed covariate? Could be a fixed covariate of a product with a higher than ncovcol+nqv, itv */
4418: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* because cotvar starts now at first ncovcol+nqv+ntv+nqtv (1 to nqtv) */
1.340 brouard 4419: }else{ /* fixed covariate */
1.345 brouard 4420: /* cotvarv=covar[Tvar[TvarFind[itv]]][i]; /\* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model *\/ */
4421: cotvarv=covar[itv][i]; /* Error: TvarFind gives the name, that is the true column of fixed covariates, but Tvar of the model */
1.340 brouard 4422: }
1.339 brouard 4423: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 4424: cotvarvold=cotvarv;
4425: }else{ /* A second product */
4426: cotvarv=cotvarv*cotvarvold;
1.339 brouard 4427: }
4428: iposold=ipos;
1.340 brouard 4429: cov[ioffset+ipos]=cotvarv;
1.339 brouard 4430: /* For products */
4431: }
4432: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates single *\/ */
4433: /* iv=TvarVDind[itv]; /\* iv, position in the model equation of time varying covariate itv *\/ */
4434: /* /\* "V1+V3+age*V1+age*V3+V1*V3" with V3 time varying *\/ */
4435: /* /\* 1 2 3 4 5 *\/ */
4436: /* /\*itv 1 *\/ */
4437: /* /\* TvarVInd[1]= 2 *\/ */
4438: /* /\* iv= Tvar[Tmodelind[itv]]-ncovcol-nqv; /\\* Counting the # varying covariate from 1 to ntveff *\\/ *\/ */
4439: /* /\* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; *\/ */
4440: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; *\/ */
4441: /* /\* k=ioffset-2-nagesqr-cptcovage+itv; /\\* position in simple model *\\/ *\/ */
4442: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; *\/ */
4443: /* cov[ioffset+iv]=cotvar[mw[mi][i]][itv][i]; */
4444: /* /\* printf(" i=%d,mi=%d,itv=%d,TmodelInvind[itv]=%d,cotvar[mw[mi][i]][itv][i]=%f\n", i, mi, itv, TvarVDind[itv],cotvar[mw[mi][i]][itv][i]); *\/ */
4445: /* } */
1.232 brouard 4446: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4447: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4448: /* /\* printf(" i=%d,mi=%d,iqtv=%d,TmodelInvQind[iqtv]=%d,cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]=%f\n", i, mi, iqtv, TmodelInvQind[iqtv],cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]); *\/ */
4449: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4450: /* } */
1.126 brouard 4451: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4452: for (j=1;j<=nlstate+ndeath;j++){
4453: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4454: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4455: }
1.214 brouard 4456:
4457: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4458: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4459: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4460: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4461: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4462: and mw[mi+1][i]. dh depends on stepm.*/
4463: newm=savm;
1.247 brouard 4464: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4465: cov[2]=agexact;
4466: if(nagesqr==1)
4467: cov[3]= agexact*agexact;
4468: for (kk=1; kk<=cptcovage;kk++) {
4469: if(!FixedV[Tvar[Tage[kk]]])
4470: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4471: else
1.341 brouard 4472: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.242 brouard 4473: }
4474: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4475: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4476: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4477: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4478: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4479: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4480: savm=oldm;
4481: oldm=newm;
1.126 brouard 4482: } /* end mult */
1.336 brouard 4483: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4484: /* But now since version 0.9 we anticipate for bias at large stepm.
4485: * If stepm is larger than one month (smallest stepm) and if the exact delay
4486: * (in months) between two waves is not a multiple of stepm, we rounded to
4487: * the nearest (and in case of equal distance, to the lowest) interval but now
4488: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4489: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4490: * probability in order to take into account the bias as a fraction of the way
4491: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4492: * -stepm/2 to stepm/2 .
4493: * For stepm=1 the results are the same as for previous versions of Imach.
4494: * For stepm > 1 the results are less biased than in previous versions.
4495: */
1.126 brouard 4496: s1=s[mw[mi][i]][i];
4497: s2=s[mw[mi+1][i]][i];
1.217 brouard 4498: /* if(s2==-1){ */
1.268 brouard 4499: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4500: /* /\* exit(1); *\/ */
4501: /* } */
1.126 brouard 4502: bbh=(double)bh[mi][i]/(double)stepm;
4503: /* bias is positive if real duration
4504: * is higher than the multiple of stepm and negative otherwise.
4505: */
4506: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4507: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4508: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4509: for (j=1,survp=0. ; j<=nlstate; j++)
4510: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4511: lli= log(survp);
1.126 brouard 4512: }else if (mle==1){
1.242 brouard 4513: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4514: } else if(mle==2){
1.242 brouard 4515: lli= (savm[s1][s2]>(double)1.e-8 ?log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]):log((1.+bbh)*out[s1][s2])); /* linear interpolation */
1.126 brouard 4516: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4517: lli= (savm[s1][s2]>(double)1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2])); /* exponential inter-extrapolation */
1.126 brouard 4518: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4519: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4520: } else{ /* mle=0 back to 1 */
1.242 brouard 4521: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4522: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4523: } /* End of if */
4524: ipmx +=1;
4525: sw += weight[i];
4526: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.342 brouard 4527: /* Printing covariates values for each contribution for checking */
1.343 brouard 4528: /* printf("num[i]=%09ld, i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",num[i],i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.126 brouard 4529: if(globpr){
1.246 brouard 4530: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4531: %11.6f %11.6f %11.6f ", \
1.242 brouard 4532: num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw,
1.268 brouard 4533: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.343 brouard 4534: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
4535: /* %11.6f %11.6f %11.6f ", \ */
4536: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
4537: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 4538: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4539: llt +=ll[k]*gipmx/gsw;
4540: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 brouard 4541: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 4542: }
1.343 brouard 4543: fprintf(ficresilk," %10.6f ", -llt);
1.335 brouard 4544: /* printf(" %10.6f\n", -llt); */
1.342 brouard 4545: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
1.343 brouard 4546: /* fprintf(ficresilk,"%09ld ", num[i]); */ /* not necessary */
4547: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
4548: fprintf(ficresilk," %g",covar[Tvar[TvarFind[kf]]][i]);
4549: }
4550: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
4551: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
4552: if(ipos!=iposold){ /* Not a product or first of a product */
4553: fprintf(ficresilk," %g",cov[ioffset+ipos]);
4554: /* printf(" %g",cov[ioffset+ipos]); */
4555: }else{
4556: fprintf(ficresilk,"*");
4557: /* printf("*"); */
1.342 brouard 4558: }
1.343 brouard 4559: iposold=ipos;
4560: }
4561: for (kk=1; kk<=cptcovage;kk++) {
4562: if(!FixedV[Tvar[Tage[kk]]]){
4563: fprintf(ficresilk," %g*age",covar[Tvar[Tage[kk]]][i]);
4564: /* printf(" %g*age",covar[Tvar[Tage[kk]]][i]); */
4565: }else{
4566: fprintf(ficresilk," %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
4567: /* printf(" %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
1.342 brouard 4568: }
1.343 brouard 4569: }
4570: /* printf("\n"); */
1.342 brouard 4571: /* } /\* End debugILK *\/ */
4572: fprintf(ficresilk,"\n");
4573: } /* End if globpr */
1.335 brouard 4574: } /* end of wave */
4575: } /* end of individual */
4576: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 4577: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 brouard 4578: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4579: if(globpr==0){ /* First time we count the contributions and weights */
4580: gipmx=ipmx;
4581: gsw=sw;
4582: }
1.343 brouard 4583: return -l;
1.126 brouard 4584: }
4585:
4586:
4587: /*************** function likelione ***********/
1.292 brouard 4588: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4589: {
4590: /* This routine should help understanding what is done with
4591: the selection of individuals/waves and
4592: to check the exact contribution to the likelihood.
4593: Plotting could be done.
1.342 brouard 4594: */
4595: void pstamp(FILE *ficres);
1.343 brouard 4596: int k, kf, kk, kvar, ncovv, iposold, ipos;
1.126 brouard 4597:
4598: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4599: strcpy(fileresilk,"ILK_");
1.202 brouard 4600: strcat(fileresilk,fileresu);
1.126 brouard 4601: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4602: printf("Problem with resultfile: %s\n", fileresilk);
4603: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4604: }
1.342 brouard 4605: pstamp(ficresilk);fprintf(ficresilk,"# model=1+age+%s\n",model);
1.214 brouard 4606: fprintf(ficresilk, "#individual(line's_record) count ageb ageend s1 s2 wave# effective_wave# number_of_matrices_product pij weight weight/gpw -2ln(pij)*weight 0pij_x 0pij_(x-stepm) cumulating_loglikeli_by_health_state(reweighted=-2ll*weightXnumber_of_contribs/sum_of_weights) and_total\n");
4607: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4608: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4609: for(k=1; k<=nlstate; k++)
4610: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
1.342 brouard 4611: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total) ");
4612:
4613: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
4614: for(kf=1;kf <= ncovf; kf++){
4615: fprintf(ficresilk,"V%d",Tvar[TvarFind[kf]]);
4616: /* printf("V%d",Tvar[TvarFind[kf]]); */
4617: }
4618: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){
1.343 brouard 4619: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
1.342 brouard 4620: if(ipos!=iposold){ /* Not a product or first of a product */
4621: /* printf(" %d",ipos); */
4622: fprintf(ficresilk," V%d",TvarVV[ncovv]);
4623: }else{
4624: /* printf("*"); */
4625: fprintf(ficresilk,"*");
1.343 brouard 4626: }
1.342 brouard 4627: iposold=ipos;
4628: }
4629: for (kk=1; kk<=cptcovage;kk++) {
4630: if(!FixedV[Tvar[Tage[kk]]]){
4631: /* printf(" %d*age(Fixed)",Tvar[Tage[kk]]); */
4632: fprintf(ficresilk," %d*age(Fixed)",Tvar[Tage[kk]]);
4633: }else{
4634: fprintf(ficresilk," %d*age(Varying)",Tvar[Tage[kk]]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
4635: /* printf(" %d*age(Varying)",Tvar[Tage[kk]]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
4636: }
4637: }
4638: /* } /\* End if debugILK *\/ */
4639: /* printf("\n"); */
4640: fprintf(ficresilk,"\n");
4641: } /* End glogpri */
1.126 brouard 4642:
1.292 brouard 4643: *fretone=(*func)(p);
1.126 brouard 4644: if(*globpri !=0){
4645: fclose(ficresilk);
1.205 brouard 4646: if (mle ==0)
4647: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4648: else if(mle >=1)
4649: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4650: fprintf(fichtm," You should at least run with mle >= 1 to get starting values corresponding to the optimized parameters in order to visualize the real contribution of each individual/wave: <a href=\"%s\">%s</a><br>\n",subdirf(fileresilk),subdirf(fileresilk));
1.274 brouard 4651: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4652:
1.207 brouard 4653: fprintf(fichtm,"<br>- The function drawn is -2Log(L) in Log scale: by state of origin <a href=\"%s-ori.png\">%s-ori.png</a><br> \
1.343 brouard 4654: <img src=\"%s-ori.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4655: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.343 brouard 4656: <img src=\"%s-dest.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
4657:
4658: for (k=1; k<= nlstate ; k++) {
4659: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j. Dot's sizes are related to corresponding weight: <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br>\n \
4660: <img src=\"%s-p%dj.png\">\n",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4661: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
4662: /* kvar=Tvar[TvarFind[kf]]; */ /* variable */
4663: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j with colored covariate V%d. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \
4664: <img src=\"%s-p%dj-%d.png\">",k,k,Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,Tvar[TvarFind[kf]]);
4665: }
4666: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Loop on the time varying extended covariates (with extension of Vn*Vm */
4667: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
4668: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
4669: /* printf("DebugILK fichtm ncovv=%d, kvar=TvarVV[ncovv]=V%d, ipos=TvarVVind[ncovv]=%d, Dummy[ipos]=%d, Typevar[ipos]=%d\n", ncovv,kvar,ipos,Dummy[ipos],Typevar[ipos]); */
4670: if(ipos!=iposold){ /* Not a product or first of a product */
4671: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
4672: /* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); */
4673: if(Dummy[ipos]==0 && Typevar[ipos]==0){ /* Only if dummy time varying: Dummy(0, 1=quant singor prod without age,2 dummy*age, 3quant*age) Typevar (0 single, 1=*age,2=Vn*vm) */
4674: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j with colored time varying dummy covariate V%d. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \
4675: <img src=\"%s-p%dj-%d.png\">",k,k,kvar,kvar,kvar,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,kvar);
4676: } /* End only for dummies time varying (single?) */
4677: }else{ /* Useless product */
4678: /* printf("*"); */
4679: /* fprintf(ficresilk,"*"); */
4680: }
4681: iposold=ipos;
4682: } /* For each time varying covariate */
4683: } /* End loop on states */
4684:
4685: /* if(debugILK){ */
4686: /* for(kf=1; kf <= ncovf; kf++){ /\* For each simple dummy covariate of the model *\/ */
4687: /* /\* kvar=Tvar[TvarFind[kf]]; *\/ /\* variable *\/ */
4688: /* for (k=1; k<= nlstate ; k++) { */
4689: /* fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j with colored covariate V%. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \ */
4690: /* <img src=\"%s-p%dj-%d.png\">",k,k,Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,Tvar[TvarFind[kf]]); */
4691: /* } */
4692: /* } */
4693: /* for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /\* Loop on the time varying extended covariates (with extension of Vn*Vm *\/ */
4694: /* ipos=TvarVVind[ncovv]; /\* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate *\/ */
4695: /* kvar=TvarVV[ncovv]; /\* TvarVV={3, 1, 3} gives the name of each varying covariate *\/ */
4696: /* /\* printf("DebugILK fichtm ncovv=%d, kvar=TvarVV[ncovv]=V%d, ipos=TvarVVind[ncovv]=%d, Dummy[ipos]=%d, Typevar[ipos]=%d\n", ncovv,kvar,ipos,Dummy[ipos],Typevar[ipos]); *\/ */
4697: /* if(ipos!=iposold){ /\* Not a product or first of a product *\/ */
4698: /* /\* fprintf(ficresilk," V%d",TvarVV[ncovv]); *\/ */
4699: /* /\* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); *\/ */
4700: /* if(Dummy[ipos]==0 && Typevar[ipos]==0){ /\* Only if dummy time varying: Dummy(0, 1=quant singor prod without age,2 dummy*age, 3quant*age) Typevar (0 single, 1=*age,2=Vn*vm) *\/ */
4701: /* for (k=1; k<= nlstate ; k++) { */
4702: /* fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \ */
4703: /* <img src=\"%s-p%dj-%d.png\">",k,k,kvar,kvar,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,kvar); */
4704: /* } /\* End state *\/ */
4705: /* } /\* End only for dummies time varying (single?) *\/ */
4706: /* }else{ /\* Useless product *\/ */
4707: /* /\* printf("*"); *\/ */
4708: /* /\* fprintf(ficresilk,"*"); *\/ */
4709: /* } */
4710: /* iposold=ipos; */
4711: /* } /\* For each time varying covariate *\/ */
4712: /* }/\* End debugILK *\/ */
1.207 brouard 4713: fflush(fichtm);
1.343 brouard 4714: }/* End globpri */
1.126 brouard 4715: return;
4716: }
4717:
4718:
4719: /*********** Maximum Likelihood Estimation ***************/
4720:
4721: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4722: {
1.319 brouard 4723: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4724: double **xi;
4725: double fret;
4726: double fretone; /* Only one call to likelihood */
4727: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4728:
4729: #ifdef NLOPT
4730: int creturn;
4731: nlopt_opt opt;
4732: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4733: double *lb;
4734: double minf; /* the minimum objective value, upon return */
4735: double * p1; /* Shifted parameters from 0 instead of 1 */
4736: myfunc_data dinst, *d = &dinst;
4737: #endif
4738:
4739:
1.126 brouard 4740: xi=matrix(1,npar,1,npar);
4741: for (i=1;i<=npar;i++)
4742: for (j=1;j<=npar;j++)
4743: xi[i][j]=(i==j ? 1.0 : 0.0);
4744: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4745: strcpy(filerespow,"POW_");
1.126 brouard 4746: strcat(filerespow,fileres);
4747: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4748: printf("Problem with resultfile: %s\n", filerespow);
4749: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4750: }
4751: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4752: for (i=1;i<=nlstate;i++)
4753: for(j=1;j<=nlstate+ndeath;j++)
4754: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4755: fprintf(ficrespow,"\n");
1.162 brouard 4756: #ifdef POWELL
1.319 brouard 4757: #ifdef LINMINORIGINAL
4758: #else /* LINMINORIGINAL */
4759:
4760: flatdir=ivector(1,npar);
4761: for (j=1;j<=npar;j++) flatdir[j]=0;
4762: #endif /*LINMINORIGINAL */
4763:
4764: #ifdef FLATSUP
4765: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4766: /* reorganizing p by suppressing flat directions */
4767: for(i=1, jk=1; i <=nlstate; i++){
4768: for(k=1; k <=(nlstate+ndeath); k++){
4769: if (k != i) {
4770: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4771: if(flatdir[jk]==1){
4772: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4773: }
4774: for(j=1; j <=ncovmodel; j++){
4775: printf("%12.7f ",p[jk]);
4776: jk++;
4777: }
4778: printf("\n");
4779: }
4780: }
4781: }
4782: /* skipping */
4783: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4784: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4785: for(k=1; k <=(nlstate+ndeath); k++){
4786: if (k != i) {
4787: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4788: if(flatdir[jk]==1){
4789: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4790: for(j=1; j <=ncovmodel; jk++,j++){
4791: printf(" p[%d]=%12.7f",jk, p[jk]);
4792: /*q[jjk]=p[jk];*/
4793: }
4794: }else{
4795: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4796: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4797: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4798: /*q[jjk]=p[jk];*/
4799: }
4800: }
4801: printf("\n");
4802: }
4803: fflush(stdout);
4804: }
4805: }
4806: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4807: #else /* FLATSUP */
1.126 brouard 4808: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4809: #endif /* FLATSUP */
4810:
4811: #ifdef LINMINORIGINAL
4812: #else
4813: free_ivector(flatdir,1,npar);
4814: #endif /* LINMINORIGINAL*/
4815: #endif /* POWELL */
1.126 brouard 4816:
1.162 brouard 4817: #ifdef NLOPT
4818: #ifdef NEWUOA
4819: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4820: #else
4821: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4822: #endif
4823: lb=vector(0,npar-1);
4824: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4825: nlopt_set_lower_bounds(opt, lb);
4826: nlopt_set_initial_step1(opt, 0.1);
4827:
4828: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4829: d->function = func;
4830: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4831: nlopt_set_min_objective(opt, myfunc, d);
4832: nlopt_set_xtol_rel(opt, ftol);
4833: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4834: printf("nlopt failed! %d\n",creturn);
4835: }
4836: else {
4837: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4838: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4839: iter=1; /* not equal */
4840: }
4841: nlopt_destroy(opt);
4842: #endif
1.319 brouard 4843: #ifdef FLATSUP
4844: /* npared = npar -flatd/ncovmodel; */
4845: /* xired= matrix(1,npared,1,npared); */
4846: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4847: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4848: /* free_matrix(xire,1,npared,1,npared); */
4849: #else /* FLATSUP */
4850: #endif /* FLATSUP */
1.126 brouard 4851: free_matrix(xi,1,npar,1,npar);
4852: fclose(ficrespow);
1.203 brouard 4853: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4854: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4855: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4856:
4857: }
4858:
4859: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4860: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4861: {
4862: double **a,**y,*x,pd;
1.203 brouard 4863: /* double **hess; */
1.164 brouard 4864: int i, j;
1.126 brouard 4865: int *indx;
4866:
4867: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4868: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4869: void lubksb(double **a, int npar, int *indx, double b[]) ;
4870: void ludcmp(double **a, int npar, int *indx, double *d) ;
4871: double gompertz(double p[]);
1.203 brouard 4872: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4873:
4874: printf("\nCalculation of the hessian matrix. Wait...\n");
4875: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4876: for (i=1;i<=npar;i++){
1.203 brouard 4877: printf("%d-",i);fflush(stdout);
4878: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4879:
4880: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4881:
4882: /* printf(" %f ",p[i]);
4883: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4884: }
4885:
4886: for (i=1;i<=npar;i++) {
4887: for (j=1;j<=npar;j++) {
4888: if (j>i) {
1.203 brouard 4889: printf(".%d-%d",i,j);fflush(stdout);
4890: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4891: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4892:
4893: hess[j][i]=hess[i][j];
4894: /*printf(" %lf ",hess[i][j]);*/
4895: }
4896: }
4897: }
4898: printf("\n");
4899: fprintf(ficlog,"\n");
4900:
4901: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4902: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4903:
4904: a=matrix(1,npar,1,npar);
4905: y=matrix(1,npar,1,npar);
4906: x=vector(1,npar);
4907: indx=ivector(1,npar);
4908: for (i=1;i<=npar;i++)
4909: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4910: ludcmp(a,npar,indx,&pd);
4911:
4912: for (j=1;j<=npar;j++) {
4913: for (i=1;i<=npar;i++) x[i]=0;
4914: x[j]=1;
4915: lubksb(a,npar,indx,x);
4916: for (i=1;i<=npar;i++){
4917: matcov[i][j]=x[i];
4918: }
4919: }
4920:
4921: printf("\n#Hessian matrix#\n");
4922: fprintf(ficlog,"\n#Hessian matrix#\n");
4923: for (i=1;i<=npar;i++) {
4924: for (j=1;j<=npar;j++) {
1.203 brouard 4925: printf("%.6e ",hess[i][j]);
4926: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4927: }
4928: printf("\n");
4929: fprintf(ficlog,"\n");
4930: }
4931:
1.203 brouard 4932: /* printf("\n#Covariance matrix#\n"); */
4933: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4934: /* for (i=1;i<=npar;i++) { */
4935: /* for (j=1;j<=npar;j++) { */
4936: /* printf("%.6e ",matcov[i][j]); */
4937: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4938: /* } */
4939: /* printf("\n"); */
4940: /* fprintf(ficlog,"\n"); */
4941: /* } */
4942:
1.126 brouard 4943: /* Recompute Inverse */
1.203 brouard 4944: /* for (i=1;i<=npar;i++) */
4945: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4946: /* ludcmp(a,npar,indx,&pd); */
4947:
4948: /* printf("\n#Hessian matrix recomputed#\n"); */
4949:
4950: /* for (j=1;j<=npar;j++) { */
4951: /* for (i=1;i<=npar;i++) x[i]=0; */
4952: /* x[j]=1; */
4953: /* lubksb(a,npar,indx,x); */
4954: /* for (i=1;i<=npar;i++){ */
4955: /* y[i][j]=x[i]; */
4956: /* printf("%.3e ",y[i][j]); */
4957: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4958: /* } */
4959: /* printf("\n"); */
4960: /* fprintf(ficlog,"\n"); */
4961: /* } */
4962:
4963: /* Verifying the inverse matrix */
4964: #ifdef DEBUGHESS
4965: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4966:
1.203 brouard 4967: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4968: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4969:
4970: for (j=1;j<=npar;j++) {
4971: for (i=1;i<=npar;i++){
1.203 brouard 4972: printf("%.2f ",y[i][j]);
4973: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4974: }
4975: printf("\n");
4976: fprintf(ficlog,"\n");
4977: }
1.203 brouard 4978: #endif
1.126 brouard 4979:
4980: free_matrix(a,1,npar,1,npar);
4981: free_matrix(y,1,npar,1,npar);
4982: free_vector(x,1,npar);
4983: free_ivector(indx,1,npar);
1.203 brouard 4984: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4985:
4986:
4987: }
4988:
4989: /*************** hessian matrix ****************/
4990: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4991: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4992: int i;
4993: int l=1, lmax=20;
1.203 brouard 4994: double k1,k2, res, fx;
1.132 brouard 4995: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4996: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4997: int k=0,kmax=10;
4998: double l1;
4999:
5000: fx=func(x);
5001: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 5002: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 5003: l1=pow(10,l);
5004: delts=delt;
5005: for(k=1 ; k <kmax; k=k+1){
5006: delt = delta*(l1*k);
5007: p2[theta]=x[theta] +delt;
1.145 brouard 5008: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 5009: p2[theta]=x[theta]-delt;
5010: k2=func(p2)-fx;
5011: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 5012: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 5013:
1.203 brouard 5014: #ifdef DEBUGHESSII
1.126 brouard 5015: printf("%d %d k1=%.12e k2=%.12e xk1=%.12e xk2=%.12e delt=%.12e res=%.12e l=%d k=%d,fx=%.12e\n",theta,theta,k1,k2,x[theta]+delt,x[theta]-delt,delt,res, l, k,fx);
5016: fprintf(ficlog,"%d %d k1=%.12e k2=%.12e xk1=%.12e xk2=%.12e delt=%.12e res=%.12e l=%d k=%d,fx=%.12e\n",theta,theta,k1,k2,x[theta]+delt,x[theta]-delt,delt,res, l, k,fx);
5017: #endif
5018: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
5019: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
5020: k=kmax;
5021: }
5022: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 5023: k=kmax; l=lmax*10;
1.126 brouard 5024: }
5025: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
5026: delts=delt;
5027: }
1.203 brouard 5028: } /* End loop k */
1.126 brouard 5029: }
5030: delti[theta]=delts;
5031: return res;
5032:
5033: }
5034:
1.203 brouard 5035: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 5036: {
5037: int i;
1.164 brouard 5038: int l=1, lmax=20;
1.126 brouard 5039: double k1,k2,k3,k4,res,fx;
1.132 brouard 5040: double p2[MAXPARM+1];
1.203 brouard 5041: int k, kmax=1;
5042: double v1, v2, cv12, lc1, lc2;
1.208 brouard 5043:
5044: int firstime=0;
1.203 brouard 5045:
1.126 brouard 5046: fx=func(x);
1.203 brouard 5047: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 5048: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 5049: p2[thetai]=x[thetai]+delti[thetai]*k;
5050: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 5051: k1=func(p2)-fx;
5052:
1.203 brouard 5053: p2[thetai]=x[thetai]+delti[thetai]*k;
5054: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 5055: k2=func(p2)-fx;
5056:
1.203 brouard 5057: p2[thetai]=x[thetai]-delti[thetai]*k;
5058: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 5059: k3=func(p2)-fx;
5060:
1.203 brouard 5061: p2[thetai]=x[thetai]-delti[thetai]*k;
5062: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 5063: k4=func(p2)-fx;
1.203 brouard 5064: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
5065: if(k1*k2*k3*k4 <0.){
1.208 brouard 5066: firstime=1;
1.203 brouard 5067: kmax=kmax+10;
1.208 brouard 5068: }
5069: if(kmax >=10 || firstime ==1){
1.246 brouard 5070: printf("Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you could increase ftol=%.2e\n",thetai,thetaj, ftol);
5071: fprintf(ficlog,"Warning: directions %d-%d, you are not estimating the Hessian at the exact maximum likelihood; you could increase ftol=%.2e\n",thetai,thetaj, ftol);
1.203 brouard 5072: printf("%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti*k=%.12e deltj*k=%.12e, xi-de*k=%.12e xj-de*k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
5073: fprintf(ficlog,"%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti*k=%.12e deltj*k=%.12e, xi-de*k=%.12e xj-de*k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
5074: }
5075: #ifdef DEBUGHESSIJ
5076: v1=hess[thetai][thetai];
5077: v2=hess[thetaj][thetaj];
5078: cv12=res;
5079: /* Computing eigen value of Hessian matrix */
5080: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5081: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5082: if ((lc2 <0) || (lc1 <0) ){
5083: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
5084: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
5085: printf("%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
5086: fprintf(ficlog,"%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4);
5087: }
1.126 brouard 5088: #endif
5089: }
5090: return res;
5091: }
5092:
1.203 brouard 5093: /* Not done yet: Was supposed to fix if not exactly at the maximum */
5094: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
5095: /* { */
5096: /* int i; */
5097: /* int l=1, lmax=20; */
5098: /* double k1,k2,k3,k4,res,fx; */
5099: /* double p2[MAXPARM+1]; */
5100: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
5101: /* int k=0,kmax=10; */
5102: /* double l1; */
5103:
5104: /* fx=func(x); */
5105: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
5106: /* l1=pow(10,l); */
5107: /* delts=delt; */
5108: /* for(k=1 ; k <kmax; k=k+1){ */
5109: /* delt = delti*(l1*k); */
5110: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
5111: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5112: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5113: /* k1=func(p2)-fx; */
5114:
5115: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5116: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5117: /* k2=func(p2)-fx; */
5118:
5119: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5120: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5121: /* k3=func(p2)-fx; */
5122:
5123: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5124: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5125: /* k4=func(p2)-fx; */
5126: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
5127: /* #ifdef DEBUGHESSIJ */
5128: /* printf("%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4); */
5129: /* fprintf(ficlog,"%d %d k=%d, k1=%.12e k2=%.12e k3=%.12e k4=%.12e delti/k=%.12e deltj/k=%.12e, xi-de/k=%.12e xj-de/k=%.12e res=%.12e k1234=%.12e,k1-2=%.12e,k3-4=%.12e\n",thetai,thetaj,k,k1,k2,k3,k4,delti[thetai]/k,delti[thetaj]/k,x[thetai]-delti[thetai]/k,x[thetaj]-delti[thetaj]/k, res,k1-k2-k3+k4,k1-k2,k3-k4); */
5130: /* #endif */
5131: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
5132: /* k=kmax; */
5133: /* } */
5134: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
5135: /* k=kmax; l=lmax*10; */
5136: /* } */
5137: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
5138: /* delts=delt; */
5139: /* } */
5140: /* } /\* End loop k *\/ */
5141: /* } */
5142: /* delti[theta]=delts; */
5143: /* return res; */
5144: /* } */
5145:
5146:
1.126 brouard 5147: /************** Inverse of matrix **************/
5148: void ludcmp(double **a, int n, int *indx, double *d)
5149: {
5150: int i,imax,j,k;
5151: double big,dum,sum,temp;
5152: double *vv;
5153:
5154: vv=vector(1,n);
5155: *d=1.0;
5156: for (i=1;i<=n;i++) {
5157: big=0.0;
5158: for (j=1;j<=n;j++)
5159: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 5160: if (big == 0.0){
5161: printf(" Singular Hessian matrix at row %d:\n",i);
5162: for (j=1;j<=n;j++) {
5163: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
5164: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
5165: }
5166: fflush(ficlog);
5167: fclose(ficlog);
5168: nrerror("Singular matrix in routine ludcmp");
5169: }
1.126 brouard 5170: vv[i]=1.0/big;
5171: }
5172: for (j=1;j<=n;j++) {
5173: for (i=1;i<j;i++) {
5174: sum=a[i][j];
5175: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
5176: a[i][j]=sum;
5177: }
5178: big=0.0;
5179: for (i=j;i<=n;i++) {
5180: sum=a[i][j];
5181: for (k=1;k<j;k++)
5182: sum -= a[i][k]*a[k][j];
5183: a[i][j]=sum;
5184: if ( (dum=vv[i]*fabs(sum)) >= big) {
5185: big=dum;
5186: imax=i;
5187: }
5188: }
5189: if (j != imax) {
5190: for (k=1;k<=n;k++) {
5191: dum=a[imax][k];
5192: a[imax][k]=a[j][k];
5193: a[j][k]=dum;
5194: }
5195: *d = -(*d);
5196: vv[imax]=vv[j];
5197: }
5198: indx[j]=imax;
5199: if (a[j][j] == 0.0) a[j][j]=TINY;
5200: if (j != n) {
5201: dum=1.0/(a[j][j]);
5202: for (i=j+1;i<=n;i++) a[i][j] *= dum;
5203: }
5204: }
5205: free_vector(vv,1,n); /* Doesn't work */
5206: ;
5207: }
5208:
5209: void lubksb(double **a, int n, int *indx, double b[])
5210: {
5211: int i,ii=0,ip,j;
5212: double sum;
5213:
5214: for (i=1;i<=n;i++) {
5215: ip=indx[i];
5216: sum=b[ip];
5217: b[ip]=b[i];
5218: if (ii)
5219: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
5220: else if (sum) ii=i;
5221: b[i]=sum;
5222: }
5223: for (i=n;i>=1;i--) {
5224: sum=b[i];
5225: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
5226: b[i]=sum/a[i][i];
5227: }
5228: }
5229:
5230: void pstamp(FILE *fichier)
5231: {
1.196 brouard 5232: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 5233: }
5234:
1.297 brouard 5235: void date2dmy(double date,double *day, double *month, double *year){
5236: double yp=0., yp1=0., yp2=0.;
5237:
5238: yp1=modf(date,&yp);/* extracts integral of date in yp and
5239: fractional in yp1 */
5240: *year=yp;
5241: yp2=modf((yp1*12),&yp);
5242: *month=yp;
5243: yp1=modf((yp2*30.5),&yp);
5244: *day=yp;
5245: if(*day==0) *day=1;
5246: if(*month==0) *month=1;
5247: }
5248:
1.253 brouard 5249:
5250:
1.126 brouard 5251: /************ Frequencies ********************/
1.251 brouard 5252: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 5253: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
5254: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 5255: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 5256: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 5257: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 5258: int iind=0, iage=0;
5259: int mi; /* Effective wave */
5260: int first;
5261: double ***freq; /* Frequencies */
1.268 brouard 5262: double *x, *y, a=0.,b=0.,r=1., sa=0., sb=0.; /* for regression, y=b+m*x and r is the correlation coefficient */
5263: int no=0, linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb);
1.284 brouard 5264: double *meanq, *stdq, *idq;
1.226 brouard 5265: double **meanqt;
5266: double *pp, **prop, *posprop, *pospropt;
5267: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
5268: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
5269: double agebegin, ageend;
5270:
5271: pp=vector(1,nlstate);
1.251 brouard 5272: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5273: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
5274: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
5275: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
5276: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 5277: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 5278: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 5279: meanqt=matrix(1,lastpass,1,nqtveff);
5280: strcpy(fileresp,"P_");
5281: strcat(fileresp,fileresu);
5282: /*strcat(fileresphtm,fileresu);*/
5283: if((ficresp=fopen(fileresp,"w"))==NULL) {
5284: printf("Problem with prevalence resultfile: %s\n", fileresp);
5285: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
5286: exit(0);
5287: }
1.240 brouard 5288:
1.226 brouard 5289: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
5290: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
5291: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5292: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5293: fflush(ficlog);
5294: exit(70);
5295: }
5296: else{
5297: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 5298: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5299: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5300: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5301: }
1.319 brouard 5302: fprintf(ficresphtm,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>Frequencies (weight=%d) and prevalence by age at begin of transition and dummy covariate value at beginning of transition</h4>\n",fileresphtm, fileresphtm, weightopt);
1.240 brouard 5303:
1.226 brouard 5304: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
5305: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
5306: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5307: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5308: fflush(ficlog);
5309: exit(70);
1.240 brouard 5310: } else{
1.226 brouard 5311: fprintf(ficresphtmfr,"<html><head>\n<title>IMaCh PHTM_Frequency table %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.319 brouard 5312: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5313: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5314: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5315: }
1.319 brouard 5316: fprintf(ficresphtmfr,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>(weight=%d) frequencies of all effective transitions of the model, by age at begin of transition, and covariate value at the begin of transition (if the covariate is a varying covariate) </h4>Unknown status is -1<br/>\n",fileresphtmfr, fileresphtmfr,weightopt);
1.240 brouard 5317:
1.253 brouard 5318: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
5319: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 5320: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5321: j1=0;
1.126 brouard 5322:
1.227 brouard 5323: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 brouard 5324: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 5325: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 5326: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 5327:
5328:
1.226 brouard 5329: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
5330: reference=low_education V1=0,V2=0
5331: med_educ V1=1 V2=0,
5332: high_educ V1=0 V2=1
1.330 brouard 5333: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 5334: */
1.249 brouard 5335: dateintsum=0;
5336: k2cpt=0;
5337:
1.253 brouard 5338: if(cptcoveff == 0 )
1.265 brouard 5339: nl=1; /* Constant and age model only */
1.253 brouard 5340: else
5341: nl=2;
1.265 brouard 5342:
5343: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
5344: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 brouard 5345: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 5346: * freq[s1][s2][iage] =0.
5347: * Loop on iind
5348: * ++freq[s1][s2][iage] weighted
5349: * end iind
5350: * if covariate and j!0
5351: * headers Variable on one line
5352: * endif cov j!=0
5353: * header of frequency table by age
5354: * Loop on age
5355: * pp[s1]+=freq[s1][s2][iage] weighted
5356: * pos+=freq[s1][s2][iage] weighted
5357: * Loop on s1 initial state
5358: * fprintf(ficresp
5359: * end s1
5360: * end age
5361: * if j!=0 computes starting values
5362: * end compute starting values
5363: * end j1
5364: * end nl
5365: */
1.253 brouard 5366: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
5367: if(nj==1)
5368: j=0; /* First pass for the constant */
1.265 brouard 5369: else{
1.335 brouard 5370: j=cptcoveff; /* Other passes for the covariate values number of simple covariates in the model V2+V1 =2 (simple dummy fixed or time varying) */
1.265 brouard 5371: }
1.251 brouard 5372: first=1;
1.332 brouard 5373: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on all dummy covariates combination of the model, ie excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
1.251 brouard 5374: posproptt=0.;
1.330 brouard 5375: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 5376: scanf("%d", i);*/
5377: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 5378: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 5379: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 5380: freq[i][s2][m]=0;
1.251 brouard 5381:
5382: for (i=1; i<=nlstate; i++) {
1.240 brouard 5383: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 5384: prop[i][m]=0;
5385: posprop[i]=0;
5386: pospropt[i]=0;
5387: }
1.283 brouard 5388: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 5389: idq[z1]=0.;
5390: meanq[z1]=0.;
5391: stdq[z1]=0.;
1.283 brouard 5392: }
5393: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 5394: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 5395: /* meanqt[m][z1]=0.; */
5396: /* } */
5397: /* } */
1.251 brouard 5398: /* dateintsum=0; */
5399: /* k2cpt=0; */
5400:
1.265 brouard 5401: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 5402: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
5403: bool=1;
5404: if(j !=0){
5405: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 brouard 5406: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
5407: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 5408: /* if(Tvaraff[z1] ==-20){ */
5409: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
5410: /* }else if(Tvaraff[z1] ==-10){ */
5411: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 5412: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 brouard 5413: /* if( iind >=imx-3) printf("Searching error iind=%d Tvaraff[z1]=%d covar[Tvaraff[z1]][iind]=%.f TnsdVar[Tvaraff[z1]]=%d, cptcoveff=%d, cptcovs=%d \n",iind, Tvaraff[z1], covar[Tvaraff[z1]][iind],TnsdVar[Tvaraff[z1]],cptcoveff, cptcovs); */
5414: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
1.338 brouard 5415: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=1+age+%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 5416: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 5417: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 5418: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 5419: /* printf("bool=%d i=%d, z1=%d, Tvaraff[%d]=%d, covar[Tvarff][%d]=%2f, codtabm(%d,%d)=%d, nbcode[Tvaraff][codtabm(%d,%d)=%d, j1=%d\n", */
5420: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
5421: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 5422: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
5423: } /* Onlyf fixed */
5424: } /* end z1 */
1.335 brouard 5425: } /* cptcoveff > 0 */
1.251 brouard 5426: } /* end any */
5427: }/* end j==0 */
1.265 brouard 5428: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 5429: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 5430: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 5431: m=mw[mi][iind];
5432: if(j!=0){
5433: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 brouard 5434: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 5435: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5436: /* iv= Tvar[Tmodelind[z1]]-ncovcol-nqv; /\* Good *\/ */
5437: iv= Tvar[Tmodelind[z1]]; /* Good *//* because cotvar starts now at first at ncovcol+nqv+ntv */
1.332 brouard 5438: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality. If covariate's
1.251 brouard 5439: value is -1, we don't select. It differs from the
5440: constant and age model which counts them. */
5441: bool=0; /* not selected */
5442: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 5443: /* i1=Tvaraff[z1]; */
5444: /* i2=TnsdVar[i1]; */
5445: /* i3=nbcode[i1][i2]; */
5446: /* i4=covar[i1][iind]; */
5447: /* if(i4 != i3){ */
5448: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 5449: bool=0;
5450: }
5451: }
5452: }
5453: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
5454: } /* end j==0 */
5455: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 5456: if(bool==1){ /*Selected */
1.251 brouard 5457: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
5458: and mw[mi+1][iind]. dh depends on stepm. */
5459: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
5460: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
5461: if(m >=firstpass && m <=lastpass){
5462: k2=anint[m][iind]+(mint[m][iind]/12.);
5463: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
5464: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
5465: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
5466: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
5467: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
5468: if (m<lastpass) {
5469: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
5470: /* printf(" num=%ld m=%d, iind=%d s1=%d s2=%d agev at m=%d\n", num[iind], m, iind,s[m][iind],s[m+1][iind], (int)agev[m][iind]); */
5471: if(s[m][iind]==-1)
5472: printf(" num=%ld m=%d, iind=%d s1=%d s2=%d agev at m=%d agebegin=%.2f ageend=%.2f, agemed=%d\n", num[iind], m, iind,s[m][iind],s[m+1][iind], (int)agev[m][iind],agebegin, ageend, (int)((agebegin+ageend)/2.));
5473: freq[s[m][iind]][s[m+1][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
1.311 brouard 5474: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
5475: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 5476: idq[z1]=idq[z1]+weight[iind];
5477: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
5478: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
5479: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 5480: }
1.284 brouard 5481: }
1.251 brouard 5482: /* if((int)agev[m][iind] == 55) */
5483: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5484: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5485: freq[s[m][iind]][s[m+1][iind]][iagemax+3] += weight[iind]; /* Total is in iagemax+3 *//* At age of beginning of transition, where status is known */
1.234 brouard 5486: }
1.251 brouard 5487: } /* end if between passes */
5488: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5489: dateintsum=dateintsum+k2; /* on all covariates ?*/
5490: k2cpt++;
5491: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5492: }
1.251 brouard 5493: }else{
5494: bool=1;
5495: }/* end bool 2 */
5496: } /* end m */
1.284 brouard 5497: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5498: /* idq[z1]=idq[z1]+weight[iind]; */
5499: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5500: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5501: /* } */
1.251 brouard 5502: } /* end bool */
5503: } /* end iind = 1 to imx */
1.319 brouard 5504: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5505: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5506:
5507:
5508: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 brouard 5509: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5510: pstamp(ficresp);
1.335 brouard 5511: if (cptcoveff>0 && j!=0){
1.265 brouard 5512: pstamp(ficresp);
1.251 brouard 5513: printf( "\n#********** Variable ");
5514: fprintf(ficresp, "\n#********** Variable ");
5515: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5516: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5517: fprintf(ficlog, "\n#********** Variable ");
1.340 brouard 5518: for (z1=1; z1<=cptcoveff; z1++){
1.251 brouard 5519: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 5520: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5521: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5522: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5523: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5524: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5525: }else{
1.330 brouard 5526: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5527: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5528: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5529: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5530: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5531: }
5532: }
5533: printf( "**********\n#");
5534: fprintf(ficresp, "**********\n#");
5535: fprintf(ficresphtm, "**********</h3>\n");
5536: fprintf(ficresphtmfr, "**********</h3>\n");
5537: fprintf(ficlog, "**********\n");
5538: }
1.284 brouard 5539: /*
5540: Printing means of quantitative variables if any
5541: */
5542: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5543: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5544: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5545: if(weightopt==1){
5546: printf(" Weighted mean and standard deviation of");
5547: fprintf(ficlog," Weighted mean and standard deviation of");
5548: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5549: }
1.311 brouard 5550: /* mu = \frac{w x}{\sum w}
5551: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5552: */
5553: printf(" fixed quantitative variable V%d on %.3g (weighted) representatives of the population : %8.5g (%8.5g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt(stdq[z1]/idq[z1]-meanq[z1]*meanq[z1]/idq[z1]/idq[z1]));
5554: fprintf(ficlog," fixed quantitative variable V%d on %.3g (weighted) representatives of the population : %8.5g (%8.5g)\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt(stdq[z1]/idq[z1]-meanq[z1]*meanq[z1]/idq[z1]/idq[z1]));
5555: fprintf(ficresphtmfr," fixed quantitative variable V%d on %.3g (weighted) representatives of the population : %8.5g (%8.5g)<p>\n", ncovcol+z1, idq[z1],meanq[z1]/idq[z1], sqrt(stdq[z1]/idq[z1]-meanq[z1]*meanq[z1]/idq[z1]/idq[z1]));
1.284 brouard 5556: }
5557: /* for (z1=1; z1<= nqtveff; z1++) { */
5558: /* for(m=1;m<=lastpass;m++){ */
5559: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5560: /* } */
5561: /* } */
1.283 brouard 5562:
1.251 brouard 5563: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 brouard 5564: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5565: fprintf(ficresp, " Age");
1.335 brouard 5566: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
5567: printf(" V%d=%d, z1=%d, Tvaraff[z1]=%d, j1=%d, TnsdVar[Tvaraff[%d]]=%d |",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])], z1, Tvaraff[z1], j1,z1,TnsdVar[Tvaraff[z1]]);
5568: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5569: }
1.251 brouard 5570: for(i=1; i<=nlstate;i++) {
1.335 brouard 5571: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5572: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5573: }
1.335 brouard 5574: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5575: fprintf(ficresphtm, "\n");
5576:
5577: /* Header of frequency table by age */
5578: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5579: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5580: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5581: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5582: if(s2!=0 && m!=0)
5583: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5584: }
1.226 brouard 5585: }
1.251 brouard 5586: fprintf(ficresphtmfr, "\n");
5587:
5588: /* For each age */
5589: for(iage=iagemin; iage <= iagemax+3; iage++){
5590: fprintf(ficresphtm,"<tr>");
5591: if(iage==iagemax+1){
5592: fprintf(ficlog,"1");
5593: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5594: }else if(iage==iagemax+2){
5595: fprintf(ficlog,"0");
5596: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5597: }else if(iage==iagemax+3){
5598: fprintf(ficlog,"Total");
5599: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5600: }else{
1.240 brouard 5601: if(first==1){
1.251 brouard 5602: first=0;
5603: printf("See log file for details...\n");
5604: }
5605: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5606: fprintf(ficlog,"Age %d", iage);
5607: }
1.265 brouard 5608: for(s1=1; s1 <=nlstate ; s1++){
5609: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5610: pp[s1] += freq[s1][m][iage];
1.251 brouard 5611: }
1.265 brouard 5612: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5613: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5614: pos += freq[s1][m][iage];
5615: if(pp[s1]>=1.e-10){
1.251 brouard 5616: if(first==1){
1.265 brouard 5617: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5618: }
1.265 brouard 5619: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5620: }else{
5621: if(first==1)
1.265 brouard 5622: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5623: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5624: }
5625: }
5626:
1.265 brouard 5627: for(s1=1; s1 <=nlstate ; s1++){
5628: /* posprop[s1]=0; */
5629: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5630: pp[s1] += freq[s1][m][iage];
5631: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5632:
5633: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5634: pos += pp[s1]; /* pos is the total number of transitions until this age */
5635: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5636: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5637: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5638: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5639: }
5640:
5641: /* Writing ficresp */
1.335 brouard 5642: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5643: if( iage <= iagemax){
5644: fprintf(ficresp," %d",iage);
5645: }
5646: }else if( nj==2){
5647: if( iage <= iagemax){
5648: fprintf(ficresp," %d",iage);
1.335 brouard 5649: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 5650: }
1.240 brouard 5651: }
1.265 brouard 5652: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5653: if(pos>=1.e-5){
1.251 brouard 5654: if(first==1)
1.265 brouard 5655: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5656: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5657: }else{
5658: if(first==1)
1.265 brouard 5659: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5660: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5661: }
5662: if( iage <= iagemax){
5663: if(pos>=1.e-5){
1.335 brouard 5664: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5665: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5666: }else if( nj==2){
5667: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5668: }
5669: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5670: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5671: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5672: } else{
1.335 brouard 5673: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5674: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5675: }
1.240 brouard 5676: }
1.265 brouard 5677: pospropt[s1] +=posprop[s1];
5678: } /* end loop s1 */
1.251 brouard 5679: /* pospropt=0.; */
1.265 brouard 5680: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5681: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5682: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5683: if(first==1){
1.265 brouard 5684: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5685: }
1.265 brouard 5686: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5687: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5688: }
1.265 brouard 5689: if(s1!=0 && m!=0)
5690: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5691: }
1.265 brouard 5692: } /* end loop s1 */
1.251 brouard 5693: posproptt=0.;
1.265 brouard 5694: for(s1=1; s1 <=nlstate; s1++){
5695: posproptt += pospropt[s1];
1.251 brouard 5696: }
5697: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5698: fprintf(ficresphtm,"</tr>\n");
1.335 brouard 5699: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 5700: if(iage <= iagemax)
5701: fprintf(ficresp,"\n");
1.240 brouard 5702: }
1.251 brouard 5703: if(first==1)
5704: printf("Others in log...\n");
5705: fprintf(ficlog,"\n");
5706: } /* end loop age iage */
1.265 brouard 5707:
1.251 brouard 5708: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5709: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5710: if(posproptt < 1.e-5){
1.265 brouard 5711: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5712: }else{
1.265 brouard 5713: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5714: }
1.226 brouard 5715: }
1.251 brouard 5716: fprintf(ficresphtm,"</tr>\n");
5717: fprintf(ficresphtm,"</table>\n");
5718: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5719: if(posproptt < 1.e-5){
1.251 brouard 5720: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5721: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5722: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5723: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5724: invalidvarcomb[j1]=1;
1.226 brouard 5725: }else{
1.338 brouard 5726: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced (or no resultline).</p>",j1);
1.251 brouard 5727: invalidvarcomb[j1]=0;
1.226 brouard 5728: }
1.251 brouard 5729: fprintf(ficresphtmfr,"</table>\n");
5730: fprintf(ficlog,"\n");
5731: if(j!=0){
5732: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5733: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5734: for(k=1; k <=(nlstate+ndeath); k++){
5735: if (k != i) {
1.265 brouard 5736: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5737: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5738: if(j1==1){ /* All dummy covariates to zero */
5739: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5740: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5741: printf("%d%d ",i,k);
5742: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5743: printf("%12.7f ln(%.0f/%.0f)= %f, OR=%f sd=%f \n",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]),freq[i][k][iagemax+3]/freq[i][i][iagemax+3], sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]));
5744: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f \n",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
5745: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5746: }
1.253 brouard 5747: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5748: for(iage=iagemin; iage <= iagemax+3; iage++){
5749: x[iage]= (double)iage;
5750: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5751: /* printf("i=%d, k=%d, s1=%d, j1=%d, jj=%d, y[%d]=%f\n",i,k,s1,j1,jj, iage, y[iage]); */
1.253 brouard 5752: }
1.268 brouard 5753: /* Some are not finite, but linreg will ignore these ages */
5754: no=0;
1.253 brouard 5755: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5756: pstart[s1]=b;
5757: pstart[s1-1]=a;
1.252 brouard 5758: }else if( j1!=1 && (j1==2 || (log(j1-1.)/log(2.)-(int)(log(j1-1.)/log(2.))) <0.010) && ( TvarsDind[(int)(log(j1-1.)/log(2.))+1]+2+nagesqr == jj) && Dummy[jj-2-nagesqr]==0){ /* We want only if the position, jj, in model corresponds to unique covariate equal to 1 in j1 combination */
5759: printf("j1=%d, jj=%d, (int)(log(j1-1.)/log(2.))+1=%d, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(int)(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
5760: printf("j1=%d, jj=%d, (log(j1-1.)/log(2.))+1=%f, TvarsDind[(int)(log(j1-1.)/log(2.))+1]=%d\n",j1, jj,(log(j1-1.)/log(2.))+1,TvarsDind[(int)(log(j1-1.)/log(2.))+1]);
1.265 brouard 5761: pstart[s1]= log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]));
1.252 brouard 5762: printf("%d%d ",i,k);
5763: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5764: printf("s1=%d,i=%d,k=%d,p[%d]=%12.7f ln((%.0f/%.0f)/(%.0f/%.0f))= %f, OR=%f sd=%f \n",s1,i,k,s1,p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3],freq[i][k][iagemax+4],freq[i][i][iagemax+4], log((freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4])),(freq[i][k][iagemax+3]/freq[i][i][iagemax+3])/(freq[i][k][iagemax+4]/freq[i][i][iagemax+4]), sqrt(1/freq[i][k][iagemax+3]+1/freq[i][i][iagemax+3]+1/freq[i][k][iagemax+4]+1/freq[i][i][iagemax+4]));
1.251 brouard 5765: }else{ /* Other cases, like quantitative fixed or varying covariates */
5766: ;
5767: }
5768: /* printf("%12.7f )", param[i][jj][k]); */
5769: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5770: s1++;
1.251 brouard 5771: } /* end jj */
5772: } /* end k!= i */
5773: } /* end k */
1.265 brouard 5774: } /* end i, s1 */
1.251 brouard 5775: } /* end j !=0 */
5776: } /* end selected combination of covariate j1 */
5777: if(j==0){ /* We can estimate starting values from the occurences in each case */
5778: printf("#Freqsummary: Starting values for the constants:\n");
5779: fprintf(ficlog,"\n");
1.265 brouard 5780: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5781: for(k=1; k <=(nlstate+ndeath); k++){
5782: if (k != i) {
5783: printf("%d%d ",i,k);
5784: fprintf(ficlog,"%d%d ",i,k);
5785: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5786: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5787: if(jj==1){ /* Age has to be done */
1.265 brouard 5788: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5789: printf("%12.7f ln(%.0f/%.0f)= %12.7f ",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
5790: fprintf(ficlog,"%12.7f ln(%.0f/%.0f)= %12.7f ",p[s1],freq[i][k][iagemax+3],freq[i][i][iagemax+3], log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]));
1.251 brouard 5791: }
5792: /* printf("%12.7f )", param[i][jj][k]); */
5793: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5794: s1++;
1.250 brouard 5795: }
1.251 brouard 5796: printf("\n");
5797: fprintf(ficlog,"\n");
1.250 brouard 5798: }
5799: }
1.284 brouard 5800: } /* end of state i */
1.251 brouard 5801: printf("#Freqsummary\n");
5802: fprintf(ficlog,"\n");
1.265 brouard 5803: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5804: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5805: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5806: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5807: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5808: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5809: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5810: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5811: /* } */
5812: }
1.265 brouard 5813: } /* end loop s1 */
1.251 brouard 5814:
5815: printf("\n");
5816: fprintf(ficlog,"\n");
5817: } /* end j=0 */
1.249 brouard 5818: } /* end j */
1.252 brouard 5819:
1.253 brouard 5820: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5821: for(i=1, jk=1; i <=nlstate; i++){
5822: for(j=1; j <=nlstate+ndeath; j++){
5823: if(j!=i){
5824: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5825: printf("%1d%1d",i,j);
5826: fprintf(ficparo,"%1d%1d",i,j);
5827: for(k=1; k<=ncovmodel;k++){
5828: /* printf(" %lf",param[i][j][k]); */
5829: /* fprintf(ficparo," %lf",param[i][j][k]); */
5830: p[jk]=pstart[jk];
5831: printf(" %f ",pstart[jk]);
5832: fprintf(ficparo," %f ",pstart[jk]);
5833: jk++;
5834: }
5835: printf("\n");
5836: fprintf(ficparo,"\n");
5837: }
5838: }
5839: }
5840: } /* end mle=-2 */
1.226 brouard 5841: dateintmean=dateintsum/k2cpt;
1.296 brouard 5842: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5843:
1.226 brouard 5844: fclose(ficresp);
5845: fclose(ficresphtm);
5846: fclose(ficresphtmfr);
1.283 brouard 5847: free_vector(idq,1,nqfveff);
1.226 brouard 5848: free_vector(meanq,1,nqfveff);
1.284 brouard 5849: free_vector(stdq,1,nqfveff);
1.226 brouard 5850: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5851: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5852: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5853: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5854: free_vector(pospropt,1,nlstate);
5855: free_vector(posprop,1,nlstate);
1.251 brouard 5856: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5857: free_vector(pp,1,nlstate);
5858: /* End of freqsummary */
5859: }
1.126 brouard 5860:
1.268 brouard 5861: /* Simple linear regression */
5862: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5863:
5864: /* y=a+bx regression */
5865: double sumx = 0.0; /* sum of x */
5866: double sumx2 = 0.0; /* sum of x**2 */
5867: double sumxy = 0.0; /* sum of x * y */
5868: double sumy = 0.0; /* sum of y */
5869: double sumy2 = 0.0; /* sum of y**2 */
5870: double sume2 = 0.0; /* sum of square or residuals */
5871: double yhat;
5872:
5873: double denom=0;
5874: int i;
5875: int ne=*no;
5876:
5877: for ( i=ifi, ne=0;i<=ila;i++) {
5878: if(!isfinite(x[i]) || !isfinite(y[i])){
5879: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5880: continue;
5881: }
5882: ne=ne+1;
5883: sumx += x[i];
5884: sumx2 += x[i]*x[i];
5885: sumxy += x[i] * y[i];
5886: sumy += y[i];
5887: sumy2 += y[i]*y[i];
5888: denom = (ne * sumx2 - sumx*sumx);
5889: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
5890: }
5891:
5892: denom = (ne * sumx2 - sumx*sumx);
5893: if (denom == 0) {
5894: // vertical, slope m is infinity
5895: *b = INFINITY;
5896: *a = 0;
5897: if (r) *r = 0;
5898: return 1;
5899: }
5900:
5901: *b = (ne * sumxy - sumx * sumy) / denom;
5902: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5903: if (r!=NULL) {
5904: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5905: sqrt((sumx2 - sumx*sumx/ne) *
5906: (sumy2 - sumy*sumy/ne));
5907: }
5908: *no=ne;
5909: for ( i=ifi, ne=0;i<=ila;i++) {
5910: if(!isfinite(x[i]) || !isfinite(y[i])){
5911: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5912: continue;
5913: }
5914: ne=ne+1;
5915: yhat = y[i] - *a -*b* x[i];
5916: sume2 += yhat * yhat ;
5917:
5918: denom = (ne * sumx2 - sumx*sumx);
5919: /* printf("ne=%d, i=%d,x[%d]=%f, y[%d]=%f sumx=%f, sumx2=%f, sumxy=%f, sumy=%f, sumy2=%f, denom=%f\n",ne,i,i,x[i],i,y[i], sumx, sumx2,sumxy, sumy, sumy2,denom); */
5920: }
5921: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5922: *sa= *sb * sqrt(sumx2/ne);
5923:
5924: return 0;
5925: }
5926:
1.126 brouard 5927: /************ Prevalence ********************/
1.227 brouard 5928: void prevalence(double ***probs, double agemin, double agemax, int **s, double **agev, int nlstate, int imx, int *Tvar, int **nbcode, int *ncodemax,double **mint,double **anint, double dateprev1,double dateprev2, int firstpass, int lastpass)
5929: {
5930: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5931: in each health status at the date of interview (if between dateprev1 and dateprev2).
5932: We still use firstpass and lastpass as another selection.
5933: */
1.126 brouard 5934:
1.227 brouard 5935: int i, m, jk, j1, bool, z1,j, iv;
5936: int mi; /* Effective wave */
5937: int iage;
5938: double agebegin, ageend;
5939:
5940: double **prop;
5941: double posprop;
5942: double y2; /* in fractional years */
5943: int iagemin, iagemax;
5944: int first; /** to stop verbosity which is redirected to log file */
5945:
5946: iagemin= (int) agemin;
5947: iagemax= (int) agemax;
5948: /*pp=vector(1,nlstate);*/
1.251 brouard 5949: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5950: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5951: j1=0;
1.222 brouard 5952:
1.227 brouard 5953: /*j=cptcoveff;*/
5954: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5955:
1.288 brouard 5956: first=0;
1.335 brouard 5957: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 5958: for (i=1; i<=nlstate; i++)
1.251 brouard 5959: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5960: prop[i][iage]=0.0;
5961: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5962: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5963: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5964:
5965: for (i=1; i<=imx; i++) { /* Each individual */
5966: bool=1;
5967: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5968: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5969: m=mw[mi][i];
5970: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5971: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5972: for (z1=1; z1<=cptcoveff; z1++){
5973: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5974: iv= Tvar[Tmodelind[z1]];/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.332 brouard 5975: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 5976: bool=0;
5977: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 5978: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 5979: bool=0;
5980: }
5981: }
5982: if(bool==1){ /* Otherwise we skip that wave/person */
5983: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5984: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5985: if(m >=firstpass && m <=lastpass){
5986: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5987: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5988: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5989: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5990: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5991: printf("Error on individual # %d agev[m][i]=%f <%d-%d or > %d+3+%d m=%d; either change agemin or agemax or fix data\n",i, agev[m][i],iagemin,AGEMARGE, iagemax,AGEMARGE,m);
5992: exit(1);
5993: }
5994: if (s[m][i]>0 && s[m][i]<=nlstate) {
5995: /*if(i>4620) printf(" i=%d m=%d s[m][i]=%d (int)agev[m][i]=%d weight[i]=%f prop=%f\n",i,m,s[m][i],(int)agev[m][m],weight[i],prop[s[m][i]][(int)agev[m][i]]);*/
5996: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5997: prop[s[m][i]][iagemax+3] += weight[i];
5998: } /* end valid statuses */
5999: } /* end selection of dates */
6000: } /* end selection of waves */
6001: } /* end bool */
6002: } /* end wave */
6003: } /* end individual */
6004: for(i=iagemin; i <= iagemax+3; i++){
6005: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
6006: posprop += prop[jk][i];
6007: }
6008:
6009: for(jk=1; jk <=nlstate ; jk++){
6010: if( i <= iagemax){
6011: if(posprop>=1.e-5){
6012: probs[i][jk][j1]= prop[jk][i]/posprop;
6013: } else{
1.288 brouard 6014: if(!first){
6015: first=1;
1.266 brouard 6016: printf("Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases\nSee others in log file...\n",jk,i,jk, j1,probs[i][jk][j1]);
6017: }else{
1.288 brouard 6018: fprintf(ficlog,"Warning Observed prevalence doesn't sum to 1 for state %d: probs[%d][%d][%d]=%lf because of lack of cases.\n",jk,i,jk, j1,probs[i][jk][j1]);
1.227 brouard 6019: }
6020: }
6021: }
6022: }/* end jk */
6023: }/* end i */
1.222 brouard 6024: /*} *//* end i1 */
1.227 brouard 6025: } /* end j1 */
1.222 brouard 6026:
1.227 brouard 6027: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
6028: /*free_vector(pp,1,nlstate);*/
1.251 brouard 6029: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 6030: } /* End of prevalence */
1.126 brouard 6031:
6032: /************* Waves Concatenation ***************/
6033:
6034: void concatwav(int wav[], int **dh, int **bh, int **mw, int **s, double *agedc, double **agev, int firstpass, int lastpass, int imx, int nlstate, int stepm)
6035: {
1.298 brouard 6036: /* Concatenates waves: wav[i] is the number of effective (useful waves in the sense that a non interview is useless) of individual i.
1.126 brouard 6037: Death is a valid wave (if date is known).
6038: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
6039: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 6040: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 6041: */
1.126 brouard 6042:
1.224 brouard 6043: int i=0, mi=0, m=0, mli=0;
1.126 brouard 6044: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
6045: double sum=0., jmean=0.;*/
1.224 brouard 6046: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 6047: int j, k=0,jk, ju, jl;
6048: double sum=0.;
6049: first=0;
1.214 brouard 6050: firstwo=0;
1.217 brouard 6051: firsthree=0;
1.218 brouard 6052: firstfour=0;
1.164 brouard 6053: jmin=100000;
1.126 brouard 6054: jmax=-1;
6055: jmean=0.;
1.224 brouard 6056:
6057: /* Treating live states */
1.214 brouard 6058: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 6059: mi=0; /* First valid wave */
1.227 brouard 6060: mli=0; /* Last valid wave */
1.309 brouard 6061: m=firstpass; /* Loop on waves */
6062: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 6063: if(m >firstpass && s[m][i]==s[m-1][i] && mint[m][i]==mint[m-1][i] && anint[m][i]==anint[m-1][i]){/* Two succesive identical information on wave m */
6064: mli=m-1;/* mw[++mi][i]=m-1; */
6065: }else if(s[m][i]>=1 || s[m][i]==-4 || s[m][i]==-5){ /* Since 0.98r4 if status=-2 vital status is really unknown, wave should be skipped */
1.309 brouard 6066: mw[++mi][i]=m; /* Valid wave: incrementing mi and updating mi; mw[mi] is the wave number of mi_th valid transition */
1.227 brouard 6067: mli=m;
1.224 brouard 6068: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
6069: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 6070: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 6071: }
1.309 brouard 6072: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 6073: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 6074: break;
1.224 brouard 6075: #else
1.317 brouard 6076: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){ /* no death date and known date of interview, case -2 (vital status unknown is warned later */
1.227 brouard 6077: if(firsthree == 0){
1.302 brouard 6078: printf("Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as 1-p_{%d%d} .\nOthers in log file only\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m, s[m][i], nlstate+ndeath);
1.227 brouard 6079: firsthree=1;
1.317 brouard 6080: }else if(firsthree >=1 && firsthree < 10){
6081: fprintf(ficlog,"Information! Unknown status for individual %ld line=%d occurred at last wave %d at known date %d/%d. Please, check if your unknown date of death %d/%d means a live state %d at wave %d. This case(%d)/wave(%d) contributes to the likelihood as 1-p_{%d%d} .\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], (int) moisdc[i], (int) andc[i], s[m][i], m, i, m, s[m][i], nlstate+ndeath);
6082: firsthree++;
6083: }else if(firsthree == 10){
6084: printf("Information, too many Information flags: no more reported to log either\n");
6085: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
6086: firsthree++;
6087: }else{
6088: firsthree++;
1.227 brouard 6089: }
1.309 brouard 6090: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 6091: mli=m;
6092: }
6093: if(s[m][i]==-2){ /* Vital status is really unknown */
6094: nbwarn++;
1.309 brouard 6095: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 6096: printf("Warning! Vital status for individual %ld (line=%d) at last wave %d interviewed at date %d/%d is unknown %d. Please, check if the vital status and the date of death %d/%d are really unknown. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\nOthers in log file only\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], (int) moisdc[i], (int) andc[i], i, m);
6097: fprintf(ficlog,"Warning! Vital status for individual %ld (line=%d) at last wave %d interviewed at date %d/%d is unknown %d. Please, check if the vital status and the date of death %d/%d are really unknown. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\n",num[i],i,lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], (int) moisdc[i], (int) andc[i], i, m);
6098: }
6099: break;
6100: }
6101: break;
1.224 brouard 6102: #endif
1.227 brouard 6103: }/* End m >= lastpass */
1.126 brouard 6104: }/* end while */
1.224 brouard 6105:
1.227 brouard 6106: /* mi is the last effective wave, m is lastpass, mw[j][i] gives the # of j-th effective wave for individual i */
1.216 brouard 6107: /* After last pass */
1.224 brouard 6108: /* Treating death states */
1.214 brouard 6109: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 6110: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
6111: /* } */
1.126 brouard 6112: mi++; /* Death is another wave */
6113: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 6114: /* Only death is a correct wave */
1.126 brouard 6115: mw[mi][i]=m;
1.257 brouard 6116: } /* else not in a death state */
1.224 brouard 6117: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 6118: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 6119: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 6120: if((andc[i]+moisdc[i]/12.) <=(anint[m][i]+mint[m][i]/12.)){ /* month of death occured before last wave month and status should have been death instead of -1 */
1.227 brouard 6121: nbwarn++;
6122: if(firstfiv==0){
1.309 brouard 6123: printf("Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d, interviewed on %d/%d and should have been coded as death instead of '%d'. This case (%d)/wave (%d) is contributing to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 6124: firstfiv=1;
6125: }else{
1.309 brouard 6126: fprintf(ficlog,"Warning! Death for individual %ld line=%d occurred at %d/%d before last wave %d, interviewed on %d/%d and should have been coded as death instead of '%d'. This case (%d)/wave (%d) is contributing to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 6127: }
1.309 brouard 6128: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
6129: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 6130: nberr++;
6131: if(firstwo==0){
1.309 brouard 6132: printf("Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d with status %d. Potential bias if other individuals are still alive on this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood. Please add a new fictitious wave at the date of last vital status scan, with a dead status. See documentation\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 6133: firstwo=1;
6134: }
1.309 brouard 6135: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred at %d/%d after last wave %d interviewed at %d/%d with status %d. Potential bias if other individuals are still alive on this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood. Please add a new fictitious wave at the date of last vital status scan, with a dead status. See documentation\n\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 6136: }
1.257 brouard 6137: }else{ /* if date of interview is unknown */
1.227 brouard 6138: /* death is known but not confirmed by death status at any wave */
6139: if(firstfour==0){
1.309 brouard 6140: printf("Error! Death for individual %ld line=%d occurred %d/%d but not confirmed by any death status for any wave, including last wave %d at unknown date %d/%d with status %d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\nOthers in log file only\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.227 brouard 6141: firstfour=1;
6142: }
1.309 brouard 6143: fprintf(ficlog,"Error! Death for individual %ld line=%d occurred %d/%d but not confirmed by any death status for any wave, including last wave %d at unknown date %d/%d with status %d. Potential bias if other individuals are still alive at this date but ignored. This case (%d)/wave (%d) is skipped, no contribution to likelihood.\n",num[i],i,(int) moisdc[i], (int) andc[i], lastpass,(int)mint[m][i],(int)anint[m][i], s[m][i], i,m );
1.214 brouard 6144: }
1.224 brouard 6145: } /* end if date of death is known */
6146: #endif
1.309 brouard 6147: wav[i]=mi; /* mi should be the last effective wave (or mli), */
6148: /* wav[i]=mw[mi][i]; */
1.126 brouard 6149: if(mi==0){
6150: nbwarn++;
6151: if(first==0){
1.227 brouard 6152: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
6153: first=1;
1.126 brouard 6154: }
6155: if(first==1){
1.227 brouard 6156: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 6157: }
6158: } /* end mi==0 */
6159: } /* End individuals */
1.214 brouard 6160: /* wav and mw are no more changed */
1.223 brouard 6161:
1.317 brouard 6162: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
6163: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
6164:
6165:
1.126 brouard 6166: for(i=1; i<=imx; i++){
6167: for(mi=1; mi<wav[i];mi++){
6168: if (stepm <=0)
1.227 brouard 6169: dh[mi][i]=1;
1.126 brouard 6170: else{
1.260 brouard 6171: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 6172: if (agedc[i] < 2*AGESUP) {
6173: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
6174: if(j==0) j=1; /* Survives at least one month after exam */
6175: else if(j<0){
6176: nberr++;
6177: printf("Error! Negative delay (%d to death) between waves %d and %d of individual %ld at line %d who is aged %.1f with statuses from %d to %d\n ",j,mw[mi][i],mw[mi+1][i],num[i], i,agev[mw[mi][i]][i],s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);
6178: j=1; /* Temporary Dangerous patch */
6179: printf(" We assumed that the date of interview was correct (and not the date of death) and postponed the death %d month(s) (one stepm) after the interview. You MUST fix the contradiction between dates.\n",stepm);
6180: fprintf(ficlog,"Error! Negative delay (%d to death) between waves %d and %d of individual %ld at line %d who is aged %.1f with statuses from %d to %d\n ",j,mw[mi][i],mw[mi+1][i],num[i], i,agev[mw[mi][i]][i],s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);
6181: fprintf(ficlog," We assumed that the date of interview was correct (and not the date of death) and postponed the death %d month(s) (one stepm) after the interview. You MUST fix the contradiction between dates.\n",stepm);
6182: }
6183: k=k+1;
6184: if (j >= jmax){
6185: jmax=j;
6186: ijmax=i;
6187: }
6188: if (j <= jmin){
6189: jmin=j;
6190: ijmin=i;
6191: }
6192: sum=sum+j;
6193: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
6194: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
6195: }
6196: }
6197: else{
6198: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 6199: /* if (j<0) printf("%d %lf %lf %d %d %d\n", i,agev[mw[mi+1][i]][i], agev[mw[mi][i]][i],j,s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]); */
1.223 brouard 6200:
1.227 brouard 6201: k=k+1;
6202: if (j >= jmax) {
6203: jmax=j;
6204: ijmax=i;
6205: }
6206: else if (j <= jmin){
6207: jmin=j;
6208: ijmin=i;
6209: }
6210: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
6211: /*printf("%d %lf %d %d %d\n", i,agev[mw[mi][i]][i],j,s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);*/
6212: if(j<0){
6213: nberr++;
6214: printf("Error! Negative delay (%d) between waves %d and %d of individual %ld at line %d who is aged %.1f with statuses from %d to %d\n ",j,mw[mi][i],mw[mi+1][i],num[i], i,agev[mw[mi][i]][i],s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);
6215: fprintf(ficlog,"Error! Negative delay (%d) between waves %d and %d of individual %ld at line %d who is aged %.1f with statuses from %d to %d\n ",j,mw[mi][i],mw[mi+1][i],num[i], i,agev[mw[mi][i]][i],s[mw[mi][i]][i] ,s[mw[mi+1][i]][i]);
6216: }
6217: sum=sum+j;
6218: }
6219: jk= j/stepm;
6220: jl= j -jk*stepm;
6221: ju= j -(jk+1)*stepm;
6222: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
6223: if(jl==0){
6224: dh[mi][i]=jk;
6225: bh[mi][i]=0;
6226: }else{ /* We want a negative bias in order to only have interpolation ie
6227: * to avoid the price of an extra matrix product in likelihood */
6228: dh[mi][i]=jk+1;
6229: bh[mi][i]=ju;
6230: }
6231: }else{
6232: if(jl <= -ju){
6233: dh[mi][i]=jk;
6234: bh[mi][i]=jl; /* bias is positive if real duration
6235: * is higher than the multiple of stepm and negative otherwise.
6236: */
6237: }
6238: else{
6239: dh[mi][i]=jk+1;
6240: bh[mi][i]=ju;
6241: }
6242: if(dh[mi][i]==0){
6243: dh[mi][i]=1; /* At least one step */
6244: bh[mi][i]=ju; /* At least one step */
6245: /* printf(" bh=%d ju=%d jl=%d dh=%d jk=%d stepm=%d %d\n",bh[mi][i],ju,jl,dh[mi][i],jk,stepm,i);*/
6246: }
6247: } /* end if mle */
1.126 brouard 6248: }
6249: } /* end wave */
6250: }
6251: jmean=sum/k;
6252: printf("Delay (in months) between two waves Min=%d (for indiviudal %ld) Max=%d (%ld) Mean=%f\n\n ",jmin, num[ijmin], jmax, num[ijmax], jmean);
1.141 brouard 6253: fprintf(ficlog,"Delay (in months) between two waves Min=%d (for indiviudal %d) Max=%d (%d) Mean=%f\n\n ",jmin, ijmin, jmax, ijmax, jmean);
1.227 brouard 6254: }
1.126 brouard 6255:
6256: /*********** Tricode ****************************/
1.220 brouard 6257: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 6258: {
6259: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
6260: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
6261: * Boring subroutine which should only output nbcode[Tvar[j]][k]
6262: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
6263: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
6264: */
1.130 brouard 6265:
1.242 brouard 6266: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
6267: int modmaxcovj=0; /* Modality max of covariates j */
6268: int cptcode=0; /* Modality max of covariates j */
6269: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 6270:
6271:
1.242 brouard 6272: /* cptcoveff=0; */
6273: /* *cptcov=0; */
1.126 brouard 6274:
1.242 brouard 6275: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 6276: for (k=1; k <= maxncov; k++)
6277: for(j=1; j<=2; j++)
6278: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 6279:
1.242 brouard 6280: /* Loop on covariates without age and products and no quantitative variable */
1.335 brouard 6281: for (k=1; k<=cptcovt; k++) { /* cptcovt: total number of covariates of the model (2) nbocc(+)+1 = 8 excepting constant and age and age*age */
1.242 brouard 6282: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
1.343 brouard 6283: /* printf("Testing k=%d, cptcovt=%d\n",k, cptcovt); */
1.339 brouard 6284: if(Dummy[k]==0 && Typevar[k] !=1 && Typevar[k] != 2){ /* Dummy covariate and not age product nor fixed product */
1.242 brouard 6285: switch(Fixed[k]) {
6286: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 6287: modmaxcovj=0;
6288: modmincovj=0;
1.242 brouard 6289: for (i=1; i<=imx; i++) { /* Loop on individuals: reads the data file to get the maximum value of the modality of this covariate Vj*/
1.339 brouard 6290: /* printf("Waiting for error tricode Tvar[%d]=%d i=%d (int)(covar[Tvar[k]][i]=%d\n",k,Tvar[k], i, (int)(covar[Tvar[k]][i])); */
1.242 brouard 6291: ij=(int)(covar[Tvar[k]][i]);
6292: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
6293: * If product of Vn*Vm, still boolean *:
6294: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
6295: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
6296: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
6297: modality of the nth covariate of individual i. */
6298: if (ij > modmaxcovj)
6299: modmaxcovj=ij;
6300: else if (ij < modmincovj)
6301: modmincovj=ij;
1.287 brouard 6302: if (ij <0 || ij >1 ){
1.311 brouard 6303: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6304: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6305: fflush(ficlog);
6306: exit(1);
1.287 brouard 6307: }
6308: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 6309: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
6310: exit(1);
6311: }else
6312: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
6313: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
6314: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
6315: /* getting the maximum value of the modality of the covariate
6316: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
6317: female ies 1, then modmaxcovj=1.
6318: */
6319: } /* end for loop on individuals i */
6320: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6321: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6322: cptcode=modmaxcovj;
6323: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
6324: /*for (i=0; i<=cptcode; i++) {*/
6325: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
6326: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6327: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6328: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
6329: if( j != -1){
6330: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
6331: covariate for which somebody answered excluding
6332: undefined. Usually 2: 0 and 1. */
6333: }
6334: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
6335: covariate for which somebody answered including
6336: undefined. Usually 3: -1, 0 and 1. */
6337: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
6338: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
6339: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 6340:
1.242 brouard 6341: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
6342: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
6343: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
6344: /* modmincovj=3; modmaxcovj = 7; */
6345: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
6346: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
6347: /* defining two dummy variables: variables V1_1 and V1_2.*/
6348: /* nbcode[Tvar[j]][ij]=k; */
6349: /* nbcode[Tvar[j]][1]=0; */
6350: /* nbcode[Tvar[j]][2]=1; */
6351: /* nbcode[Tvar[j]][3]=2; */
6352: /* To be continued (not working yet). */
6353: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 6354:
6355: /* for (i=modmincovj; i<=modmaxcovj; i++) { */ /* i= 1 to 2 for dichotomous, or from 1 to 3 or from -1 or 0 to 1 currently*/
6356: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
6357: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
6358: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
6359: /*, could be restored in the future */
6360: for (i=0; i<=1; i++) { /* i= 1 to 2 for dichotomous, or from 1 to 3 or from -1 or 0 to 1 currently*/
1.242 brouard 6361: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
6362: break;
6363: }
6364: ij++;
1.287 brouard 6365: nbcode[Tvar[k]][ij]=i; /* stores the original value of modality i in an array nbcode, ij modality from 1 to last non-nul modality. nbcode[1][1]=0 nbcode[1][2]=1 . Could be -1*/
1.242 brouard 6366: cptcode = ij; /* New max modality for covar j */
6367: } /* end of loop on modality i=-1 to 1 or more */
6368: break;
6369: case 1: /* Testing on varying covariate, could be simple and
6370: * should look at waves or product of fixed *
6371: * varying. No time to test -1, assuming 0 and 1 only */
6372: ij=0;
6373: for(i=0; i<=1;i++){
6374: nbcode[Tvar[k]][++ij]=i;
6375: }
6376: break;
6377: default:
6378: break;
6379: } /* end switch */
6380: } /* end dummy test */
1.342 brouard 6381: if(Dummy[k]==1 && Typevar[k] !=1 && Fixed ==0){ /* Fixed Quantitative covariate and not age product */
1.311 brouard 6382: for (i=1; i<=imx; i++) { /* Loop on individuals: reads the data file to get the maximum value of the modality of this covariate Vj*/
1.335 brouard 6383: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
6384: printf("Error k=%d \n",k);
6385: exit(1);
6386: }
1.311 brouard 6387: if(isnan(covar[Tvar[k]][i])){
6388: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6389: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6390: fflush(ficlog);
6391: exit(1);
6392: }
6393: }
1.335 brouard 6394: } /* end Quanti */
1.287 brouard 6395: } /* end of loop on model-covariate k. nbcode[Tvark][1]=-1, nbcode[Tvark][1]=0 and nbcode[Tvark][2]=1 sets the value of covariate k*/
1.242 brouard 6396:
6397: for (k=-1; k< maxncov; k++) Ndum[k]=0;
6398: /* Look at fixed dummy (single or product) covariates to check empty modalities */
6399: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
6400: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
6401: ij=Tvar[i]; /* Tvar 5,4,3,6,5,7,1,4 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V4*age */
6402: Ndum[ij]++; /* Count the # of 1, 2 etc: {1,1,1,2,2,1,1} because V1 once, V2 once, two V4 and V5 in above */
6403: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
6404: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
6405:
6406: ij=0;
6407: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
1.335 brouard 6408: for (k=1; k<= cptcovt; k++) { /* cptcovt: total number of covariates of the model (2) nbocc(+)+1 = 8 excepting constant and age and age*age */
6409: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 6410: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
6411: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 brouard 6412: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
6413: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
6414: /* Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product*/
1.242 brouard 6415: /* If product not in single variable we don't print results */
6416: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 brouard 6417: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
6418: /* k= 1 2 3 4 5 6 7 8 9 */
6419: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
6420: /* ij 1 2 3 */
6421: /* Tvaraff[ij]= 4 3 1 */
6422: /* Tmodelind[ij]=2 3 9 */
6423: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 6424: Tvaraff[ij]=Tvar[k]; /* For printing combination *//* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, Tvar {5, 4, 3, 6, 5, 2, 7, 1, 1} Tvaraff={4, 3, 1} V4, V3, V1*/
6425: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
6426: TmodelInvind[ij]=Tvar[k]- ncovcol-nqv; /* Inverse TmodelInvind[2=V4]=2 second dummy varying cov (V4)4-1-1 {0, 2, 1, } TmodelInvind[3]=1 */
6427: if(Fixed[k]!=0)
6428: anyvaryingduminmodel=1;
6429: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
6430: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
6431: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
6432: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
6433: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
6434: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
6435: }
6436: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
6437: /* ij--; */
6438: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 brouard 6439: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 6440: * because they can be excluded from the model and real
6441: * if in the model but excluded because missing values, but how to get k from ij?*/
6442: for(j=ij+1; j<= cptcovt; j++){
6443: Tvaraff[j]=0;
6444: Tmodelind[j]=0;
6445: }
6446: for(j=ntveff+1; j<= cptcovt; j++){
6447: TmodelInvind[j]=0;
6448: }
6449: /* To be sorted */
6450: ;
6451: }
1.126 brouard 6452:
1.145 brouard 6453:
1.126 brouard 6454: /*********** Health Expectancies ****************/
6455:
1.235 brouard 6456: void evsij(double ***eij, double x[], int nlstate, int stepm, int bage, int fage, double **oldm, double **savm, int cij, int estepm,char strstart[], int nres )
1.126 brouard 6457:
6458: {
6459: /* Health expectancies, no variances */
1.329 brouard 6460: /* cij is the combination in the list of combination of dummy covariates */
6461: /* strstart is a string of time at start of computing */
1.164 brouard 6462: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 6463: int nhstepma, nstepma; /* Decreasing with age */
6464: double age, agelim, hf;
6465: double ***p3mat;
6466: double eip;
6467:
1.238 brouard 6468: /* pstamp(ficreseij); */
1.126 brouard 6469: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
6470: fprintf(ficreseij,"# Age");
6471: for(i=1; i<=nlstate;i++){
6472: for(j=1; j<=nlstate;j++){
6473: fprintf(ficreseij," e%1d%1d ",i,j);
6474: }
6475: fprintf(ficreseij," e%1d. ",i);
6476: }
6477: fprintf(ficreseij,"\n");
6478:
6479:
6480: if(estepm < stepm){
6481: printf ("Problem %d lower than %d\n",estepm, stepm);
6482: }
6483: else hstepm=estepm;
6484: /* We compute the life expectancy from trapezoids spaced every estepm months
6485: * This is mainly to measure the difference between two models: for example
6486: * if stepm=24 months pijx are given only every 2 years and by summing them
6487: * we are calculating an estimate of the Life Expectancy assuming a linear
6488: * progression in between and thus overestimating or underestimating according
6489: * to the curvature of the survival function. If, for the same date, we
6490: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6491: * to compare the new estimate of Life expectancy with the same linear
6492: * hypothesis. A more precise result, taking into account a more precise
6493: * curvature will be obtained if estepm is as small as stepm. */
6494:
6495: /* For example we decided to compute the life expectancy with the smallest unit */
6496: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6497: nhstepm is the number of hstepm from age to agelim
6498: nstepm is the number of stepm from age to agelin.
1.270 brouard 6499: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6500: and note for a fixed period like estepm months */
6501: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6502: survival function given by stepm (the optimization length). Unfortunately it
6503: means that if the survival funtion is printed only each two years of age and if
6504: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6505: results. So we changed our mind and took the option of the best precision.
6506: */
6507: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6508:
6509: agelim=AGESUP;
6510: /* If stepm=6 months */
6511: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6512: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6513:
6514: /* nhstepm age range expressed in number of stepm */
6515: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6516: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6517: /* if (stepm >= YEARM) hstepm=1;*/
6518: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6519: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6520:
6521: for (age=bage; age<=fage; age ++){
6522: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6523: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6524: /* if (stepm >= YEARM) hstepm=1;*/
6525: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6526:
6527: /* If stepm=6 months */
6528: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6529: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 6530: /* printf("HELLO evsij Entering hpxij age=%d cij=%d hstepm=%d x[1]=%f nres=%d\n",(int) age, cij, hstepm, x[1], nres); */
1.235 brouard 6531: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6532:
6533: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6534:
6535: printf("%d|",(int)age);fflush(stdout);
6536: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6537:
6538: /* Computing expectancies */
6539: for(i=1; i<=nlstate;i++)
6540: for(j=1; j<=nlstate;j++)
6541: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6542: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6543:
6544: /* if((int)age==70)printf("i=%2d,j=%2d,h=%2d,age=%3d,%9.4f,%9.4f,%9.4f\n",i,j,h,(int)age,p3mat[i][j][h],hf,eij[i][j][(int)age]);*/
6545:
6546: }
6547:
6548: fprintf(ficreseij,"%3.0f",age );
6549: for(i=1; i<=nlstate;i++){
6550: eip=0;
6551: for(j=1; j<=nlstate;j++){
6552: eip +=eij[i][j][(int)age];
6553: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6554: }
6555: fprintf(ficreseij,"%9.4f", eip );
6556: }
6557: fprintf(ficreseij,"\n");
6558:
6559: }
6560: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6561: printf("\n");
6562: fprintf(ficlog,"\n");
6563:
6564: }
6565:
1.235 brouard 6566: void cvevsij(double ***eij, double x[], int nlstate, int stepm, int bage, int fage, double **oldm, double **savm, int cij, int estepm,double delti[],double **matcov,char strstart[], int nres )
1.126 brouard 6567:
6568: {
6569: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6570: to initial status i, ei. .
1.126 brouard 6571: */
1.336 brouard 6572: /* Very time consuming function, but already optimized with precov */
1.126 brouard 6573: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6574: int nhstepma, nstepma; /* Decreasing with age */
6575: double age, agelim, hf;
6576: double ***p3matp, ***p3matm, ***varhe;
6577: double **dnewm,**doldm;
6578: double *xp, *xm;
6579: double **gp, **gm;
6580: double ***gradg, ***trgradg;
6581: int theta;
6582:
6583: double eip, vip;
6584:
6585: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6586: xp=vector(1,npar);
6587: xm=vector(1,npar);
6588: dnewm=matrix(1,nlstate*nlstate,1,npar);
6589: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6590:
6591: pstamp(ficresstdeij);
6592: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6593: fprintf(ficresstdeij,"# Age");
6594: for(i=1; i<=nlstate;i++){
6595: for(j=1; j<=nlstate;j++)
6596: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6597: fprintf(ficresstdeij," e%1d. ",i);
6598: }
6599: fprintf(ficresstdeij,"\n");
6600:
6601: pstamp(ficrescveij);
6602: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6603: fprintf(ficrescveij,"# Age");
6604: for(i=1; i<=nlstate;i++)
6605: for(j=1; j<=nlstate;j++){
6606: cptj= (j-1)*nlstate+i;
6607: for(i2=1; i2<=nlstate;i2++)
6608: for(j2=1; j2<=nlstate;j2++){
6609: cptj2= (j2-1)*nlstate+i2;
6610: if(cptj2 <= cptj)
6611: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6612: }
6613: }
6614: fprintf(ficrescveij,"\n");
6615:
6616: if(estepm < stepm){
6617: printf ("Problem %d lower than %d\n",estepm, stepm);
6618: }
6619: else hstepm=estepm;
6620: /* We compute the life expectancy from trapezoids spaced every estepm months
6621: * This is mainly to measure the difference between two models: for example
6622: * if stepm=24 months pijx are given only every 2 years and by summing them
6623: * we are calculating an estimate of the Life Expectancy assuming a linear
6624: * progression in between and thus overestimating or underestimating according
6625: * to the curvature of the survival function. If, for the same date, we
6626: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6627: * to compare the new estimate of Life expectancy with the same linear
6628: * hypothesis. A more precise result, taking into account a more precise
6629: * curvature will be obtained if estepm is as small as stepm. */
6630:
6631: /* For example we decided to compute the life expectancy with the smallest unit */
6632: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6633: nhstepm is the number of hstepm from age to agelim
6634: nstepm is the number of stepm from age to agelin.
6635: Look at hpijx to understand the reason of that which relies in memory size
6636: and note for a fixed period like estepm months */
6637: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6638: survival function given by stepm (the optimization length). Unfortunately it
6639: means that if the survival funtion is printed only each two years of age and if
6640: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6641: results. So we changed our mind and took the option of the best precision.
6642: */
6643: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6644:
6645: /* If stepm=6 months */
6646: /* nhstepm age range expressed in number of stepm */
6647: agelim=AGESUP;
6648: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6649: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6650: /* if (stepm >= YEARM) hstepm=1;*/
6651: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6652:
6653: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6654: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6655: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6656: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6657: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6658: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6659:
6660: for (age=bage; age<=fage; age ++){
6661: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6662: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6663: /* if (stepm >= YEARM) hstepm=1;*/
6664: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6665:
1.126 brouard 6666: /* If stepm=6 months */
6667: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6668: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6669:
6670: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6671:
1.126 brouard 6672: /* Computing Variances of health expectancies */
6673: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6674: decrease memory allocation */
6675: for(theta=1; theta <=npar; theta++){
6676: for(i=1; i<=npar; i++){
1.222 brouard 6677: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6678: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6679: }
1.235 brouard 6680: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6681: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6682:
1.126 brouard 6683: for(j=1; j<= nlstate; j++){
1.222 brouard 6684: for(i=1; i<=nlstate; i++){
6685: for(h=0; h<=nhstepm-1; h++){
6686: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6687: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6688: }
6689: }
1.126 brouard 6690: }
1.218 brouard 6691:
1.126 brouard 6692: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6693: for(h=0; h<=nhstepm-1; h++){
6694: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6695: }
1.126 brouard 6696: }/* End theta */
6697:
6698:
6699: for(h=0; h<=nhstepm-1; h++)
6700: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6701: for(theta=1; theta <=npar; theta++)
6702: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6703:
1.218 brouard 6704:
1.222 brouard 6705: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6706: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6707: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6708:
1.222 brouard 6709: printf("%d|",(int)age);fflush(stdout);
6710: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6711: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6712: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6713: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6714: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6715: for(ij=1;ij<=nlstate*nlstate;ij++)
6716: for(ji=1;ji<=nlstate*nlstate;ji++)
6717: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6718: }
6719: }
1.320 brouard 6720: /* if((int)age ==50){ */
6721: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6722: /* } */
1.126 brouard 6723: /* Computing expectancies */
1.235 brouard 6724: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6725: for(i=1; i<=nlstate;i++)
6726: for(j=1; j<=nlstate;j++)
1.222 brouard 6727: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6728: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6729:
1.222 brouard 6730: /* if((int)age==70)printf("i=%2d,j=%2d,h=%2d,age=%3d,%9.4f,%9.4f,%9.4f\n",i,j,h,(int)age,p3mat[i][j][h],hf,eij[i][j][(int)age]);*/
1.218 brouard 6731:
1.222 brouard 6732: }
1.269 brouard 6733:
6734: /* Standard deviation of expectancies ij */
1.126 brouard 6735: fprintf(ficresstdeij,"%3.0f",age );
6736: for(i=1; i<=nlstate;i++){
6737: eip=0.;
6738: vip=0.;
6739: for(j=1; j<=nlstate;j++){
1.222 brouard 6740: eip += eij[i][j][(int)age];
6741: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6742: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6743: fprintf(ficresstdeij," %9.4f (%.4f)", eij[i][j][(int)age], sqrt(varhe[(j-1)*nlstate+i][(j-1)*nlstate+i][(int)age]) );
1.126 brouard 6744: }
6745: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6746: }
6747: fprintf(ficresstdeij,"\n");
1.218 brouard 6748:
1.269 brouard 6749: /* Variance of expectancies ij */
1.126 brouard 6750: fprintf(ficrescveij,"%3.0f",age );
6751: for(i=1; i<=nlstate;i++)
6752: for(j=1; j<=nlstate;j++){
1.222 brouard 6753: cptj= (j-1)*nlstate+i;
6754: for(i2=1; i2<=nlstate;i2++)
6755: for(j2=1; j2<=nlstate;j2++){
6756: cptj2= (j2-1)*nlstate+i2;
6757: if(cptj2 <= cptj)
6758: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6759: }
1.126 brouard 6760: }
6761: fprintf(ficrescveij,"\n");
1.218 brouard 6762:
1.126 brouard 6763: }
6764: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6765: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6766: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6767: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6768: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6769: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6770: printf("\n");
6771: fprintf(ficlog,"\n");
1.218 brouard 6772:
1.126 brouard 6773: free_vector(xm,1,npar);
6774: free_vector(xp,1,npar);
6775: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6776: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6777: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6778: }
1.218 brouard 6779:
1.126 brouard 6780: /************ Variance ******************/
1.235 brouard 6781: void varevsij(char optionfilefiname[], double ***vareij, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **prlim, double ftolpl, int *ncvyearp, int ij, int estepm, int cptcov, int cptcod, int popbased, int mobilav, char strstart[], int nres)
1.218 brouard 6782: {
1.279 brouard 6783: /** Variance of health expectancies
6784: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6785: * double **newm;
6786: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6787: */
1.218 brouard 6788:
6789: /* int movingaverage(); */
6790: double **dnewm,**doldm;
6791: double **dnewmp,**doldmp;
6792: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6793: int first=0;
1.218 brouard 6794: int k;
6795: double *xp;
1.279 brouard 6796: double **gp, **gm; /**< for var eij */
6797: double ***gradg, ***trgradg; /**< for var eij */
6798: double **gradgp, **trgradgp; /**< for var p point j */
6799: double *gpp, *gmp; /**< for var p point j */
6800: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6801: double ***p3mat;
6802: double age,agelim, hf;
6803: /* double ***mobaverage; */
6804: int theta;
6805: char digit[4];
6806: char digitp[25];
6807:
6808: char fileresprobmorprev[FILENAMELENGTH];
6809:
6810: if(popbased==1){
6811: if(mobilav!=0)
6812: strcpy(digitp,"-POPULBASED-MOBILAV_");
6813: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6814: }
6815: else
6816: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6817:
1.218 brouard 6818: /* if (mobilav!=0) { */
6819: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6820: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6821: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6822: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6823: /* } */
6824: /* } */
6825:
6826: strcpy(fileresprobmorprev,"PRMORPREV-");
6827: sprintf(digit,"%-d",ij);
6828: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6829: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6830: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6831: strcat(fileresprobmorprev,fileresu);
6832: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6833: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6834: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6835: }
6836: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6837: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6838: pstamp(ficresprobmorprev);
6839: fprintf(ficresprobmorprev,"# probabilities of dying before estepm=%d months for people of exact age and weighted probabilities w1*p1j+w2*p2j+... stand dev in()\n",estepm);
1.238 brouard 6840: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.337 brouard 6841:
6842: /* We use TinvDoQresult[nres][resultmodel[nres][j] we sort according to the equation model and the resultline: it is a choice */
6843: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ /\* To be done*\/ */
6844: /* fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
6845: /* } */
6846: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */ /* To be done*/
1.344 brouard 6847: /* fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); */
1.337 brouard 6848: fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 6849: }
1.337 brouard 6850: /* for(j=1;j<=cptcoveff;j++) */
6851: /* fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]); */
1.238 brouard 6852: fprintf(ficresprobmorprev,"\n");
6853:
1.218 brouard 6854: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6855: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6856: fprintf(ficresprobmorprev," p.%-d SE",j);
6857: for(i=1; i<=nlstate;i++)
6858: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6859: }
6860: fprintf(ficresprobmorprev,"\n");
6861:
6862: fprintf(ficgp,"\n# Routine varevsij");
6863: fprintf(ficgp,"\nunset title \n");
6864: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6865: fprintf(fichtm,"\n<li><h4> Computing probabilities of dying over estepm months as a weighted average (i.e global mortality independent of initial healh state)</h4></li>\n");
6866: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6867:
1.218 brouard 6868: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6869: pstamp(ficresvij);
6870: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6871: if(popbased==1)
6872: fprintf(ficresvij,"the age specific prevalence observed (cross-sectionally) in the population i.e cross-sectionally\n in each health state (popbased=1) (mobilav=%d\n",mobilav);
6873: else
6874: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6875: fprintf(ficresvij,"# Age");
6876: for(i=1; i<=nlstate;i++)
6877: for(j=1; j<=nlstate;j++)
6878: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6879: fprintf(ficresvij,"\n");
6880:
6881: xp=vector(1,npar);
6882: dnewm=matrix(1,nlstate,1,npar);
6883: doldm=matrix(1,nlstate,1,nlstate);
6884: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6885: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6886:
6887: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6888: gpp=vector(nlstate+1,nlstate+ndeath);
6889: gmp=vector(nlstate+1,nlstate+ndeath);
6890: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6891:
1.218 brouard 6892: if(estepm < stepm){
6893: printf ("Problem %d lower than %d\n",estepm, stepm);
6894: }
6895: else hstepm=estepm;
6896: /* For example we decided to compute the life expectancy with the smallest unit */
6897: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6898: nhstepm is the number of hstepm from age to agelim
6899: nstepm is the number of stepm from age to agelim.
6900: Look at function hpijx to understand why because of memory size limitations,
6901: we decided (b) to get a life expectancy respecting the most precise curvature of the
6902: survival function given by stepm (the optimization length). Unfortunately it
6903: means that if the survival funtion is printed every two years of age and if
6904: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6905: results. So we changed our mind and took the option of the best precision.
6906: */
6907: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6908: agelim = AGESUP;
6909: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6910: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6911: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6912: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6913: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6914: gp=matrix(0,nhstepm,1,nlstate);
6915: gm=matrix(0,nhstepm,1,nlstate);
6916:
6917:
6918: for(theta=1; theta <=npar; theta++){
6919: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6920: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6921: }
1.279 brouard 6922: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6923: * returns into prlim .
1.288 brouard 6924: */
1.242 brouard 6925: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6926:
6927: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6928: if (popbased==1) {
6929: if(mobilav ==0){
6930: for(i=1; i<=nlstate;i++)
6931: prlim[i][i]=probs[(int)age][i][ij];
6932: }else{ /* mobilav */
6933: for(i=1; i<=nlstate;i++)
6934: prlim[i][i]=mobaverage[(int)age][i][ij];
6935: }
6936: }
1.295 brouard 6937: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6938: */
6939: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres); /* Returns p3mat[i][j][h] for h=0 to nhstepm */
1.292 brouard 6940: /**< And for each alive state j, sums over i \f$ w^i_x {}{h}_p^{ij}x\f$, which are the probability
1.279 brouard 6941: * at horizon h in state j including mortality.
6942: */
1.218 brouard 6943: for(j=1; j<= nlstate; j++){
6944: for(h=0; h<=nhstepm; h++){
6945: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6946: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6947: }
6948: }
1.279 brouard 6949: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6950: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6951: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6952: */
6953: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6954: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6955: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6956: }
6957:
6958: /* Again with minus shift */
1.218 brouard 6959:
6960: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6961: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6962:
1.242 brouard 6963: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6964:
6965: if (popbased==1) {
6966: if(mobilav ==0){
6967: for(i=1; i<=nlstate;i++)
6968: prlim[i][i]=probs[(int)age][i][ij];
6969: }else{ /* mobilav */
6970: for(i=1; i<=nlstate;i++)
6971: prlim[i][i]=mobaverage[(int)age][i][ij];
6972: }
6973: }
6974:
1.235 brouard 6975: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6976:
6977: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6978: for(h=0; h<=nhstepm; h++){
6979: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6980: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6981: }
6982: }
6983: /* This for computing probability of death (h=1 means
6984: computed over hstepm matrices product = hstepm*stepm months)
6985: as a weighted average of prlim.
6986: */
6987: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6988: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6989: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6990: }
1.279 brouard 6991: /* end shifting computations */
6992:
6993: /**< Computing gradient matrix at horizon h
6994: */
1.218 brouard 6995: for(j=1; j<= nlstate; j++) /* vareij */
6996: for(h=0; h<=nhstepm; h++){
6997: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6998: }
1.279 brouard 6999: /**< Gradient of overall mortality p.3 (or p.j)
7000: */
7001: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 7002: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
7003: }
7004:
7005: } /* End theta */
1.279 brouard 7006:
7007: /* We got the gradient matrix for each theta and state j */
1.218 brouard 7008: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
7009:
7010: for(h=0; h<=nhstepm; h++) /* veij */
7011: for(j=1; j<=nlstate;j++)
7012: for(theta=1; theta <=npar; theta++)
7013: trgradg[h][j][theta]=gradg[h][theta][j];
7014:
7015: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
7016: for(theta=1; theta <=npar; theta++)
7017: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 7018: /**< as well as its transposed matrix
7019: */
1.218 brouard 7020:
7021: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
7022: for(i=1;i<=nlstate;i++)
7023: for(j=1;j<=nlstate;j++)
7024: vareij[i][j][(int)age] =0.;
1.279 brouard 7025:
7026: /* Computing trgradg by matcov by gradg at age and summing over h
7027: * and k (nhstepm) formula 15 of article
7028: * Lievre-Brouard-Heathcote
7029: */
7030:
1.218 brouard 7031: for(h=0;h<=nhstepm;h++){
7032: for(k=0;k<=nhstepm;k++){
7033: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
7034: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
7035: for(i=1;i<=nlstate;i++)
7036: for(j=1;j<=nlstate;j++)
7037: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
7038: }
7039: }
7040:
1.279 brouard 7041: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
7042: * p.j overall mortality formula 49 but computed directly because
7043: * we compute the grad (wix pijx) instead of grad (pijx),even if
7044: * wix is independent of theta.
7045: */
1.218 brouard 7046: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
7047: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
7048: for(j=nlstate+1;j<=nlstate+ndeath;j++)
7049: for(i=nlstate+1;i<=nlstate+ndeath;i++)
7050: varppt[j][i]=doldmp[j][i];
7051: /* end ppptj */
7052: /* x centered again */
7053:
1.242 brouard 7054: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 7055:
7056: if (popbased==1) {
7057: if(mobilav ==0){
7058: for(i=1; i<=nlstate;i++)
7059: prlim[i][i]=probs[(int)age][i][ij];
7060: }else{ /* mobilav */
7061: for(i=1; i<=nlstate;i++)
7062: prlim[i][i]=mobaverage[(int)age][i][ij];
7063: }
7064: }
7065:
7066: /* This for computing probability of death (h=1 means
7067: computed over hstepm (estepm) matrices product = hstepm*stepm months)
7068: as a weighted average of prlim.
7069: */
1.235 brouard 7070: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 7071: for(j=nlstate+1;j<=nlstate+ndeath;j++){
7072: for(i=1,gmp[j]=0.;i<= nlstate; i++)
7073: gmp[j] += prlim[i][i]*p3mat[i][j][1];
7074: }
7075: /* end probability of death */
7076:
7077: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
7078: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
7079: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
7080: for(i=1; i<=nlstate;i++){
7081: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
7082: }
7083: }
7084: fprintf(ficresprobmorprev,"\n");
7085:
7086: fprintf(ficresvij,"%.0f ",age );
7087: for(i=1; i<=nlstate;i++)
7088: for(j=1; j<=nlstate;j++){
7089: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
7090: }
7091: fprintf(ficresvij,"\n");
7092: free_matrix(gp,0,nhstepm,1,nlstate);
7093: free_matrix(gm,0,nhstepm,1,nlstate);
7094: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
7095: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
7096: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7097: } /* End age */
7098: free_vector(gpp,nlstate+1,nlstate+ndeath);
7099: free_vector(gmp,nlstate+1,nlstate+ndeath);
7100: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
7101: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
7102: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
7103: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
7104: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
7105: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
7106: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
7107: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
7108: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
7109: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
7110: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
7111: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
7112: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
7113: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
7114: fprintf(fichtm,"\n<br> Probability is computed over estepm=%d months. <br> <img src=\"%s%s.svg\"> <br>\n", estepm,subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
7115: /* fprintf(fichtm,"\n<br> Probability is computed over estepm=%d months and then divided by estepm and multiplied by %.0f in order to have the probability to die over a year <br> <img src=\"varmuptjgr%s%s.svg\"> <br>\n", stepm,YEARM,digitp,digit);
1.126 brouard 7116: */
1.218 brouard 7117: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
7118: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 7119:
1.218 brouard 7120: free_vector(xp,1,npar);
7121: free_matrix(doldm,1,nlstate,1,nlstate);
7122: free_matrix(dnewm,1,nlstate,1,npar);
7123: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7124: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
7125: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7126: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7127: fclose(ficresprobmorprev);
7128: fflush(ficgp);
7129: fflush(fichtm);
7130: } /* end varevsij */
1.126 brouard 7131:
7132: /************ Variance of prevlim ******************/
1.269 brouard 7133: void varprevlim(char fileresvpl[], FILE *ficresvpl, double **varpl, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **prlim, double ftolpl, int *ncvyearp, int ij, char strstart[], int nres)
1.126 brouard 7134: {
1.205 brouard 7135: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 7136: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 7137:
1.268 brouard 7138: double **dnewmpar,**doldm;
1.126 brouard 7139: int i, j, nhstepm, hstepm;
7140: double *xp;
7141: double *gp, *gm;
7142: double **gradg, **trgradg;
1.208 brouard 7143: double **mgm, **mgp;
1.126 brouard 7144: double age,agelim;
7145: int theta;
7146:
7147: pstamp(ficresvpl);
1.288 brouard 7148: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 7149: fprintf(ficresvpl,"# Age ");
7150: if(nresult >=1)
7151: fprintf(ficresvpl," Result# ");
1.126 brouard 7152: for(i=1; i<=nlstate;i++)
7153: fprintf(ficresvpl," %1d-%1d",i,i);
7154: fprintf(ficresvpl,"\n");
7155:
7156: xp=vector(1,npar);
1.268 brouard 7157: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 7158: doldm=matrix(1,nlstate,1,nlstate);
7159:
7160: hstepm=1*YEARM; /* Every year of age */
7161: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7162: agelim = AGESUP;
7163: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
7164: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7165: if (stepm >= YEARM) hstepm=1;
7166: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7167: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 7168: mgp=matrix(1,npar,1,nlstate);
7169: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 7170: gp=vector(1,nlstate);
7171: gm=vector(1,nlstate);
7172:
7173: for(theta=1; theta <=npar; theta++){
7174: for(i=1; i<=npar; i++){ /* Computes gradient */
7175: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7176: }
1.288 brouard 7177: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7178: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7179: /* else */
7180: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7181: for(i=1;i<=nlstate;i++){
1.126 brouard 7182: gp[i] = prlim[i][i];
1.208 brouard 7183: mgp[theta][i] = prlim[i][i];
7184: }
1.126 brouard 7185: for(i=1; i<=npar; i++) /* Computes gradient */
7186: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 7187: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7188: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7189: /* else */
7190: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7191: for(i=1;i<=nlstate;i++){
1.126 brouard 7192: gm[i] = prlim[i][i];
1.208 brouard 7193: mgm[theta][i] = prlim[i][i];
7194: }
1.126 brouard 7195: for(i=1;i<=nlstate;i++)
7196: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 7197: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 7198: } /* End theta */
7199:
7200: trgradg =matrix(1,nlstate,1,npar);
7201:
7202: for(j=1; j<=nlstate;j++)
7203: for(theta=1; theta <=npar; theta++)
7204: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 7205: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7206: /* printf("\nmgm mgp %d ",(int)age); */
7207: /* for(j=1; j<=nlstate;j++){ */
7208: /* printf(" %d ",j); */
7209: /* for(theta=1; theta <=npar; theta++) */
7210: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7211: /* printf("\n "); */
7212: /* } */
7213: /* } */
7214: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7215: /* printf("\n gradg %d ",(int)age); */
7216: /* for(j=1; j<=nlstate;j++){ */
7217: /* printf("%d ",j); */
7218: /* for(theta=1; theta <=npar; theta++) */
7219: /* printf("%d %lf ",theta,gradg[theta][j]); */
7220: /* printf("\n "); */
7221: /* } */
7222: /* } */
1.126 brouard 7223:
7224: for(i=1;i<=nlstate;i++)
7225: varpl[i][(int)age] =0.;
1.209 brouard 7226: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 7227: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7228: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7229: }else{
1.268 brouard 7230: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7231: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7232: }
1.126 brouard 7233: for(i=1;i<=nlstate;i++)
7234: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7235:
7236: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 7237: if(nresult >=1)
7238: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 7239: for(i=1; i<=nlstate;i++){
1.126 brouard 7240: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 7241: /* for(j=1;j<=nlstate;j++) */
7242: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
7243: }
1.126 brouard 7244: fprintf(ficresvpl,"\n");
7245: free_vector(gp,1,nlstate);
7246: free_vector(gm,1,nlstate);
1.208 brouard 7247: free_matrix(mgm,1,npar,1,nlstate);
7248: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 7249: free_matrix(gradg,1,npar,1,nlstate);
7250: free_matrix(trgradg,1,nlstate,1,npar);
7251: } /* End age */
7252:
7253: free_vector(xp,1,npar);
7254: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 7255: free_matrix(dnewmpar,1,nlstate,1,nlstate);
7256:
7257: }
7258:
7259:
7260: /************ Variance of backprevalence limit ******************/
1.269 brouard 7261: void varbrevlim(char fileresvbl[], FILE *ficresvbl, double **varbpl, double **matcov, double x[], double delti[], int nlstate, int stepm, double bage, double fage, double **oldm, double **savm, double **bprlim, double ftolpl, int mobilavproj, int *ncvyearp, int ij, char strstart[], int nres)
1.268 brouard 7262: {
7263: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
7264: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
7265:
7266: double **dnewmpar,**doldm;
7267: int i, j, nhstepm, hstepm;
7268: double *xp;
7269: double *gp, *gm;
7270: double **gradg, **trgradg;
7271: double **mgm, **mgp;
7272: double age,agelim;
7273: int theta;
7274:
7275: pstamp(ficresvbl);
7276: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
7277: fprintf(ficresvbl,"# Age ");
7278: if(nresult >=1)
7279: fprintf(ficresvbl," Result# ");
7280: for(i=1; i<=nlstate;i++)
7281: fprintf(ficresvbl," %1d-%1d",i,i);
7282: fprintf(ficresvbl,"\n");
7283:
7284: xp=vector(1,npar);
7285: dnewmpar=matrix(1,nlstate,1,npar);
7286: doldm=matrix(1,nlstate,1,nlstate);
7287:
7288: hstepm=1*YEARM; /* Every year of age */
7289: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7290: agelim = AGEINF;
7291: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
7292: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7293: if (stepm >= YEARM) hstepm=1;
7294: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7295: gradg=matrix(1,npar,1,nlstate);
7296: mgp=matrix(1,npar,1,nlstate);
7297: mgm=matrix(1,npar,1,nlstate);
7298: gp=vector(1,nlstate);
7299: gm=vector(1,nlstate);
7300:
7301: for(theta=1; theta <=npar; theta++){
7302: for(i=1; i<=npar; i++){ /* Computes gradient */
7303: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7304: }
7305: if(mobilavproj > 0 )
7306: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7307: else
7308: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7309: for(i=1;i<=nlstate;i++){
7310: gp[i] = bprlim[i][i];
7311: mgp[theta][i] = bprlim[i][i];
7312: }
7313: for(i=1; i<=npar; i++) /* Computes gradient */
7314: xp[i] = x[i] - (i==theta ?delti[theta]:0);
7315: if(mobilavproj > 0 )
7316: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7317: else
7318: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7319: for(i=1;i<=nlstate;i++){
7320: gm[i] = bprlim[i][i];
7321: mgm[theta][i] = bprlim[i][i];
7322: }
7323: for(i=1;i<=nlstate;i++)
7324: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
7325: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
7326: } /* End theta */
7327:
7328: trgradg =matrix(1,nlstate,1,npar);
7329:
7330: for(j=1; j<=nlstate;j++)
7331: for(theta=1; theta <=npar; theta++)
7332: trgradg[j][theta]=gradg[theta][j];
7333: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7334: /* printf("\nmgm mgp %d ",(int)age); */
7335: /* for(j=1; j<=nlstate;j++){ */
7336: /* printf(" %d ",j); */
7337: /* for(theta=1; theta <=npar; theta++) */
7338: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7339: /* printf("\n "); */
7340: /* } */
7341: /* } */
7342: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7343: /* printf("\n gradg %d ",(int)age); */
7344: /* for(j=1; j<=nlstate;j++){ */
7345: /* printf("%d ",j); */
7346: /* for(theta=1; theta <=npar; theta++) */
7347: /* printf("%d %lf ",theta,gradg[theta][j]); */
7348: /* printf("\n "); */
7349: /* } */
7350: /* } */
7351:
7352: for(i=1;i<=nlstate;i++)
7353: varbpl[i][(int)age] =0.;
7354: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
7355: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7356: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7357: }else{
7358: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7359: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7360: }
7361: for(i=1;i<=nlstate;i++)
7362: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7363:
7364: fprintf(ficresvbl,"%.0f ",age );
7365: if(nresult >=1)
7366: fprintf(ficresvbl,"%d ",nres );
7367: for(i=1; i<=nlstate;i++)
7368: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
7369: fprintf(ficresvbl,"\n");
7370: free_vector(gp,1,nlstate);
7371: free_vector(gm,1,nlstate);
7372: free_matrix(mgm,1,npar,1,nlstate);
7373: free_matrix(mgp,1,npar,1,nlstate);
7374: free_matrix(gradg,1,npar,1,nlstate);
7375: free_matrix(trgradg,1,nlstate,1,npar);
7376: } /* End age */
7377:
7378: free_vector(xp,1,npar);
7379: free_matrix(doldm,1,nlstate,1,npar);
7380: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 7381:
7382: }
7383:
7384: /************ Variance of one-step probabilities ******************/
7385: void varprob(char optionfilefiname[], double **matcov, double x[], double delti[], int nlstate, double bage, double fage, int ij, int *Tvar, int **nbcode, int *ncodemax, char strstart[])
1.222 brouard 7386: {
7387: int i, j=0, k1, l1, tj;
7388: int k2, l2, j1, z1;
7389: int k=0, l;
7390: int first=1, first1, first2;
1.326 brouard 7391: int nres=0; /* New */
1.222 brouard 7392: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
7393: double **dnewm,**doldm;
7394: double *xp;
7395: double *gp, *gm;
7396: double **gradg, **trgradg;
7397: double **mu;
7398: double age, cov[NCOVMAX+1];
7399: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
7400: int theta;
7401: char fileresprob[FILENAMELENGTH];
7402: char fileresprobcov[FILENAMELENGTH];
7403: char fileresprobcor[FILENAMELENGTH];
7404: double ***varpij;
7405:
7406: strcpy(fileresprob,"PROB_");
7407: strcat(fileresprob,fileres);
7408: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
7409: printf("Problem with resultfile: %s\n", fileresprob);
7410: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
7411: }
7412: strcpy(fileresprobcov,"PROBCOV_");
7413: strcat(fileresprobcov,fileresu);
7414: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
7415: printf("Problem with resultfile: %s\n", fileresprobcov);
7416: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
7417: }
7418: strcpy(fileresprobcor,"PROBCOR_");
7419: strcat(fileresprobcor,fileresu);
7420: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
7421: printf("Problem with resultfile: %s\n", fileresprobcor);
7422: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
7423: }
7424: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7425: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7426: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7427: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7428: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7429: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7430: pstamp(ficresprob);
7431: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
7432: fprintf(ficresprob,"# Age");
7433: pstamp(ficresprobcov);
7434: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
7435: fprintf(ficresprobcov,"# Age");
7436: pstamp(ficresprobcor);
7437: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
7438: fprintf(ficresprobcor,"# Age");
1.126 brouard 7439:
7440:
1.222 brouard 7441: for(i=1; i<=nlstate;i++)
7442: for(j=1; j<=(nlstate+ndeath);j++){
7443: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
7444: fprintf(ficresprobcov," p%1d-%1d ",i,j);
7445: fprintf(ficresprobcor," p%1d-%1d ",i,j);
7446: }
7447: /* fprintf(ficresprob,"\n");
7448: fprintf(ficresprobcov,"\n");
7449: fprintf(ficresprobcor,"\n");
7450: */
7451: xp=vector(1,npar);
7452: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7453: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7454: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
7455: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
7456: first=1;
7457: fprintf(ficgp,"\n# Routine varprob");
7458: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
7459: fprintf(fichtm,"\n");
7460:
1.288 brouard 7461: fprintf(fichtm,"\n<li><h4> <a href=\"%s\">Matrix of variance-covariance of one-step probabilities (drawings)</a></h4> this page is important in order to visualize confidence intervals and especially correlation between disability and recovery, or more generally, way in and way back. File %s</li>\n",optionfilehtmcov,optionfilehtmcov);
1.222 brouard 7462: fprintf(fichtmcov,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>Matrix of variance-covariance of pairs of step probabilities</h4>\n",optionfilehtmcov, optionfilehtmcov);
7463: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 7464: and drawn. It helps understanding how is the covariance between two incidences.\
7465: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 7466: fprintf(fichtmcov,"\n<br> Contour plot corresponding to x'cov<sup>-1</sup>x = 4 (where x is the column vector (pij,pkl)) are drawn. \
1.126 brouard 7467: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
7468: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
7469: standard deviations wide on each axis. <br>\
7470: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
7471: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
7472: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
7473:
1.222 brouard 7474: cov[1]=1;
7475: /* tj=cptcoveff; */
1.225 brouard 7476: tj = (int) pow(2,cptcoveff);
1.222 brouard 7477: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
7478: j1=0;
1.332 brouard 7479:
7480: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
7481: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.342 brouard 7482: /* printf("Varprob TKresult[nres]=%d j1=%d, nres=%d, cptcovn=%d, cptcoveff=%d tj=%d cptcovs=%d\n", TKresult[nres], j1, nres, cptcovn, cptcoveff, tj, cptcovs); */
1.332 brouard 7483: if(tj != 1 && TKresult[nres]!= j1)
7484: continue;
7485:
7486: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
7487: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
7488: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 7489: if (cptcovn>0) {
1.334 brouard 7490: fprintf(ficresprob, "\n#********** Variable ");
7491: fprintf(ficresprobcov, "\n#********** Variable ");
7492: fprintf(ficgp, "\n#********** Variable ");
7493: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
7494: fprintf(ficresprobcor, "\n#********** Variable ");
7495:
7496: /* Including quantitative variables of the resultline to be done */
7497: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
1.343 brouard 7498: /* printf("Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model); */
1.338 brouard 7499: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model);
7500: /* fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s resultline[%d]=%s \n",nres, z1, modelresult[nres][z1], model, nres, resultline[nres]); */
1.334 brouard 7501: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
7502: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
7503: fprintf(ficresprob,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
7504: fprintf(ficresprobcov,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
7505: fprintf(ficgp,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
7506: fprintf(fichtmcov,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
7507: fprintf(ficresprobcor,"V%d=%d ",Tvresult[nres][z1],Tresult[nres][z1]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
7508: fprintf(ficresprob,"fixed ");
7509: fprintf(ficresprobcov,"fixed ");
7510: fprintf(ficgp,"fixed ");
7511: fprintf(fichtmcov,"fixed ");
7512: fprintf(ficresprobcor,"fixed ");
7513: }else{
7514: fprintf(ficresprob,"varyi ");
7515: fprintf(ficresprobcov,"varyi ");
7516: fprintf(ficgp,"varyi ");
7517: fprintf(fichtmcov,"varyi ");
7518: fprintf(ficresprobcor,"varyi ");
7519: }
7520: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
7521: /* For each selected (single) quantitative value */
1.337 brouard 7522: fprintf(ficresprob," V%d=%lg ",Tvqresult[nres][z1],Tqresult[nres][z1]);
1.334 brouard 7523: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
7524: fprintf(ficresprob,"fixed ");
7525: fprintf(ficresprobcov,"fixed ");
7526: fprintf(ficgp,"fixed ");
7527: fprintf(fichtmcov,"fixed ");
7528: fprintf(ficresprobcor,"fixed ");
7529: }else{
7530: fprintf(ficresprob,"varyi ");
7531: fprintf(ficresprobcov,"varyi ");
7532: fprintf(ficgp,"varyi ");
7533: fprintf(fichtmcov,"varyi ");
7534: fprintf(ficresprobcor,"varyi ");
7535: }
7536: }else{
7537: printf("Error in varprob() Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=V%d cptcovs=%d, cptcoveff=%d \n", nres, z1, Dummy[modelresult[nres][z1]],nres,z1,modelresult[nres][z1],cptcovs, cptcoveff); /* end if dummy or quanti */
7538: fprintf(ficlog,"Error in varprob() Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=V%d cptcovs=%d, cptcoveff=%d \n", nres, z1, Dummy[modelresult[nres][z1]],nres,z1,modelresult[nres][z1],cptcovs, cptcoveff); /* end if dummy or quanti */
7539: exit(1);
7540: }
7541: } /* End loop on variable of this resultline */
7542: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 7543: fprintf(ficresprob, "**********\n#\n");
7544: fprintf(ficresprobcov, "**********\n#\n");
7545: fprintf(ficgp, "**********\n#\n");
7546: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
7547: fprintf(ficresprobcor, "**********\n#");
7548: if(invalidvarcomb[j1]){
7549: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7550: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7551: continue;
7552: }
7553: }
7554: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7555: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7556: gp=vector(1,(nlstate)*(nlstate+ndeath));
7557: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 7558: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 7559: cov[2]=age;
7560: if(nagesqr==1)
7561: cov[3]= age*age;
1.334 brouard 7562: /* New code end of combination but for each resultline */
7563: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
7564: if(Typevar[k1]==1){ /* A product with age */
7565: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 7566: }else{
1.334 brouard 7567: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 7568: }
1.334 brouard 7569: }/* End of loop on model equation */
7570: /* Old code */
7571: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
7572: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
7573: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
7574: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
7575: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
7576: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
7577: /* * 1 1 1 1 1 */
7578: /* * 2 2 1 1 1 */
7579: /* * 3 1 2 1 1 */
7580: /* *\/ */
7581: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
7582: /* } */
7583: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
7584: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
7585: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
7586: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
7587: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
7588: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
7589: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7590: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
7591: /* printf("Internal IMaCh error, don't know which value for quantitative covariate with age, Tage[k]%d, k=%d, Tvar[Tage[k]]=V%d, age=%d\n",Tage[k],k ,Tvar[Tage[k]], (int)cov[2]); */
7592: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
7593: /* /\* exit(1); *\/ */
7594: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
7595: /* } */
7596: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7597: /* } */
7598: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
7599: /* if(Dummy[Tvard[k][1]]==0){ */
7600: /* if(Dummy[Tvard[k][2]]==0){ */
7601: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])]; */
7602: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7603: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
7604: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
7605: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
7606: /* } */
7607: /* }else{ */
7608: /* if(Dummy[Tvard[k][2]]==0){ */
7609: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
7610: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
7611: /* }else{ */
7612: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
7613: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
7614: /* } */
7615: /* } */
7616: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7617: /* } */
1.326 brouard 7618: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7619: for(theta=1; theta <=npar; theta++){
7620: for(i=1; i<=npar; i++)
7621: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7622:
1.222 brouard 7623: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7624:
1.222 brouard 7625: k=0;
7626: for(i=1; i<= (nlstate); i++){
7627: for(j=1; j<=(nlstate+ndeath);j++){
7628: k=k+1;
7629: gp[k]=pmmij[i][j];
7630: }
7631: }
1.220 brouard 7632:
1.222 brouard 7633: for(i=1; i<=npar; i++)
7634: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7635:
1.222 brouard 7636: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7637: k=0;
7638: for(i=1; i<=(nlstate); i++){
7639: for(j=1; j<=(nlstate+ndeath);j++){
7640: k=k+1;
7641: gm[k]=pmmij[i][j];
7642: }
7643: }
1.220 brouard 7644:
1.222 brouard 7645: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7646: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7647: }
1.126 brouard 7648:
1.222 brouard 7649: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7650: for(theta=1; theta <=npar; theta++)
7651: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7652:
1.222 brouard 7653: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7654: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7655:
1.222 brouard 7656: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7657:
1.222 brouard 7658: k=0;
7659: for(i=1; i<=(nlstate); i++){
7660: for(j=1; j<=(nlstate+ndeath);j++){
7661: k=k+1;
7662: mu[k][(int) age]=pmmij[i][j];
7663: }
7664: }
7665: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7666: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7667: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7668:
1.222 brouard 7669: /*printf("\n%d ",(int)age);
7670: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7671: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7672: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7673: }*/
1.220 brouard 7674:
1.222 brouard 7675: fprintf(ficresprob,"\n%d ",(int)age);
7676: fprintf(ficresprobcov,"\n%d ",(int)age);
7677: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7678:
1.222 brouard 7679: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7680: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7681: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7682: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7683: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7684: }
7685: i=0;
7686: for (k=1; k<=(nlstate);k++){
7687: for (l=1; l<=(nlstate+ndeath);l++){
7688: i++;
7689: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7690: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7691: for (j=1; j<=i;j++){
7692: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7693: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7694: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7695: }
7696: }
7697: }/* end of loop for state */
7698: } /* end of loop for age */
7699: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7700: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7701: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7702: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7703:
7704: /* Confidence intervalle of pij */
7705: /*
7706: fprintf(ficgp,"\nunset parametric;unset label");
7707: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7708: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7709: fprintf(fichtm,"\n<br>Probability with confidence intervals expressed in year<sup>-1</sup> :<a href=\"pijgr%s.png\">pijgr%s.png</A>, ",optionfilefiname,optionfilefiname);
7710: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7711: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7712: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7713: */
7714:
7715: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7716: first1=1;first2=2;
7717: for (k2=1; k2<=(nlstate);k2++){
7718: for (l2=1; l2<=(nlstate+ndeath);l2++){
7719: if(l2==k2) continue;
7720: j=(k2-1)*(nlstate+ndeath)+l2;
7721: for (k1=1; k1<=(nlstate);k1++){
7722: for (l1=1; l1<=(nlstate+ndeath);l1++){
7723: if(l1==k1) continue;
7724: i=(k1-1)*(nlstate+ndeath)+l1;
7725: if(i<=j) continue;
7726: for (age=bage; age<=fage; age ++){
7727: if ((int)age %5==0){
7728: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7729: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7730: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7731: mu1=mu[i][(int) age]/stepm*YEARM ;
7732: mu2=mu[j][(int) age]/stepm*YEARM;
7733: c12=cv12/sqrt(v1*v2);
7734: /* Computing eigen value of matrix of covariance */
7735: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7736: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7737: if ((lc2 <0) || (lc1 <0) ){
7738: if(first2==1){
7739: first1=0;
7740: printf("Strange: j1=%d One eigen value of 2x2 matrix of covariance is negative, lc1=%11.3e, lc2=%11.3e, v1=%11.3e, v2=%11.3e, cv12=%11.3e.\n It means that the matrix was not well estimated (varpij), for i=%2d, j=%2d, age=%4d .\n See files %s and %s. Probably WRONG RESULTS. See log file for details...\n", j1, lc1, lc2, v1, v2, cv12, i, j, (int)age,fileresprobcov, fileresprobcor);
7741: }
7742: fprintf(ficlog,"Strange: j1=%d One eigen value of 2x2 matrix of covariance is negative, lc1=%11.3e, lc2=%11.3e, v1=%11.3e, v2=%11.3e, cv12=%11.3e.\n It means that the matrix was not well estimated (varpij), for i=%2d, j=%2d, age=%4d .\n See files %s and %s. Probably WRONG RESULTS.\n", j1, lc1, lc2, v1, v2, cv12, i, j, (int)age,fileresprobcov, fileresprobcor);fflush(ficlog);
7743: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7744: /* lc2=fabs(lc2); */
7745: }
1.220 brouard 7746:
1.222 brouard 7747: /* Eigen vectors */
1.280 brouard 7748: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7749: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7750: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7751: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7752: }else
7753: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7754: /*v21=sqrt(1.-v11*v11); *//* error */
7755: v21=(lc1-v1)/cv12*v11;
7756: v12=-v21;
7757: v22=v11;
7758: tnalp=v21/v11;
7759: if(first1==1){
7760: first1=0;
7761: printf("%d %d%d-%d%d mu %.4e %.4e Var %.4e %.4e cor %.3f cov %.4e Eig %.3e %.3e 1stv %.3f %.3f tang %.3f\nOthers in log...\n",(int) age,k1,l1,k2,l2,mu1,mu2,v1,v2,c12,cv12,lc1,lc2,v11,v21,tnalp);
7762: }
7763: fprintf(ficlog,"%d %d%d-%d%d mu %.4e %.4e Var %.4e %.4e cor %.3f cov %.4e Eig %.3e %.3e 1stv %.3f %.3f tan %.3f\n",(int) age,k1,l1,k2,l2,mu1,mu2,v1,v2,c12,cv12,lc1,lc2,v11,v21,tnalp);
7764: /*printf(fignu*/
7765: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7766: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7767: if(first==1){
7768: first=0;
7769: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7770: fprintf(ficgp,"\nset parametric;unset label");
7771: fprintf(ficgp,"\nset log y;set log x; set xlabel \"p%1d%1d (year-1)\";set ylabel \"p%1d%1d (year-1)\"",k1,l1,k2,l2);
7772: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7773: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7774: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7775: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7776: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7777: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7778: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7779: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7780: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7781: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7782: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7783: fprintf(ficgp,"\nplot [-pi:pi] %11.3e+ %.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)), %11.3e +%.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)) not", \
1.280 brouard 7784: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7785: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7786: }else{
7787: first=0;
7788: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7789: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7790: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7791: fprintf(ficgp,"\nreplot %11.3e+ %.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)), %11.3e +%.3f*(%11.3e*%11.3e*cos(t)+%11.3e*%11.3e*sin(t)) not", \
1.266 brouard 7792: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7793: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7794: }/* if first */
7795: } /* age mod 5 */
7796: } /* end loop age */
7797: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7798: first=1;
7799: } /*l12 */
7800: } /* k12 */
7801: } /*l1 */
7802: }/* k1 */
1.332 brouard 7803: } /* loop on combination of covariates j1 */
1.326 brouard 7804: } /* loop on nres */
1.222 brouard 7805: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7806: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7807: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7808: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7809: free_vector(xp,1,npar);
7810: fclose(ficresprob);
7811: fclose(ficresprobcov);
7812: fclose(ficresprobcor);
7813: fflush(ficgp);
7814: fflush(fichtmcov);
7815: }
1.126 brouard 7816:
7817:
7818: /******************* Printing html file ***********/
1.201 brouard 7819: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7820: int lastpass, int stepm, int weightopt, char model[],\
7821: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7822: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7823: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7824: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7825: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7826: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7827: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7828: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7829: </ul>");
1.319 brouard 7830: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7831: /* </ul>", model); */
1.214 brouard 7832: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7833: fprintf(fichtm,"<li>- Observed frequency between two states (during the period defined between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf): <a href=\"%s\">%s</a> (html file)<br/>\n",
7834: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 7835: fprintf(fichtm,"<li> - Observed prevalence (cross-sectional prevalence) in each state (during the period defined between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf): <a href=\"%s\">%s</a> (html file) ",
1.213 brouard 7836: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7837: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7838: fprintf(fichtm,"\
7839: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7840: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7841: fprintf(fichtm,"\
1.217 brouard 7842: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7843: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7844: fprintf(fichtm,"\
1.288 brouard 7845: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7846: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7847: fprintf(fichtm,"\
1.288 brouard 7848: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7849: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7850: fprintf(fichtm,"\
1.211 brouard 7851: - (a) Life expectancies by health status at initial age, e<sub>i.</sub> (b) health expectancies by health status at initial age, e<sub>ij</sub> . If one or more covariates are included, specific tables for each value of the covariate are output in sequences within the same file (estepm=%2d months): \
1.126 brouard 7852: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7853: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7854: if(prevfcast==1){
7855: fprintf(fichtm,"\
7856: - Prevalence projections by age and states: \
1.201 brouard 7857: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7858: }
1.126 brouard 7859:
7860:
1.225 brouard 7861: m=pow(2,cptcoveff);
1.222 brouard 7862: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7863:
1.317 brouard 7864: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7865:
7866: jj1=0;
7867:
7868: fprintf(fichtm," \n<ul>");
1.337 brouard 7869: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7870: /* k1=nres; */
1.338 brouard 7871: k1=TKresult[nres];
7872: if(TKresult[nres]==0)k1=1; /* To be checked for no result */
1.337 brouard 7873: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7874: /* if(m != 1 && TKresult[nres]!= k1) */
7875: /* continue; */
1.264 brouard 7876: jj1++;
7877: if (cptcovn > 0) {
7878: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
1.337 brouard 7879: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
7880: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7881: }
1.337 brouard 7882: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7883: /* fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7884: /* } */
7885: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7886: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7887: /* } */
1.264 brouard 7888: fprintf(fichtm,"\">");
7889:
7890: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7891: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 7892: for (cpt=1; cpt<=cptcovs;cpt++){
7893: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7894: }
1.337 brouard 7895: /* fprintf(fichtm,"************ Results for covariates"); */
7896: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7897: /* fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7898: /* } */
7899: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7900: /* fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7901: /* } */
1.264 brouard 7902: if(invalidvarcomb[k1]){
7903: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7904: continue;
7905: }
7906: fprintf(fichtm,"</a></li>");
7907: } /* cptcovn >0 */
7908: }
1.317 brouard 7909: fprintf(fichtm," \n</ul>");
1.264 brouard 7910:
1.222 brouard 7911: jj1=0;
1.237 brouard 7912:
1.337 brouard 7913: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7914: /* k1=nres; */
1.338 brouard 7915: k1=TKresult[nres];
7916: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 7917: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7918: /* if(m != 1 && TKresult[nres]!= k1) */
7919: /* continue; */
1.220 brouard 7920:
1.222 brouard 7921: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7922: jj1++;
7923: if (cptcovn > 0) {
1.264 brouard 7924: fprintf(fichtm,"\n<p><a name=\"rescov");
1.337 brouard 7925: for (cpt=1; cpt<=cptcovs;cpt++){
7926: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7927: }
1.337 brouard 7928: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7929: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7930: /* } */
1.264 brouard 7931: fprintf(fichtm,"\"</a>");
7932:
1.222 brouard 7933: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 7934: for (cpt=1; cpt<=cptcovs;cpt++){
7935: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
7936: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 7937: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7938: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7939: }
1.230 brouard 7940: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.338 brouard 7941: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7942: if(invalidvarcomb[k1]){
7943: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7944: printf("\nCombination (%d) ignored because no cases \n",k1);
7945: continue;
7946: }
7947: }
7948: /* aij, bij */
1.259 brouard 7949: fprintf(fichtm,"<br>- Logit model (yours is: logit(pij)=log(pij/pii)= aij+ bij age+%s) as a function of age: <a href=\"%s_%d-1-%d.svg\">%s_%d-1-%d.svg</a><br> \
1.241 brouard 7950: <img src=\"%s_%d-1-%d.svg\">",model,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres);
1.222 brouard 7951: /* Pij */
1.241 brouard 7952: fprintf(fichtm,"<br>\n- P<sub>ij</sub> or conditional probabilities to be observed in state j being in state i, %d (stepm) months before: <a href=\"%s_%d-2-%d.svg\">%s_%d-2-%d.svg</a><br> \
7953: <img src=\"%s_%d-2-%d.svg\">",stepm,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres);
1.222 brouard 7954: /* Quasi-incidences */
7955: fprintf(fichtm,"<br>\n- I<sub>ij</sub> or Conditional probabilities to be observed in state j being in state i %d (stepm) months\
1.220 brouard 7956: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7957: incidence (rates) are the limit when h tends to zero of the ratio of the probability <sub>h</sub>P<sub>ij</sub> \
1.241 brouard 7958: divided by h: <sub>h</sub>P<sub>ij</sub>/h : <a href=\"%s_%d-3-%d.svg\">%s_%d-3-%d.svg</a><br> \
7959: <img src=\"%s_%d-3-%d.svg\">",stepm,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres,subdirf2(optionfilefiname,"PE_"),k1,nres);
1.222 brouard 7960: /* Survival functions (period) in state j */
7961: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7962: fprintf(fichtm,"<br>\n- Survival functions in state %d. And probability to be observed in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, cpt, nlstate, subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
7963: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7964: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7965: }
7966: /* State specific survival functions (period) */
7967: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7968: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7969: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7970: <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> ", cpt, nlstate, cpt, subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
7971: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7972: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7973: }
1.288 brouard 7974: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7975: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7976: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability for a person being in state (1 to %d) at different ages, to be in state %d some years after. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.338 brouard 7977: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
1.329 brouard 7978: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7979: }
1.296 brouard 7980: if(prevbcast==1){
1.288 brouard 7981: /* Backward prevalence in each health state */
1.222 brouard 7982: for(cpt=1; cpt<=nlstate;cpt++){
1.338 brouard 7983: fprintf(fichtm,"<br>\n- Convergence to mixed (stable) back prevalence in state %d. Or probability for a person to be in state %d at a younger age, knowing that she/he was in state (1 to %d) at different older ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, cpt, nlstate, subdirf2(optionfilefiname,"PB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
7984: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJB_"),subdirf2(optionfilefiname,"PIJB_"));
7985: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 7986: }
1.217 brouard 7987: }
1.222 brouard 7988: if(prevfcast==1){
1.288 brouard 7989: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7990: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7991: fprintf(fichtm,"<br>\n- Projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), from year %.1f up to year %.1f tending to period (stable) forward prevalence in state %d. Or probability to be in state %d being in an observed weighted state (from 1 to %d). <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a>", dateprev1, dateprev2, mobilavproj, dateprojd, dateprojf, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
7992: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7993: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7994: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7995: }
7996: }
1.296 brouard 7997: if(prevbcast==1){
1.268 brouard 7998: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7999: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 8000: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
8001: from year %.1f up to year %.1f (probably close to stable [mixed] back prevalence in state %d (randomness in cross-sectional prevalence is not taken into \
8002: account but can visually be appreciated). Or probability to have been in an state %d, knowing that the person was in either state (1 or %d) \
1.314 brouard 8003: with weights corresponding to observed prevalence at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a>", dateprev1, dateprev2, mobilavproj, dateback1, dateback2, cpt, cpt, nlstate, subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8004: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
8005: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 8006: }
8007: }
1.220 brouard 8008:
1.222 brouard 8009: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 8010: fprintf(fichtm,"\n<br>- Life expectancy by health state (%d) at initial age and its decomposition into health expectancies in each alive state (1 to %d) (or area under each survival functions): <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a>",cpt,nlstate,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres,subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
8011: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
8012: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 8013: }
8014: /* } /\* end i1 *\/ */
1.337 brouard 8015: }/* End k1=nres */
1.222 brouard 8016: fprintf(fichtm,"</ul>");
1.126 brouard 8017:
1.222 brouard 8018: fprintf(fichtm,"\
1.126 brouard 8019: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 8020: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 8021: - 95%% confidence intervals and Wald tests of the estimated parameters are in the log file if optimization has been done (mle != 0).<br> \
1.197 brouard 8022: But because parameters are usually highly correlated (a higher incidence of disability \
8023: and a higher incidence of recovery can give very close observed transition) it might \
8024: be very useful to look not only at linear confidence intervals estimated from the \
8025: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
8026: (parameters) of the logistic regression, it might be more meaningful to visualize the \
8027: covariance matrix of the one-step probabilities. \
8028: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 8029:
1.222 brouard 8030: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
8031: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
8032: fprintf(fichtm,"\
1.126 brouard 8033: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 8034: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 8035:
1.222 brouard 8036: fprintf(fichtm,"\
1.126 brouard 8037: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 8038: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
8039: fprintf(fichtm,"\
1.126 brouard 8040: - Variances and covariances of health expectancies by age and <b>initial health status</b> (cov(e<sup>ij</sup>,e<sup>kl</sup>)(estepm=%2d months): \
8041: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 8042: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 8043: fprintf(fichtm,"\
1.126 brouard 8044: - (a) Health expectancies by health status at initial age (e<sup>ij</sup>) and standard errors (in parentheses) (b) life expectancies and standard errors (e<sup>i.</sup>=e<sup>i1</sup>+e<sup>i2</sup>+...)(estepm=%2d months): \
8045: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 8046: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 8047: fprintf(fichtm,"\
1.288 brouard 8048: - Variances and covariances of health expectancies by age. Status (i) based health expectancies (in state j), e<sup>ij</sup> are weighted by the forward (period) prevalences in each state i (if popbased=1, an additional computation is done using the cross-sectional prevalences, i.e population based) (estepm=%d months): <a href=\"%s\">%s</a><br>\n",
1.222 brouard 8049: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
8050: fprintf(fichtm,"\
1.128 brouard 8051: - Total life expectancy and total health expectancies to be spent in each health state e<sup>.j</sup> with their standard errors (if popbased=1, an additional computation is done using the cross-sectional prevalences, i.e population based) (estepm=%d months): <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 8052: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
8053: fprintf(fichtm,"\
1.288 brouard 8054: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 8055: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 8056:
8057: /* if(popforecast==1) fprintf(fichtm,"\n */
8058: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
8059: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
8060: /* <br>",fileres,fileres,fileres,fileres); */
8061: /* else */
1.338 brouard 8062: /* fprintf(fichtm,"\n No population forecast: popforecast = %d (instead of 1) or stepm = %d (instead of 1) or model=1+age+%s (instead of .)<br><br></li>\n",popforecast, stepm, model); */
1.222 brouard 8063: fflush(fichtm);
1.126 brouard 8064:
1.225 brouard 8065: m=pow(2,cptcoveff);
1.222 brouard 8066: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 8067:
1.317 brouard 8068: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
8069:
8070: jj1=0;
8071:
8072: fprintf(fichtm," \n<ul>");
1.337 brouard 8073: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8074: /* k1=nres; */
1.338 brouard 8075: k1=TKresult[nres];
1.337 brouard 8076: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
8077: /* if(m != 1 && TKresult[nres]!= k1) */
8078: /* continue; */
1.317 brouard 8079: jj1++;
8080: if (cptcovn > 0) {
8081: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
1.337 brouard 8082: for (cpt=1; cpt<=cptcovs;cpt++){
8083: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8084: }
8085: fprintf(fichtm,"\">");
8086:
8087: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
8088: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 8089: for (cpt=1; cpt<=cptcovs;cpt++){
8090: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8091: }
8092: if(invalidvarcomb[k1]){
8093: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
8094: continue;
8095: }
8096: fprintf(fichtm,"</a></li>");
8097: } /* cptcovn >0 */
1.337 brouard 8098: } /* End nres */
1.317 brouard 8099: fprintf(fichtm," \n</ul>");
8100:
1.222 brouard 8101: jj1=0;
1.237 brouard 8102:
1.241 brouard 8103: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8104: /* k1=nres; */
1.338 brouard 8105: k1=TKresult[nres];
8106: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8107: /* for(k1=1; k1<=m;k1++){ */
8108: /* if(m != 1 && TKresult[nres]!= k1) */
8109: /* continue; */
1.222 brouard 8110: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
8111: jj1++;
1.126 brouard 8112: if (cptcovn > 0) {
1.317 brouard 8113: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
1.337 brouard 8114: for (cpt=1; cpt<=cptcovs;cpt++){
8115: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8116: }
8117: fprintf(fichtm,"\"</a>");
8118:
1.126 brouard 8119: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 8120: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcoveff number of variables */
8121: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
8122: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 8123: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 8124: }
1.237 brouard 8125:
1.338 brouard 8126: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 8127:
1.222 brouard 8128: if(invalidvarcomb[k1]){
8129: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
8130: continue;
8131: }
1.337 brouard 8132: } /* If cptcovn >0 */
1.126 brouard 8133: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 8134: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 8135: prevalence (with 95%% confidence interval) in state (%d): <a href=\"%s_%d-%d-%d.svg\"> %s_%d-%d-%d.svg</a>",mobilav,cpt,subdirf2(optionfilefiname,"V_"),cpt,k1,nres,subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
8136: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
8137: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 8138: }
8139: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 8140: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 8141: true period expectancies (those weighted with period prevalences are also\
8142: drawn in addition to the population based expectancies computed using\
1.314 brouard 8143: observed and cahotic prevalences: <a href=\"%s_%d-%d.svg\">%s_%d-%d.svg</a>",nlstate, subdirf2(optionfilefiname,"E_"),k1,nres,subdirf2(optionfilefiname,"E_"),k1,nres);
8144: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
8145: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 8146: /* } /\* end i1 *\/ */
1.241 brouard 8147: }/* End nres */
1.222 brouard 8148: fprintf(fichtm,"</ul>");
8149: fflush(fichtm);
1.126 brouard 8150: }
8151:
8152: /******************* Gnuplot file **************/
1.296 brouard 8153: void printinggnuplot(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double bage, double fage , int prevfcast, int prevbcast, char pathc[], double p[], int offyear, int offbyear){
1.126 brouard 8154:
8155: char dirfileres[132],optfileres[132];
1.264 brouard 8156: char gplotcondition[132], gplotlabel[132];
1.343 brouard 8157: int cpt=0,k1=0,i=0,k=0,j=0,jk=0,k2=0,k3=0,k4=0,kf=0,kvar=0,kk=0,ipos=0,iposold=0,ij=0, ijp=0, l=0;
1.211 brouard 8158: int lv=0, vlv=0, kl=0;
1.130 brouard 8159: int ng=0;
1.201 brouard 8160: int vpopbased;
1.223 brouard 8161: int ioffset; /* variable offset for columns */
1.270 brouard 8162: int iyearc=1; /* variable column for year of projection */
8163: int iagec=1; /* variable column for age of projection */
1.235 brouard 8164: int nres=0; /* Index of resultline */
1.266 brouard 8165: int istart=1; /* For starting graphs in projections */
1.219 brouard 8166:
1.126 brouard 8167: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
8168: /* printf("Problem with file %s",optionfilegnuplot); */
8169: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
8170: /* } */
8171:
8172: /*#ifdef windows */
8173: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 8174: /*#endif */
1.225 brouard 8175: m=pow(2,cptcoveff);
1.126 brouard 8176:
1.274 brouard 8177: /* diagram of the model */
8178: fprintf(ficgp,"\n#Diagram of the model \n");
8179: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
8180: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
8181: fprintf(ficgp,"\n#Peripheral arrows\nset for [i=1:%d] for [j=1:%d] arrow i*10+j from cos(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.95*(cos(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) - cos(pi*((1-(%d/2)*2./%d)/2+(j-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta2:0)), -0.95*(sin(pi*((1-(%d/2)*2./%d)/2+(i-1)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) - sin(pi*((1-(%d/2)*2./%d)/2+(j-1)*2./%d))+( i!=j?(i-j)/abs(i-j)*delta2:0)) ls (i < j? 1:2)\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
8182:
1.343 brouard 8183: fprintf(ficgp,"\n#Centripete arrows (turning in other direction (1-i) instead of (i-1)) \nset for [i=1:%d] for [j=1:%d] arrow (%d+1)*10+i from cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.80*(cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) ), -0.80*(sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) + yoff ) ls 4\n",nlstate, nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
1.274 brouard 8184: fprintf(ficgp,"\n#show arrow\nunset label\n");
8185: fprintf(ficgp,"\n#States labels, starting from 2 (2-i) instead of (1-i), was (i-1)\nset for [i=1:%d] label i sprintf(\"State %%d\",i) center at cos(pi*((1-(%d/2)*2./%d)/2+(2-i)*2./%d)), yoff+sin(pi*((1-(%d/2)*2./%d)/2+(2-i)*2./%d)) font \"helvetica, 16\" tc rgbcolor \"blue\"\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
8186: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
8187: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
8188: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
8189: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
8190:
1.202 brouard 8191: /* Contribution to likelihood */
8192: /* Plot the probability implied in the likelihood */
1.223 brouard 8193: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
8194: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
8195: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
8196: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 8197: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 8198: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
8199: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 8200: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
8201: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
8202: fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$13):6 t \"All sample, transitions colored by destination\" with dots lc variable; set out;\n",subdirf(fileresilk));
8203: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
8204: fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$13):5 t \"All sample, transitions colored by origin\" with dots lc variable; set out;\n\n",subdirf(fileresilk));
8205: for (i=1; i<= nlstate ; i ++) {
8206: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
8207: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
8208: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($12/4.):6 t \"p%d%d\" with points pointtype 7 ps variable lc variable \\\n",i,1,i,1);
8209: for (j=2; j<= nlstate+ndeath ; j ++) {
8210: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($12/4.):6 t \"p%d%d\" with points pointtype 7 ps variable lc variable ",i,j,i,j);
8211: }
8212: fprintf(ficgp,";\nset out; unset ylabel;\n");
8213: }
8214: /* unset log; plot "rrtest1_sorted_4/ILK_rrtest1_sorted_4.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with points lc variable */
8215: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
8216: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
8217: fprintf(ficgp,"\nset out;unset log\n");
8218: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 8219:
1.343 brouard 8220: /* Plot the probability implied in the likelihood by covariate value */
8221: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
8222: /* if(debugILK==1){ */
8223: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
8224: kvar=Tvar[TvarFind[kf]]; /* variable */
8225: k=18+Tvar[TvarFind[kf]];/*offset because there are 18 columns in the ILK_ file */
8226: for (i=1; i<= nlstate ; i ++) {
8227: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
8228: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
8229: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable \\\n",i,1,k,k,i,1,kvar);
8230: for (j=2; j<= nlstate+ndeath ; j ++) {
8231: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable ",i,j,k,k,i,j,kvar);
8232: }
8233: fprintf(ficgp,";\nset out; unset ylabel;\n");
8234: }
8235: } /* End of each covariate dummy */
8236: for(ncovv=1, iposold=0, kk=0; ncovv <= ncovvt ; ncovv++){
8237: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
8238: * kmodel = 1 2 3 4 5 6 7 8 9
8239: * varying 1 2 3 4 5
8240: * ncovv 1 2 3 4 5 6 7 8
8241: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
8242: * TvarVVind[ncovv]=kmodel 2 3 7 7 8 8 9 9
8243: * TvarFind[kmodel] 1 0 0 0 0 0 0 0 0
8244: * kdata ncovcol=[V1 V2] nqv=0 ntv=[V3 V4] nqtv=V5
8245: * Dummy[kmodel] 0 0 1 2 2 3 1 1 1
8246: */
8247: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
8248: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
8249: /* printf("DebugILK ficgp ncovv=%d, kvar=TvarVV[ncovv]=%d, ipos=TvarVVind[ncovv]=%d, Dummy[ipos]=%d, Typevar[ipos]=%d\n", ncovv,kvar,ipos,Dummy[ipos],Typevar[ipos]); */
8250: if(ipos!=iposold){ /* Not a product or first of a product */
8251: /* printf(" %d",ipos); */
8252: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
8253: /* printf(" DebugILK ficgp suite ipos=%d != iposold=%d\n", ipos, iposold); */
8254: kk++; /* Position of the ncovv column in ILK_ */
8255: k=18+ncovf+kk; /*offset because there are 18 columns in the ILK_ file plus ncovf fixed covariate */
8256: if(Dummy[ipos]==0 && Typevar[ipos]==0){ /* Only if dummy time varying: Dummy(0, 1=quant singor prod without age,2 dummy*age, 3quant*age) Typevar (0 single, 1=*age,2=Vn*vm) */
8257: for (i=1; i<= nlstate ; i ++) {
8258: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
8259: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
8260:
8261: if(gnuplotversion >=5.2){ /* Former gnuplot versions do not have variable pointsize!! */
8262: /* printf("DebugILK gnuplotversion=%g >=5.2\n",gnuplotversion); */
8263: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable \\\n",i,1,k,k,i,1,kvar);
8264: for (j=2; j<= nlstate+ndeath ; j ++) {
8265: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable ",i,j,k,k,i,j,kvar);
8266: }
8267: }else{
8268: /* printf("DebugILK gnuplotversion=%g <5.2\n",gnuplotversion); */
8269: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt 7 ps 0.4 lc variable \\\n",i,1,k,i,1,kvar);
8270: for (j=2; j<= nlstate+ndeath ; j ++) {
8271: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt 7 ps 0.4 lc variable ",i,j,k,i,j,kvar);
8272: }
8273: }
8274: fprintf(ficgp,";\nset out; unset ylabel;\n");
8275: }
8276: }/* End if dummy varying */
8277: }else{ /*Product */
8278: /* printf("*"); */
8279: /* fprintf(ficresilk,"*"); */
8280: }
8281: iposold=ipos;
8282: } /* For each time varying covariate */
8283: /* } /\* debugILK==1 *\/ */
8284: /* unset log; plot "rrtest1_sorted_4/ILK_rrtest1_sorted_4.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with points lc variable */
8285: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
8286: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
8287: fprintf(ficgp,"\nset out;unset log\n");
8288: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
8289:
8290:
8291:
1.126 brouard 8292: strcpy(dirfileres,optionfilefiname);
8293: strcpy(optfileres,"vpl");
1.223 brouard 8294: /* 1eme*/
1.238 brouard 8295: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
1.337 brouard 8296: /* for (k1=1; k1<= m ; k1 ++){ /\* For each valid combination of covariate *\/ */
1.236 brouard 8297: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8298: k1=TKresult[nres];
1.338 brouard 8299: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.238 brouard 8300: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.337 brouard 8301: /* if(m != 1 && TKresult[nres]!= k1) */
8302: /* continue; */
1.238 brouard 8303: /* We are interested in selected combination by the resultline */
1.246 brouard 8304: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 8305: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 8306: strcpy(gplotlabel,"(");
1.337 brouard 8307: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8308: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8309: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8310:
8311: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate k get corresponding value lv for combination k1 *\/ */
8312: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the value of the covariate corresponding to k1 combination *\\/ *\/ */
8313: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8314: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8315: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8316: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8317: /* vlv= nbcode[Tvaraff[k]][lv]; /\* vlv is the value of the covariate lv, 0 or 1 *\/ */
8318: /* /\* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv *\/ */
8319: /* /\* printf(" V%d=%d ",Tvaraff[k],vlv); *\/ */
8320: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8321: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8322: /* } */
8323: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8324: /* /\* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); *\/ */
8325: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8326: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.264 brouard 8327: }
8328: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 8329: /* printf("\n#\n"); */
1.238 brouard 8330: fprintf(ficgp,"\n#\n");
8331: if(invalidvarcomb[k1]){
1.260 brouard 8332: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 8333: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8334: continue;
8335: }
1.235 brouard 8336:
1.241 brouard 8337: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
8338: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 8339: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
1.338 brouard 8340: fprintf(ficgp,"set title \"Alive state %d %s model=1+age+%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 8341: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \nset ter svg size 640, 480\nplot [%.f:%.f] \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",ageminpar,fage,subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
8342: /* fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \nset ter svg size 640, 480\nplot [%.f:%.f] \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",ageminpar,fage,subdirf2(fileresu,"VPL_"),k1-1,k1-1,nres); */
8343: /* k1-1 error should be nres-1*/
1.238 brouard 8344: for (i=1; i<= nlstate ; i ++) {
8345: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8346: else fprintf(ficgp," %%*lf (%%*lf)");
8347: }
1.288 brouard 8348: fprintf(ficgp,"\" t\"Forward prevalence\" w l lt 0,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 8349: for (i=1; i<= nlstate ; i ++) {
8350: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8351: else fprintf(ficgp," %%*lf (%%*lf)");
8352: }
1.260 brouard 8353: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 8354: for (i=1; i<= nlstate ; i ++) {
8355: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8356: else fprintf(ficgp," %%*lf (%%*lf)");
8357: }
1.265 brouard 8358: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
8359:
8360: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
8361: if(cptcoveff ==0){
1.271 brouard 8362: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 8363: }else{
8364: kl=0;
8365: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8366: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8367: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 8368: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8369: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8370: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8371: vlv= nbcode[Tvaraff[k]][lv];
8372: kl++;
8373: /* kl=6+(cpt-1)*(nlstate+1)+1+(i-1); /\* 6+(1-1)*(2+1)+1+(1-1)=7, 6+(2-1)(2+1)+1+(1-1)=10 *\/ */
8374: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8375: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8376: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8377: if(k==cptcoveff){
8378: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
8379: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
8380: }else{
8381: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
8382: kl++;
8383: }
8384: } /* end covariate */
8385: } /* end if no covariate */
8386:
1.296 brouard 8387: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 8388: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 8389: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 8390: if(cptcoveff ==0){
1.245 brouard 8391: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 8392: }else{
8393: kl=0;
8394: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8395: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8396: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 8397: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8398: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8399: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8400: /* vlv= nbcode[Tvaraff[k]][lv]; */
8401: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 8402: kl++;
1.238 brouard 8403: /* kl=6+(cpt-1)*(nlstate+1)+1+(i-1); /\* 6+(1-1)*(2+1)+1+(1-1)=7, 6+(2-1)(2+1)+1+(1-1)=10 *\/ */
8404: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8405: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8406: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8407: if(k==cptcoveff){
1.245 brouard 8408: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 8409: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 8410: }else{
1.332 brouard 8411: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 8412: kl++;
8413: }
8414: } /* end covariate */
8415: } /* end if no covariate */
1.296 brouard 8416: if(prevbcast == 1){
1.268 brouard 8417: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
8418: /* k1-1 error should be nres-1*/
8419: for (i=1; i<= nlstate ; i ++) {
8420: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8421: else fprintf(ficgp," %%*lf (%%*lf)");
8422: }
1.271 brouard 8423: fprintf(ficgp,"\" t\"Backward (stable) prevalence\" w l lt 6 dt 3,\"%s\" every :::%d::%d u 1:($2==%d ? $3+1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
1.268 brouard 8424: for (i=1; i<= nlstate ; i ++) {
8425: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8426: else fprintf(ficgp," %%*lf (%%*lf)");
8427: }
1.276 brouard 8428: fprintf(ficgp,"\" t\"95%% CI\" w l lt 4,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
1.268 brouard 8429: for (i=1; i<= nlstate ; i ++) {
8430: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8431: else fprintf(ficgp," %%*lf (%%*lf)");
8432: }
1.274 brouard 8433: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 8434: } /* end if backprojcast */
1.296 brouard 8435: } /* end if prevbcast */
1.276 brouard 8436: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
8437: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 8438: } /* nres */
1.337 brouard 8439: /* } /\* k1 *\/ */
1.201 brouard 8440: } /* cpt */
1.235 brouard 8441:
8442:
1.126 brouard 8443: /*2 eme*/
1.337 brouard 8444: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8445: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8446: k1=TKresult[nres];
1.338 brouard 8447: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8448: /* if(m != 1 && TKresult[nres]!= k1) */
8449: /* continue; */
1.238 brouard 8450: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 8451: strcpy(gplotlabel,"(");
1.337 brouard 8452: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8453: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8454: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8455: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8456: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8457: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8458: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8459: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8460: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8461: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8462: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8463: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8464: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8465: /* } */
8466: /* /\* for(k=1; k <= ncovds; k++){ *\/ */
8467: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8468: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8469: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8470: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 8471: }
1.264 brouard 8472: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8473: fprintf(ficgp,"\n#\n");
1.223 brouard 8474: if(invalidvarcomb[k1]){
8475: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8476: continue;
8477: }
1.219 brouard 8478:
1.241 brouard 8479: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 8480: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 8481: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
8482: if(vpopbased==0){
1.238 brouard 8483: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 8484: }else
1.238 brouard 8485: fprintf(ficgp,"\nreplot ");
8486: for (i=1; i<= nlstate+1 ; i ++) {
8487: k=2*i;
1.261 brouard 8488: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ?$4 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1, vpopbased);
1.238 brouard 8489: for (j=1; j<= nlstate+1 ; j ++) {
8490: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8491: else fprintf(ficgp," %%*lf (%%*lf)");
8492: }
8493: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
8494: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 8495: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4-$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 8496: for (j=1; j<= nlstate+1 ; j ++) {
8497: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8498: else fprintf(ficgp," %%*lf (%%*lf)");
8499: }
8500: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 8501: fprintf(ficgp,"\"%s\" every :::%d::%d u 1:($2==%d && $4!=0 ? $4+$5*2 : 1/0) \"%%lf %%lf %%lf",subdirf2(fileresu,"T_"),nres-1,nres-1,vpopbased);
1.238 brouard 8502: for (j=1; j<= nlstate+1 ; j ++) {
8503: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8504: else fprintf(ficgp," %%*lf (%%*lf)");
8505: }
8506: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
8507: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
8508: } /* state */
8509: } /* vpopbased */
1.264 brouard 8510: fprintf(ficgp,"\nset out;set out \"%s_%d-%d.svg\"; replot; set out; unset label;\n",subdirf2(optionfilefiname,"E_"),k1,nres); /* Buggy gnuplot */
1.238 brouard 8511: } /* end nres */
1.337 brouard 8512: /* } /\* k1 end 2 eme*\/ */
1.238 brouard 8513:
8514:
8515: /*3eme*/
1.337 brouard 8516: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8517: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8518: k1=TKresult[nres];
1.338 brouard 8519: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8520: /* if(m != 1 && TKresult[nres]!= k1) */
8521: /* continue; */
1.238 brouard 8522:
1.332 brouard 8523: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 8524: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 8525: strcpy(gplotlabel,"(");
1.337 brouard 8526: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8527: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8528: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8529: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8530: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8531: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8532: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8533: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8534: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8535: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8536: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8537: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8538: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8539: /* } */
8540: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8541: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8542: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8543: }
1.264 brouard 8544: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8545: fprintf(ficgp,"\n#\n");
8546: if(invalidvarcomb[k1]){
8547: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8548: continue;
8549: }
8550:
8551: /* k=2+nlstate*(2*cpt-2); */
8552: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 8553: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 8554: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 8555: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 8556: plot [%.f:%.f] \"%s\" every :::%d::%d u 1:%d t \"e%d1\" w l",ageminpar,fage,subdirf2(fileresu,"E_"),nres-1,nres-1,k,cpt);
1.238 brouard 8557: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8558: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8559: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
8560: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8561: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8562: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 8563:
1.238 brouard 8564: */
8565: for (i=1; i< nlstate ; i ++) {
1.261 brouard 8566: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+i,cpt,i+1);
1.238 brouard 8567: /* fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d%d\" w l",subdirf2(fileres,"e"),k1-1,k1-1,k+2*i,cpt,i+1);*/
1.219 brouard 8568:
1.238 brouard 8569: }
1.261 brouard 8570: fprintf(ficgp," ,\"%s\" every :::%d::%d u 1:%d t \"e%d.\" w l",subdirf2(fileresu,"E_"),nres-1,nres-1,k+nlstate,cpt);
1.238 brouard 8571: }
1.264 brouard 8572: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 8573: } /* end nres */
1.337 brouard 8574: /* } /\* end kl 3eme *\/ */
1.126 brouard 8575:
1.223 brouard 8576: /* 4eme */
1.201 brouard 8577: /* Survival functions (period) from state i in state j by initial state i */
1.337 brouard 8578: /* for (k1=1; k1<=m; k1++){ /\* For each covariate and each value *\/ */
1.238 brouard 8579: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8580: k1=TKresult[nres];
1.338 brouard 8581: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8582: /* if(m != 1 && TKresult[nres]!= k1) */
8583: /* continue; */
1.238 brouard 8584: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 8585: strcpy(gplotlabel,"(");
1.337 brouard 8586: fprintf(ficgp,"\n#\n#\n# Survival functions in state %d : 'LIJ_' files, cov=%d state=%d", cpt, k1, cpt);
8587: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8588: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8589: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8590: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8591: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8592: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8593: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8594: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8595: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8596: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8597: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8598: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8599: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8600: /* } */
8601: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8602: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8603: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8604: }
1.264 brouard 8605: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8606: fprintf(ficgp,"\n#\n");
8607: if(invalidvarcomb[k1]){
8608: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8609: continue;
1.223 brouard 8610: }
1.238 brouard 8611:
1.241 brouard 8612: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 8613: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.238 brouard 8614: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8615: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8616: k=3;
8617: for (i=1; i<= nlstate ; i ++){
8618: if(i==1){
8619: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8620: }else{
8621: fprintf(ficgp,", '' ");
8622: }
8623: l=(nlstate+ndeath)*(i-1)+1;
8624: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8625: for (j=2; j<= nlstate+ndeath ; j ++)
8626: fprintf(ficgp,"+$%d",k+l+j-1);
8627: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
8628: } /* nlstate */
1.264 brouard 8629: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8630: } /* end cpt state*/
8631: } /* end nres */
1.337 brouard 8632: /* } /\* end covariate k1 *\/ */
1.238 brouard 8633:
1.220 brouard 8634: /* 5eme */
1.201 brouard 8635: /* Survival functions (period) from state i in state j by final state j */
1.337 brouard 8636: /* for (k1=1; k1<= m ; k1++){ /\* For each covariate combination if any *\/ */
1.238 brouard 8637: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8638: k1=TKresult[nres];
1.338 brouard 8639: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8640: /* if(m != 1 && TKresult[nres]!= k1) */
8641: /* continue; */
1.238 brouard 8642: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 8643: strcpy(gplotlabel,"(");
1.238 brouard 8644: fprintf(ficgp,"\n#\n#\n# Survival functions in state j and all livestates from state i by final state j: 'lij' files, cov=%d state=%d",k1, cpt);
1.337 brouard 8645: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8646: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8647: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8648: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8649: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8650: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8651: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8652: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8653: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8654: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8655: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8656: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8657: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8658: /* } */
8659: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8660: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8661: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8662: }
1.264 brouard 8663: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8664: fprintf(ficgp,"\n#\n");
8665: if(invalidvarcomb[k1]){
8666: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8667: continue;
8668: }
1.227 brouard 8669:
1.241 brouard 8670: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8671: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.238 brouard 8672: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8673: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8674: k=3;
8675: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8676: if(j==1)
8677: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8678: else
8679: fprintf(ficgp,", '' ");
8680: l=(nlstate+ndeath)*(cpt-1) +j;
8681: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8682: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8683: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8684: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8685: } /* nlstate */
8686: fprintf(ficgp,", '' ");
8687: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8688: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8689: l=(nlstate+ndeath)*(cpt-1) +j;
8690: if(j < nlstate)
8691: fprintf(ficgp,"$%d +",k+l);
8692: else
8693: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8694: }
1.264 brouard 8695: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8696: } /* end cpt state*/
1.337 brouard 8697: /* } /\* end covariate *\/ */
1.238 brouard 8698: } /* end nres */
1.227 brouard 8699:
1.220 brouard 8700: /* 6eme */
1.202 brouard 8701: /* CV preval stable (period) for each covariate */
1.337 brouard 8702: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8703: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8704: k1=TKresult[nres];
1.338 brouard 8705: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8706: /* if(m != 1 && TKresult[nres]!= k1) */
8707: /* continue; */
1.255 brouard 8708: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8709: strcpy(gplotlabel,"(");
1.288 brouard 8710: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8711: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8712: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8713: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8714: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8715: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8716: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8717: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8718: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8719: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8720: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8721: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8722: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8723: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8724: /* } */
8725: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8726: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8727: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8728: }
1.264 brouard 8729: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8730: fprintf(ficgp,"\n#\n");
1.223 brouard 8731: if(invalidvarcomb[k1]){
1.227 brouard 8732: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8733: continue;
1.223 brouard 8734: }
1.227 brouard 8735:
1.241 brouard 8736: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8737: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.126 brouard 8738: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8739: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8740: k=3; /* Offset */
1.255 brouard 8741: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8742: if(i==1)
8743: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8744: else
8745: fprintf(ficgp,", '' ");
1.255 brouard 8746: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8747: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8748: for (j=2; j<= nlstate ; j ++)
8749: fprintf(ficgp,"+$%d",k+l+j-1);
8750: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8751: } /* nlstate */
1.264 brouard 8752: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8753: } /* end cpt state*/
8754: } /* end covariate */
1.227 brouard 8755:
8756:
1.220 brouard 8757: /* 7eme */
1.296 brouard 8758: if(prevbcast == 1){
1.288 brouard 8759: /* CV backward prevalence for each covariate */
1.337 brouard 8760: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8761: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8762: k1=TKresult[nres];
1.338 brouard 8763: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8764: /* if(m != 1 && TKresult[nres]!= k1) */
8765: /* continue; */
1.268 brouard 8766: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8767: strcpy(gplotlabel,"(");
1.288 brouard 8768: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8769: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8770: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8771: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8772: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8773: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8774: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8775: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8776: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8777: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8778: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8779: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8780: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8781: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8782: /* } */
8783: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8784: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8785: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8786: }
1.264 brouard 8787: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8788: fprintf(ficgp,"\n#\n");
8789: if(invalidvarcomb[k1]){
8790: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8791: continue;
8792: }
8793:
1.241 brouard 8794: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8795: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 8796: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8797: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8798: k=3; /* Offset */
1.268 brouard 8799: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8800: if(i==1)
8801: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8802: else
8803: fprintf(ficgp,", '' ");
8804: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8805: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8806: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8807: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l+(cpt-1)+i-1); /\* a vérifier *\/ */
1.255 brouard 8808: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8809: /* for (j=2; j<= nlstate ; j ++) */
8810: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8811: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8812: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8813: } /* nlstate */
1.264 brouard 8814: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8815: } /* end cpt state*/
8816: } /* end covariate */
1.296 brouard 8817: } /* End if prevbcast */
1.218 brouard 8818:
1.223 brouard 8819: /* 8eme */
1.218 brouard 8820: if(prevfcast==1){
1.288 brouard 8821: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8822:
1.337 brouard 8823: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8824: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8825: k1=TKresult[nres];
1.338 brouard 8826: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8827: /* if(m != 1 && TKresult[nres]!= k1) */
8828: /* continue; */
1.211 brouard 8829: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8830: strcpy(gplotlabel,"(");
1.288 brouard 8831: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8832: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8833: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8834: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8835: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8836: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8837: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8838: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8839: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8840: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8841: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8842: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8843: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8844: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8845: /* } */
8846: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8847: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8848: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8849: }
1.264 brouard 8850: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8851: fprintf(ficgp,"\n#\n");
8852: if(invalidvarcomb[k1]){
8853: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8854: continue;
8855: }
8856:
8857: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8858: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8859: fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
1.227 brouard 8860: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8861: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8862:
8863: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8864: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8865: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8866: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8867: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8868: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8869: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8870: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8871: if(i==istart){
1.227 brouard 8872: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8873: }else{
8874: fprintf(ficgp,",\\\n '' ");
8875: }
8876: if(cptcoveff ==0){ /* No covariate */
8877: ioffset=2; /* Age is in 2 */
8878: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8879: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8880: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8881: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8882: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8883: if(i==nlstate+1){
1.270 brouard 8884: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8885: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8886: fprintf(ficgp,",\\\n '' ");
8887: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8888: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8889: offyear, \
1.268 brouard 8890: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8891: }else
1.227 brouard 8892: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8893: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8894: }else{ /* more than 2 covariates */
1.270 brouard 8895: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8896: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8897: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8898: iyearc=ioffset-1;
8899: iagec=ioffset;
1.227 brouard 8900: fprintf(ficgp," u %d:(",ioffset);
8901: kl=0;
8902: strcpy(gplotcondition,"(");
8903: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8904: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8905: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8906: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8907: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8908: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8909: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8910: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8911: kl++;
8912: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8913: kl++;
8914: if(k <cptcoveff && cptcoveff>1)
8915: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8916: }
8917: strcpy(gplotcondition+strlen(gplotcondition),")");
8918: /* kl=6+(cpt-1)*(nlstate+1)+1+(i-1); /\* 6+(1-1)*(2+1)+1+(1-1)=7, 6+(2-1)(2+1)+1+(1-1)=10 *\/ */
8919: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8920: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8921: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8922: if(i==nlstate+1){
1.270 brouard 8923: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8924: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8925: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8926: fprintf(ficgp," u %d:(",iagec);
8927: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8928: iyearc, iagec, offyear, \
8929: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8930: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
1.227 brouard 8931: }else{
8932: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8933: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8934: }
8935: } /* end if covariate */
8936: } /* nlstate */
1.264 brouard 8937: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8938: } /* end cpt state*/
8939: } /* end covariate */
8940: } /* End if prevfcast */
1.227 brouard 8941:
1.296 brouard 8942: if(prevbcast==1){
1.268 brouard 8943: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8944:
1.337 brouard 8945: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.268 brouard 8946: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8947: k1=TKresult[nres];
1.338 brouard 8948: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8949: /* if(m != 1 && TKresult[nres]!= k1) */
8950: /* continue; */
1.268 brouard 8951: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8952: strcpy(gplotlabel,"(");
8953: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
1.337 brouard 8954: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8955: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8956: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8957: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8958: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8959: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8960: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8961: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8962: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8963: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8964: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8965: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8966: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8967: /* } */
8968: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8969: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8970: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.268 brouard 8971: }
8972: strcpy(gplotlabel+strlen(gplotlabel),")");
8973: fprintf(ficgp,"\n#\n");
8974: if(invalidvarcomb[k1]){
8975: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8976: continue;
8977: }
8978:
8979: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8980: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8981: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8982: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8983: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8984:
8985: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8986: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8987: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8988: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8989: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8990: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8991: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8992: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8993: if(i==istart){
8994: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8995: }else{
8996: fprintf(ficgp,",\\\n '' ");
8997: }
8998: if(cptcoveff ==0){ /* No covariate */
8999: ioffset=2; /* Age is in 2 */
9000: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
9001: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
9002: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
9003: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
9004: fprintf(ficgp," u %d:(", ioffset);
9005: if(i==nlstate+1){
1.270 brouard 9006: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 9007: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
9008: fprintf(ficgp,",\\\n '' ");
9009: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 9010: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 9011: offbyear, \
9012: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
9013: }else
9014: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
9015: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
9016: }else{ /* more than 2 covariates */
1.270 brouard 9017: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
9018: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
9019: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
9020: iyearc=ioffset-1;
9021: iagec=ioffset;
1.268 brouard 9022: fprintf(ficgp," u %d:(",ioffset);
9023: kl=0;
9024: strcpy(gplotcondition,"(");
1.337 brouard 9025: for (k=1; k<=cptcovs; k++){ /* For each covariate k of the resultline, get corresponding value lv for combination k1 */
1.338 brouard 9026: if(Dummy[modelresult[nres][k]]==0){ /* To be verified */
1.337 brouard 9027: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
9028: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
9029: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
9030: lv=Tvresult[nres][k];
9031: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
9032: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
9033: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
9034: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
9035: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
9036: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
9037: kl++;
9038: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
9039: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%lg " ,kl,Tvresult[nres][k], kl+1,TinvDoQresult[nres][Tvresult[nres][k]]);
9040: kl++;
1.338 brouard 9041: if(k <cptcovs && cptcovs>1)
1.337 brouard 9042: sprintf(gplotcondition+strlen(gplotcondition)," && ");
9043: }
1.268 brouard 9044: }
9045: strcpy(gplotcondition+strlen(gplotcondition),")");
9046: /* kl=6+(cpt-1)*(nlstate+1)+1+(i-1); /\* 6+(1-1)*(2+1)+1+(1-1)=7, 6+(2-1)(2+1)+1+(1-1)=10 *\/ */
9047: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
9048: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
9049: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
9050: if(i==nlstate+1){
1.270 brouard 9051: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
9052: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 9053: fprintf(ficgp,",\\\n '' ");
1.270 brouard 9054: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 9055: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 9056: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
9057: iyearc,iagec,offbyear, \
9058: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 9059: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
9060: }else{
9061: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
9062: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
9063: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
9064: }
9065: } /* end if covariate */
9066: } /* nlstate */
9067: fprintf(ficgp,"\nset out; unset label;\n");
9068: } /* end cpt state*/
9069: } /* end covariate */
1.296 brouard 9070: } /* End if prevbcast */
1.268 brouard 9071:
1.227 brouard 9072:
1.238 brouard 9073: /* 9eme writing MLE parameters */
9074: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 9075: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 9076: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 9077: for(k=1; k <=(nlstate+ndeath); k++){
9078: if (k != i) {
1.227 brouard 9079: fprintf(ficgp,"# current state %d\n",k);
9080: for(j=1; j <=ncovmodel; j++){
9081: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
9082: jk++;
9083: }
9084: fprintf(ficgp,"\n");
1.126 brouard 9085: }
9086: }
1.223 brouard 9087: }
1.187 brouard 9088: fprintf(ficgp,"##############\n#\n");
1.227 brouard 9089:
1.145 brouard 9090: /*goto avoid;*/
1.238 brouard 9091: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
9092: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 9093: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
9094: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
9095: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
9096: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
9097: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9098: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
9099: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9100: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
9101: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
9102: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9103: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
9104: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
9105: fprintf(ficgp,"#\n");
1.223 brouard 9106: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 9107: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.338 brouard 9108: fprintf(ficgp,"#model=1+age+%s \n",model);
1.238 brouard 9109: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 9110: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
1.337 brouard 9111: /* for(k1=1; k1 <=m; k1++) /\* For each combination of covariate *\/ */
1.237 brouard 9112: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 9113: /* k1=nres; */
1.338 brouard 9114: k1=TKresult[nres];
9115: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 9116: fprintf(ficgp,"\n\n# Resultline k1=%d ",k1);
1.264 brouard 9117: strcpy(gplotlabel,"(");
1.276 brouard 9118: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.337 brouard 9119: for (k=1; k<=cptcovs; k++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
9120: /* for each resultline nres, and position k, Tvresult[nres][k] gives the name of the variable and
9121: TinvDoQresult[nres][Tvresult[nres][k]] gives its value double or integer) */
9122: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
9123: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
9124: }
9125: /* if(m != 1 && TKresult[nres]!= k1) */
9126: /* continue; */
9127: /* fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1); */
9128: /* strcpy(gplotlabel,"("); */
9129: /* /\*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*\/ */
9130: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
9131: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
9132: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
9133: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
9134: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
9135: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
9136: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
9137: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
9138: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
9139: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
9140: /* } */
9141: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9142: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
9143: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
9144: /* } */
1.264 brouard 9145: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 9146: fprintf(ficgp,"\n#\n");
1.264 brouard 9147: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 9148: fprintf(ficgp,"\nset key outside ");
9149: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
9150: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 9151: fprintf(ficgp,"\nset ter svg size 640, 480 ");
9152: if (ng==1){
9153: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
9154: fprintf(ficgp,"\nunset log y");
9155: }else if (ng==2){
9156: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
9157: fprintf(ficgp,"\nset log y");
9158: }else if (ng==3){
9159: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
9160: fprintf(ficgp,"\nset log y");
9161: }else
9162: fprintf(ficgp,"\nunset title ");
9163: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
9164: i=1;
9165: for(k2=1; k2<=nlstate; k2++) {
9166: k3=i;
9167: for(k=1; k<=(nlstate+ndeath); k++) {
9168: if (k != k2){
9169: switch( ng) {
9170: case 1:
9171: if(nagesqr==0)
9172: fprintf(ficgp," p%d+p%d*x",i,i+1);
9173: else /* nagesqr =1 */
9174: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9175: break;
9176: case 2: /* ng=2 */
9177: if(nagesqr==0)
9178: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
9179: else /* nagesqr =1 */
9180: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9181: break;
9182: case 3:
9183: if(nagesqr==0)
9184: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
9185: else /* nagesqr =1 */
9186: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
9187: break;
9188: }
9189: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 9190: ijp=1; /* product no age */
9191: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
9192: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 9193: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 9194: switch(Typevar[j]){
9195: case 1:
9196: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9197: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
9198: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9199: if(DummyV[j]==0){/* Bug valgrind */
9200: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
9201: }else{ /* quantitative */
9202: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9203: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9204: }
9205: ij++;
1.268 brouard 9206: }
1.237 brouard 9207: }
1.329 brouard 9208: }
9209: break;
9210: case 2:
9211: if(cptcovprod >0){
9212: if(j==Tprod[ijp]) { /* */
9213: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9214: if(ijp <=cptcovprod) { /* Product */
9215: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9216: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9217: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],nbcode[Tvard[ijp][2]][codtabm(k1,j)]); */
9218: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9219: }else{ /* Vn is dummy and Vm is quanti */
9220: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9221: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9222: }
9223: }else{ /* Vn*Vm Vn is quanti */
9224: if(DummyV[Tvard[ijp][2]]==0){
9225: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9226: }else{ /* Both quanti */
9227: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9228: }
1.268 brouard 9229: }
1.329 brouard 9230: ijp++;
1.237 brouard 9231: }
1.329 brouard 9232: } /* end Tprod */
9233: }
9234: break;
9235: case 0:
9236: /* simple covariate */
1.264 brouard 9237: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 9238: if(Dummy[j]==0){
9239: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
9240: }else{ /* quantitative */
9241: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 9242: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 9243: }
1.329 brouard 9244: /* end simple */
9245: break;
9246: default:
9247: break;
9248: } /* end switch */
1.237 brouard 9249: } /* end j */
1.329 brouard 9250: }else{ /* k=k2 */
9251: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
9252: fprintf(ficgp," (1.");i=i-ncovmodel;
9253: }else
9254: i=i-ncovmodel;
1.223 brouard 9255: }
1.227 brouard 9256:
1.223 brouard 9257: if(ng != 1){
9258: fprintf(ficgp,")/(1");
1.227 brouard 9259:
1.264 brouard 9260: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 9261: if(nagesqr==0)
1.264 brouard 9262: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 9263: else /* nagesqr =1 */
1.264 brouard 9264: fprintf(ficgp,"+exp(p%d+p%d*x+p%d*x*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1,k3+(cpt-1)*ncovmodel+1+nagesqr);
1.217 brouard 9265:
1.223 brouard 9266: ij=1;
1.329 brouard 9267: ijp=1;
9268: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
9269: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
9270: switch(Typevar[j]){
9271: case 1:
9272: if(cptcovage >0){
9273: if(j==Tage[ij]) { /* Bug valgrind */
9274: if(ij <=cptcovage) { /* Bug valgrind */
9275: if(DummyV[j]==0){/* Bug valgrind */
9276: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
9277: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
9278: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
9279: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
9280: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9281: }else{ /* quantitative */
9282: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9283: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9284: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9285: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9286: }
9287: ij++;
9288: }
9289: }
9290: }
9291: break;
9292: case 2:
9293: if(cptcovprod >0){
9294: if(j==Tprod[ijp]) { /* */
9295: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9296: if(ijp <=cptcovprod) { /* Product */
9297: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9298: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9299: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],nbcode[Tvard[ijp][2]][codtabm(k1,j)]); */
9300: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9301: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
9302: }else{ /* Vn is dummy and Vm is quanti */
9303: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9304: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9305: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9306: }
9307: }else{ /* Vn*Vm Vn is quanti */
9308: if(DummyV[Tvard[ijp][2]]==0){
9309: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9310: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
9311: }else{ /* Both quanti */
9312: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9313: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9314: }
9315: }
9316: ijp++;
9317: }
9318: } /* end Tprod */
9319: } /* end if */
9320: break;
9321: case 0:
9322: /* simple covariate */
9323: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
9324: if(Dummy[j]==0){
9325: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9326: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
9327: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9328: }else{ /* quantitative */
9329: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
9330: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
9331: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9332: }
9333: /* end simple */
9334: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
9335: break;
9336: default:
9337: break;
9338: } /* end switch */
1.223 brouard 9339: }
9340: fprintf(ficgp,")");
9341: }
9342: fprintf(ficgp,")");
9343: if(ng ==2)
1.276 brouard 9344: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"p%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 9345: else /* ng= 3 */
1.276 brouard 9346: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"i%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.329 brouard 9347: }else{ /* end ng <> 1 */
1.223 brouard 9348: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 9349: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"logit(p%d%d)\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 9350: }
9351: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
9352: fprintf(ficgp,",");
9353: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
9354: fprintf(ficgp,",");
9355: i=i+ncovmodel;
9356: } /* end k */
9357: } /* end k2 */
1.276 brouard 9358: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
9359: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.337 brouard 9360: } /* end resultline */
1.223 brouard 9361: } /* end ng */
9362: /* avoid: */
9363: fflush(ficgp);
1.126 brouard 9364: } /* end gnuplot */
9365:
9366:
9367: /*************** Moving average **************/
1.219 brouard 9368: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 9369: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 9370:
1.222 brouard 9371: int i, cpt, cptcod;
9372: int modcovmax =1;
9373: int mobilavrange, mob;
9374: int iage=0;
1.288 brouard 9375: int firstA1=0, firstA2=0;
1.222 brouard 9376:
1.266 brouard 9377: double sum=0., sumr=0.;
1.222 brouard 9378: double age;
1.266 brouard 9379: double *sumnewp, *sumnewm, *sumnewmr;
9380: double *agemingood, *agemaxgood;
9381: double *agemingoodr, *agemaxgoodr;
1.222 brouard 9382:
9383:
1.278 brouard 9384: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
9385: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 9386:
9387: sumnewp = vector(1,ncovcombmax);
9388: sumnewm = vector(1,ncovcombmax);
1.266 brouard 9389: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 9390: agemingood = vector(1,ncovcombmax);
1.266 brouard 9391: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 9392: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 9393: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 9394:
9395: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 9396: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 9397: sumnewp[cptcod]=0.;
1.266 brouard 9398: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
9399: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 9400: }
9401: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
9402:
1.266 brouard 9403: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
9404: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 9405: else mobilavrange=mobilav;
9406: for (age=bage; age<=fage; age++)
9407: for (i=1; i<=nlstate;i++)
9408: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
9409: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9410: /* We keep the original values on the extreme ages bage, fage and for
9411: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
9412: we use a 5 terms etc. until the borders are no more concerned.
9413: */
9414: for (mob=3;mob <=mobilavrange;mob=mob+2){
9415: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 9416: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
9417: sumnewm[cptcod]=0.;
9418: for (i=1; i<=nlstate;i++){
1.222 brouard 9419: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
9420: for (cpt=1;cpt<=(mob-1)/2;cpt++){
9421: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
9422: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
9423: }
9424: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 9425: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9426: } /* end i */
9427: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
9428: } /* end cptcod */
1.222 brouard 9429: }/* end age */
9430: }/* end mob */
1.266 brouard 9431: }else{
9432: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 9433: return -1;
1.266 brouard 9434: }
9435:
9436: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 9437: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
9438: if(invalidvarcomb[cptcod]){
9439: printf("\nCombination (%d) ignored because no cases \n",cptcod);
9440: continue;
9441: }
1.219 brouard 9442:
1.266 brouard 9443: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
9444: sumnewm[cptcod]=0.;
9445: sumnewmr[cptcod]=0.;
9446: for (i=1; i<=nlstate;i++){
9447: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9448: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9449: }
9450: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9451: agemingoodr[cptcod]=age;
9452: }
9453: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9454: agemingood[cptcod]=age;
9455: }
9456: } /* age */
9457: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 9458: sumnewm[cptcod]=0.;
1.266 brouard 9459: sumnewmr[cptcod]=0.;
1.222 brouard 9460: for (i=1; i<=nlstate;i++){
9461: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9462: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9463: }
9464: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9465: agemaxgoodr[cptcod]=age;
1.222 brouard 9466: }
9467: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 9468: agemaxgood[cptcod]=age;
9469: }
9470: } /* age */
9471: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
9472: /* but they will change */
1.288 brouard 9473: firstA1=0;firstA2=0;
1.266 brouard 9474: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
9475: sumnewm[cptcod]=0.;
9476: sumnewmr[cptcod]=0.;
9477: for (i=1; i<=nlstate;i++){
9478: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9479: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9480: }
9481: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9482: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9483: agemaxgoodr[cptcod]=age; /* age min */
9484: for (i=1; i<=nlstate;i++)
9485: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9486: }else{ /* bad we change the value with the values of good ages */
9487: for (i=1; i<=nlstate;i++){
9488: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
9489: } /* i */
9490: } /* end bad */
9491: }else{
9492: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9493: agemaxgood[cptcod]=age;
9494: }else{ /* bad we change the value with the values of good ages */
9495: for (i=1; i<=nlstate;i++){
9496: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
9497: } /* i */
9498: } /* end bad */
9499: }/* end else */
9500: sum=0.;sumr=0.;
9501: for (i=1; i<=nlstate;i++){
9502: sum+=mobaverage[(int)age][i][cptcod];
9503: sumr+=probs[(int)age][i][cptcod];
9504: }
9505: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 9506: if(!firstA1){
9507: firstA1=1;
9508: printf("Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d. Others in log file...\n",cptcod,sumr, (int)age, (int)bage);
9509: }
9510: fprintf(ficlog,"Moving average A1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.266 brouard 9511: } /* end bad */
9512: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9513: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 9514: if(!firstA2){
9515: firstA2=1;
9516: printf("Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d. Others in log file...\n",cptcod,sumr, (int)age, (int)bage);
9517: }
9518: fprintf(ficlog,"Moving average A2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase bage=%d\n",cptcod,sumr, (int)age, (int)bage);
1.222 brouard 9519: } /* end bad */
9520: }/* age */
1.266 brouard 9521:
9522: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 9523: sumnewm[cptcod]=0.;
1.266 brouard 9524: sumnewmr[cptcod]=0.;
1.222 brouard 9525: for (i=1; i<=nlstate;i++){
9526: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9527: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9528: }
9529: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9530: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
9531: agemingoodr[cptcod]=age;
9532: for (i=1; i<=nlstate;i++)
9533: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9534: }else{ /* bad we change the value with the values of good ages */
9535: for (i=1; i<=nlstate;i++){
9536: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
9537: } /* i */
9538: } /* end bad */
9539: }else{
9540: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9541: agemingood[cptcod]=age;
9542: }else{ /* bad */
9543: for (i=1; i<=nlstate;i++){
9544: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
9545: } /* i */
9546: } /* end bad */
9547: }/* end else */
9548: sum=0.;sumr=0.;
9549: for (i=1; i<=nlstate;i++){
9550: sum+=mobaverage[(int)age][i][cptcod];
9551: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 9552: }
1.266 brouard 9553: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 9554: printf("Moving average B1: For this combination of covariate cptcod=%d, we can't get a smoothed prevalence which sums to one (%f) at any descending age! age=%d, could you decrease fage=%d?\n",cptcod, sum, (int) age, (int)fage);
1.266 brouard 9555: } /* end bad */
9556: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9557: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 9558: printf("Moving average B2: For this combination of covariate cptcod=%d, the raw prevalence doesn't sums to one (%f) even with smoothed values at young ages! age=%d, could you increase fage=%d\n",cptcod,sumr, (int)age, (int)fage);
1.222 brouard 9559: } /* end bad */
9560: }/* age */
1.266 brouard 9561:
1.222 brouard 9562:
9563: for (age=bage; age<=fage; age++){
1.235 brouard 9564: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 9565: sumnewp[cptcod]=0.;
9566: sumnewm[cptcod]=0.;
9567: for (i=1; i<=nlstate;i++){
9568: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
9569: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9570: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
9571: }
9572: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
9573: }
9574: /* printf("\n"); */
9575: /* } */
1.266 brouard 9576:
1.222 brouard 9577: /* brutal averaging */
1.266 brouard 9578: /* for (i=1; i<=nlstate;i++){ */
9579: /* for (age=1; age<=bage; age++){ */
9580: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
9581: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9582: /* } */
9583: /* for (age=fage; age<=AGESUP; age++){ */
9584: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
9585: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9586: /* } */
9587: /* } /\* end i status *\/ */
9588: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
9589: /* for (age=1; age<=AGESUP; age++){ */
9590: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
9591: /* mobaverage[(int)age][i][cptcod]=0.; */
9592: /* } */
9593: /* } */
1.222 brouard 9594: }/* end cptcod */
1.266 brouard 9595: free_vector(agemaxgoodr,1, ncovcombmax);
9596: free_vector(agemaxgood,1, ncovcombmax);
9597: free_vector(agemingood,1, ncovcombmax);
9598: free_vector(agemingoodr,1, ncovcombmax);
9599: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 9600: free_vector(sumnewm,1, ncovcombmax);
9601: free_vector(sumnewp,1, ncovcombmax);
9602: return 0;
9603: }/* End movingaverage */
1.218 brouard 9604:
1.126 brouard 9605:
1.296 brouard 9606:
1.126 brouard 9607: /************** Forecasting ******************/
1.296 brouard 9608: /* void prevforecast(char fileres[], double dateintmean, double anprojd, double mprojd, double jprojd, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double anprojf, double p[], int cptcoveff)*/
9609: void prevforecast(char fileres[], double dateintmean, double dateprojd, double dateprojf, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double ***prev, double bage, double fage, int firstpass, int lastpass, double p[], int cptcoveff){
9610: /* dateintemean, mean date of interviews
9611: dateprojd, year, month, day of starting projection
9612: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 9613: agemin, agemax range of age
9614: dateprev1 dateprev2 range of dates during which prevalence is computed
9615: */
1.296 brouard 9616: /* double anprojd, mprojd, jprojd; */
9617: /* double anprojf, mprojf, jprojf; */
1.267 brouard 9618: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 9619: double agec; /* generic age */
1.296 brouard 9620: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 9621: double *popeffectif,*popcount;
9622: double ***p3mat;
1.218 brouard 9623: /* double ***mobaverage; */
1.126 brouard 9624: char fileresf[FILENAMELENGTH];
9625:
9626: agelim=AGESUP;
1.211 brouard 9627: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9628: in each health status at the date of interview (if between dateprev1 and dateprev2).
9629: We still use firstpass and lastpass as another selection.
9630: */
1.214 brouard 9631: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9632: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 9633:
1.201 brouard 9634: strcpy(fileresf,"F_");
9635: strcat(fileresf,fileresu);
1.126 brouard 9636: if((ficresf=fopen(fileresf,"w"))==NULL) {
9637: printf("Problem with forecast resultfile: %s\n", fileresf);
9638: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
9639: }
1.235 brouard 9640: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
9641: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 9642:
1.225 brouard 9643: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 9644:
9645:
9646: stepsize=(int) (stepm+YEARM-1)/YEARM;
9647: if (stepm<=12) stepsize=1;
9648: if(estepm < stepm){
9649: printf ("Problem %d lower than %d\n",estepm, stepm);
9650: }
1.270 brouard 9651: else{
9652: hstepm=estepm;
9653: }
9654: if(estepm > stepm){ /* Yes every two year */
9655: stepsize=2;
9656: }
1.296 brouard 9657: hstepm=hstepm/stepm;
1.126 brouard 9658:
1.296 brouard 9659:
9660: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9661: /* fractional in yp1 *\/ */
9662: /* aintmean=yp; */
9663: /* yp2=modf((yp1*12),&yp); */
9664: /* mintmean=yp; */
9665: /* yp1=modf((yp2*30.5),&yp); */
9666: /* jintmean=yp; */
9667: /* if(jintmean==0) jintmean=1; */
9668: /* if(mintmean==0) mintmean=1; */
1.126 brouard 9669:
1.296 brouard 9670:
9671: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
9672: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
9673: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 9674: i1=pow(2,cptcoveff);
1.126 brouard 9675: if (cptcovn < 1){i1=1;}
9676:
1.296 brouard 9677: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 9678:
9679: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 9680:
1.126 brouard 9681: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 9682: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9683: for(k=1; k<=i1;k++){ /* We want to find the combination k corresponding to the values of the dummies given in this resut line (to be cleaned one day) */
1.253 brouard 9684: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9685: continue;
1.227 brouard 9686: if(invalidvarcomb[k]){
9687: printf("\nCombination (%d) projection ignored because no cases \n",k);
9688: continue;
9689: }
9690: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
9691: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9692: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); */
9693: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 9694: }
1.235 brouard 9695: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 9696: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 9697: }
1.227 brouard 9698: fprintf(ficresf," yearproj age");
9699: for(j=1; j<=nlstate+ndeath;j++){
9700: for(i=1; i<=nlstate;i++)
9701: fprintf(ficresf," p%d%d",i,j);
9702: fprintf(ficresf," wp.%d",j);
9703: }
1.296 brouard 9704: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 9705: fprintf(ficresf,"\n");
1.296 brouard 9706: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 9707: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
9708: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 9709: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
9710: nhstepm = nhstepm/hstepm;
9711: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9712: oldm=oldms;savm=savms;
1.268 brouard 9713: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 9714: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 9715: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 9716: for (h=0; h<=nhstepm; h++){
9717: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 9718: break;
9719: }
9720: }
9721: fprintf(ficresf,"\n");
9722: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9723: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
9724: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* TnsdVar[Tvaraff] correct */
1.296 brouard 9725: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9726:
9727: for(j=1; j<=nlstate+ndeath;j++) {
9728: ppij=0.;
9729: for(i=1; i<=nlstate;i++) {
1.278 brouard 9730: if (mobilav>=1)
9731: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9732: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9733: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9734: }
1.268 brouard 9735: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9736: } /* end i */
9737: fprintf(ficresf," %.3f", ppij);
9738: }/* end j */
1.227 brouard 9739: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9740: } /* end agec */
1.266 brouard 9741: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9742: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9743: } /* end yearp */
9744: } /* end k */
1.219 brouard 9745:
1.126 brouard 9746: fclose(ficresf);
1.215 brouard 9747: printf("End of Computing forecasting \n");
9748: fprintf(ficlog,"End of Computing forecasting\n");
9749:
1.126 brouard 9750: }
9751:
1.269 brouard 9752: /************** Back Forecasting ******************/
1.296 brouard 9753: /* void prevbackforecast(char fileres[], double ***prevacurrent, double anback1, double mback1, double jback1, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double anback2, double p[], int cptcoveff){ */
9754: void prevbackforecast(char fileres[], double ***prevacurrent, double dateintmean, double dateprojd, double dateprojf, double ageminpar, double agemax, double dateprev1, double dateprev2, int mobilav, double bage, double fage, int firstpass, int lastpass, double p[], int cptcoveff){
9755: /* back1, year, month, day of starting backprojection
1.267 brouard 9756: agemin, agemax range of age
9757: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9758: anback2 year of end of backprojection (same day and month as back1).
9759: prevacurrent and prev are prevalences.
1.267 brouard 9760: */
9761: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9762: double agec; /* generic age */
1.302 brouard 9763: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9764: double *popeffectif,*popcount;
9765: double ***p3mat;
9766: /* double ***mobaverage; */
9767: char fileresfb[FILENAMELENGTH];
9768:
1.268 brouard 9769: agelim=AGEINF;
1.267 brouard 9770: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9771: in each health status at the date of interview (if between dateprev1 and dateprev2).
9772: We still use firstpass and lastpass as another selection.
9773: */
9774: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9775: /* firstpass, lastpass, stepm, weightopt, model); */
9776:
9777: /*Do we need to compute prevalence again?*/
9778:
9779: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9780:
9781: strcpy(fileresfb,"FB_");
9782: strcat(fileresfb,fileresu);
9783: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9784: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9785: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9786: }
9787: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9788: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9789:
9790: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9791:
9792:
9793: stepsize=(int) (stepm+YEARM-1)/YEARM;
9794: if (stepm<=12) stepsize=1;
9795: if(estepm < stepm){
9796: printf ("Problem %d lower than %d\n",estepm, stepm);
9797: }
1.270 brouard 9798: else{
9799: hstepm=estepm;
9800: }
9801: if(estepm >= stepm){ /* Yes every two year */
9802: stepsize=2;
9803: }
1.267 brouard 9804:
9805: hstepm=hstepm/stepm;
1.296 brouard 9806: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9807: /* fractional in yp1 *\/ */
9808: /* aintmean=yp; */
9809: /* yp2=modf((yp1*12),&yp); */
9810: /* mintmean=yp; */
9811: /* yp1=modf((yp2*30.5),&yp); */
9812: /* jintmean=yp; */
9813: /* if(jintmean==0) jintmean=1; */
9814: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9815:
9816: i1=pow(2,cptcoveff);
9817: if (cptcovn < 1){i1=1;}
9818:
1.296 brouard 9819: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9820: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9821:
9822: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9823:
9824: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9825: for(k=1; k<=i1;k++){
9826: if(i1 != 1 && TKresult[nres]!= k)
9827: continue;
9828: if(invalidvarcomb[k]){
9829: printf("\nCombination (%d) projection ignored because no cases \n",k);
9830: continue;
9831: }
1.268 brouard 9832: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9833: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9834: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.267 brouard 9835: }
9836: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9837: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9838: }
9839: fprintf(ficresfb," yearbproj age");
9840: for(j=1; j<=nlstate+ndeath;j++){
9841: for(i=1; i<=nlstate;i++)
1.268 brouard 9842: fprintf(ficresfb," b%d%d",i,j);
9843: fprintf(ficresfb," b.%d",j);
1.267 brouard 9844: }
1.296 brouard 9845: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9846: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9847: fprintf(ficresfb,"\n");
1.296 brouard 9848: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9849: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9850: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9851: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9852: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9853: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9854: nhstepm = nhstepm/hstepm;
9855: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9856: oldm=oldms;savm=savms;
1.268 brouard 9857: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9858: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9859: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9860: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9861: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9862: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9863: for (h=0; h<=nhstepm; h++){
1.268 brouard 9864: if (h*hstepm/YEARM*stepm ==-yearp) {
9865: break;
9866: }
9867: }
9868: fprintf(ficresfb,"\n");
9869: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9870: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.296 brouard 9871: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9872: for(i=1; i<=nlstate+ndeath;i++) {
9873: ppij=0.;ppi=0.;
9874: for(j=1; j<=nlstate;j++) {
9875: /* if (mobilav==1) */
1.269 brouard 9876: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9877: ppi=ppi+prevacurrent[(int)agec][j][k];
9878: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9879: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9880: /* else { */
9881: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9882: /* } */
1.268 brouard 9883: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9884: } /* end j */
9885: if(ppi <0.99){
9886: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9887: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9888: }
9889: fprintf(ficresfb," %.3f", ppij);
9890: }/* end j */
1.267 brouard 9891: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9892: } /* end agec */
9893: } /* end yearp */
9894: } /* end k */
1.217 brouard 9895:
1.267 brouard 9896: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9897:
1.267 brouard 9898: fclose(ficresfb);
9899: printf("End of Computing Back forecasting \n");
9900: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9901:
1.267 brouard 9902: }
1.217 brouard 9903:
1.269 brouard 9904: /* Variance of prevalence limit: varprlim */
9905: void varprlim(char fileresu[], int nresult, double ***prevacurrent, int mobilavproj, double bage, double fage, double **prlim, int *ncvyearp, double ftolpl, double p[], double **matcov, double *delti, int stepm, int cptcoveff){
1.288 brouard 9906: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9907:
9908: char fileresvpl[FILENAMELENGTH];
9909: FILE *ficresvpl;
9910: double **oldm, **savm;
9911: double **varpl; /* Variances of prevalence limits by age */
9912: int i1, k, nres, j ;
9913:
9914: strcpy(fileresvpl,"VPL_");
9915: strcat(fileresvpl,fileresu);
9916: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9917: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9918: exit(0);
9919: }
1.288 brouard 9920: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9921: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9922:
9923: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9924: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9925:
9926: i1=pow(2,cptcoveff);
9927: if (cptcovn < 1){i1=1;}
9928:
1.337 brouard 9929: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9930: k=TKresult[nres];
1.338 brouard 9931: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 9932: /* for(k=1; k<=i1;k++){ /\* We find the combination equivalent to result line values of dummies *\/ */
1.269 brouard 9933: if(i1 != 1 && TKresult[nres]!= k)
9934: continue;
9935: fprintf(ficresvpl,"\n#****** ");
9936: printf("\n#****** ");
9937: fprintf(ficlog,"\n#****** ");
1.337 brouard 9938: for(j=1;j<=cptcovs;j++) {
9939: fprintf(ficresvpl,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9940: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9941: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9942: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9943: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.269 brouard 9944: }
1.337 brouard 9945: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
9946: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9947: /* fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9948: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9949: /* } */
1.269 brouard 9950: fprintf(ficresvpl,"******\n");
9951: printf("******\n");
9952: fprintf(ficlog,"******\n");
9953:
9954: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9955: oldm=oldms;savm=savms;
9956: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9957: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9958: /*}*/
9959: }
9960:
9961: fclose(ficresvpl);
1.288 brouard 9962: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9963: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9964:
9965: }
9966: /* Variance of back prevalence: varbprlim */
9967: void varbprlim(char fileresu[], int nresult, double ***prevacurrent, int mobilavproj, double bage, double fage, double **bprlim, int *ncvyearp, double ftolpl, double p[], double **matcov, double *delti, int stepm, int cptcoveff){
9968: /*------- Variance of back (stable) prevalence------*/
9969:
9970: char fileresvbl[FILENAMELENGTH];
9971: FILE *ficresvbl;
9972:
9973: double **oldm, **savm;
9974: double **varbpl; /* Variances of back prevalence limits by age */
9975: int i1, k, nres, j ;
9976:
9977: strcpy(fileresvbl,"VBL_");
9978: strcat(fileresvbl,fileresu);
9979: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9980: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9981: exit(0);
9982: }
9983: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9984: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9985:
9986:
9987: i1=pow(2,cptcoveff);
9988: if (cptcovn < 1){i1=1;}
9989:
1.337 brouard 9990: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9991: k=TKresult[nres];
1.338 brouard 9992: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 9993: /* for(k=1; k<=i1;k++){ */
9994: /* if(i1 != 1 && TKresult[nres]!= k) */
9995: /* continue; */
1.269 brouard 9996: fprintf(ficresvbl,"\n#****** ");
9997: printf("\n#****** ");
9998: fprintf(ficlog,"\n#****** ");
1.337 brouard 9999: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */
1.338 brouard 10000: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
10001: fprintf(ficresvbl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
10002: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
1.337 brouard 10003: /* for(j=1;j<=cptcoveff;j++) { */
10004: /* fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
10005: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
10006: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
10007: /* } */
10008: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
10009: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
10010: /* fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
10011: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
1.269 brouard 10012: }
10013: fprintf(ficresvbl,"******\n");
10014: printf("******\n");
10015: fprintf(ficlog,"******\n");
10016:
10017: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
10018: oldm=oldms;savm=savms;
10019:
10020: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
10021: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
10022: /*}*/
10023: }
10024:
10025: fclose(ficresvbl);
10026: printf("done variance-covariance of back prevalence\n");fflush(stdout);
10027: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
10028:
10029: } /* End of varbprlim */
10030:
1.126 brouard 10031: /************** Forecasting *****not tested NB*************/
1.227 brouard 10032: /* void populforecast(char fileres[], double anpyram,double mpyram,double jpyram,double ageminpar, double agemax,double dateprev1, double dateprev2s, int mobilav, double agedeb, double fage, int popforecast, char popfile[], double anpyram1,double p[], int i2){ */
1.126 brouard 10033:
1.227 brouard 10034: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
10035: /* int *popage; */
10036: /* double calagedatem, agelim, kk1, kk2; */
10037: /* double *popeffectif,*popcount; */
10038: /* double ***p3mat,***tabpop,***tabpopprev; */
10039: /* /\* double ***mobaverage; *\/ */
10040: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 10041:
1.227 brouard 10042: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10043: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10044: /* agelim=AGESUP; */
10045: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 10046:
1.227 brouard 10047: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 10048:
10049:
1.227 brouard 10050: /* strcpy(filerespop,"POP_"); */
10051: /* strcat(filerespop,fileresu); */
10052: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
10053: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
10054: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
10055: /* } */
10056: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
10057: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 10058:
1.227 brouard 10059: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 10060:
1.227 brouard 10061: /* /\* if (mobilav!=0) { *\/ */
10062: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
10063: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
10064: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
10065: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
10066: /* /\* } *\/ */
10067: /* /\* } *\/ */
1.126 brouard 10068:
1.227 brouard 10069: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
10070: /* if (stepm<=12) stepsize=1; */
1.126 brouard 10071:
1.227 brouard 10072: /* agelim=AGESUP; */
1.126 brouard 10073:
1.227 brouard 10074: /* hstepm=1; */
10075: /* hstepm=hstepm/stepm; */
1.218 brouard 10076:
1.227 brouard 10077: /* if (popforecast==1) { */
10078: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
10079: /* printf("Problem with population file : %s\n",popfile);exit(0); */
10080: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
10081: /* } */
10082: /* popage=ivector(0,AGESUP); */
10083: /* popeffectif=vector(0,AGESUP); */
10084: /* popcount=vector(0,AGESUP); */
1.126 brouard 10085:
1.227 brouard 10086: /* i=1; */
10087: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 10088:
1.227 brouard 10089: /* imx=i; */
10090: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
10091: /* } */
1.218 brouard 10092:
1.227 brouard 10093: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
10094: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
10095: /* k=k+1; */
10096: /* fprintf(ficrespop,"\n#******"); */
10097: /* for(j=1;j<=cptcoveff;j++) { */
10098: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
10099: /* } */
10100: /* fprintf(ficrespop,"******\n"); */
10101: /* fprintf(ficrespop,"# Age"); */
10102: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
10103: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 10104:
1.227 brouard 10105: /* for (cpt=0; cpt<=0;cpt++) { */
10106: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 10107:
1.227 brouard 10108: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
10109: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
10110: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 10111:
1.227 brouard 10112: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10113: /* oldm=oldms;savm=savms; */
10114: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 10115:
1.227 brouard 10116: /* for (h=0; h<=nhstepm; h++){ */
10117: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
10118: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
10119: /* } */
10120: /* for(j=1; j<=nlstate+ndeath;j++) { */
10121: /* kk1=0.;kk2=0; */
10122: /* for(i=1; i<=nlstate;i++) { */
10123: /* if (mobilav==1) */
10124: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
10125: /* else { */
10126: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
10127: /* } */
10128: /* } */
10129: /* if (h==(int)(calagedatem+12*cpt)){ */
10130: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
10131: /* /\*fprintf(ficrespop," %.3f", kk1); */
10132: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
10133: /* } */
10134: /* } */
10135: /* for(i=1; i<=nlstate;i++){ */
10136: /* kk1=0.; */
10137: /* for(j=1; j<=nlstate;j++){ */
10138: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
10139: /* } */
10140: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
10141: /* } */
1.218 brouard 10142:
1.227 brouard 10143: /* if (h==(int)(calagedatem+12*cpt)) */
10144: /* for(j=1; j<=nlstate;j++) */
10145: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
10146: /* } */
10147: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10148: /* } */
10149: /* } */
1.218 brouard 10150:
1.227 brouard 10151: /* /\******\/ */
1.218 brouard 10152:
1.227 brouard 10153: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
10154: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
10155: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
10156: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
10157: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 10158:
1.227 brouard 10159: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10160: /* oldm=oldms;savm=savms; */
10161: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
10162: /* for (h=0; h<=nhstepm; h++){ */
10163: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
10164: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
10165: /* } */
10166: /* for(j=1; j<=nlstate+ndeath;j++) { */
10167: /* kk1=0.;kk2=0; */
10168: /* for(i=1; i<=nlstate;i++) { */
10169: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
10170: /* } */
10171: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
10172: /* } */
10173: /* } */
10174: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10175: /* } */
10176: /* } */
10177: /* } */
10178: /* } */
1.218 brouard 10179:
1.227 brouard 10180: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 10181:
1.227 brouard 10182: /* if (popforecast==1) { */
10183: /* free_ivector(popage,0,AGESUP); */
10184: /* free_vector(popeffectif,0,AGESUP); */
10185: /* free_vector(popcount,0,AGESUP); */
10186: /* } */
10187: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10188: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10189: /* fclose(ficrespop); */
10190: /* } /\* End of popforecast *\/ */
1.218 brouard 10191:
1.126 brouard 10192: int fileappend(FILE *fichier, char *optionfich)
10193: {
10194: if((fichier=fopen(optionfich,"a"))==NULL) {
10195: printf("Problem with file: %s\n", optionfich);
10196: fprintf(ficlog,"Problem with file: %s\n", optionfich);
10197: return (0);
10198: }
10199: fflush(fichier);
10200: return (1);
10201: }
10202:
10203:
10204: /**************** function prwizard **********************/
10205: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
10206: {
10207:
10208: /* Wizard to print covariance matrix template */
10209:
1.164 brouard 10210: char ca[32], cb[32];
10211: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 10212: int numlinepar;
10213:
10214: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10215: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10216: for(i=1; i <=nlstate; i++){
10217: jj=0;
10218: for(j=1; j <=nlstate+ndeath; j++){
10219: if(j==i) continue;
10220: jj++;
10221: /*ca[0]= k+'a'-1;ca[1]='\0';*/
10222: printf("%1d%1d",i,j);
10223: fprintf(ficparo,"%1d%1d",i,j);
10224: for(k=1; k<=ncovmodel;k++){
10225: /* printf(" %lf",param[i][j][k]); */
10226: /* fprintf(ficparo," %lf",param[i][j][k]); */
10227: printf(" 0.");
10228: fprintf(ficparo," 0.");
10229: }
10230: printf("\n");
10231: fprintf(ficparo,"\n");
10232: }
10233: }
10234: printf("# Scales (for hessian or gradient estimation)\n");
10235: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
10236: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
10237: for(i=1; i <=nlstate; i++){
10238: jj=0;
10239: for(j=1; j <=nlstate+ndeath; j++){
10240: if(j==i) continue;
10241: jj++;
10242: fprintf(ficparo,"%1d%1d",i,j);
10243: printf("%1d%1d",i,j);
10244: fflush(stdout);
10245: for(k=1; k<=ncovmodel;k++){
10246: /* printf(" %le",delti3[i][j][k]); */
10247: /* fprintf(ficparo," %le",delti3[i][j][k]); */
10248: printf(" 0.");
10249: fprintf(ficparo," 0.");
10250: }
10251: numlinepar++;
10252: printf("\n");
10253: fprintf(ficparo,"\n");
10254: }
10255: }
10256: printf("# Covariance matrix\n");
10257: /* # 121 Var(a12)\n\ */
10258: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10259: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10260: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10261: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10262: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10263: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10264: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10265: fflush(stdout);
10266: fprintf(ficparo,"# Covariance matrix\n");
10267: /* # 121 Var(a12)\n\ */
10268: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10269: /* # ...\n\ */
10270: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10271:
10272: for(itimes=1;itimes<=2;itimes++){
10273: jj=0;
10274: for(i=1; i <=nlstate; i++){
10275: for(j=1; j <=nlstate+ndeath; j++){
10276: if(j==i) continue;
10277: for(k=1; k<=ncovmodel;k++){
10278: jj++;
10279: ca[0]= k+'a'-1;ca[1]='\0';
10280: if(itimes==1){
10281: printf("#%1d%1d%d",i,j,k);
10282: fprintf(ficparo,"#%1d%1d%d",i,j,k);
10283: }else{
10284: printf("%1d%1d%d",i,j,k);
10285: fprintf(ficparo,"%1d%1d%d",i,j,k);
10286: /* printf(" %.5le",matcov[i][j]); */
10287: }
10288: ll=0;
10289: for(li=1;li <=nlstate; li++){
10290: for(lj=1;lj <=nlstate+ndeath; lj++){
10291: if(lj==li) continue;
10292: for(lk=1;lk<=ncovmodel;lk++){
10293: ll++;
10294: if(ll<=jj){
10295: cb[0]= lk +'a'-1;cb[1]='\0';
10296: if(ll<jj){
10297: if(itimes==1){
10298: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10299: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10300: }else{
10301: printf(" 0.");
10302: fprintf(ficparo," 0.");
10303: }
10304: }else{
10305: if(itimes==1){
10306: printf(" Var(%s%1d%1d)",ca,i,j);
10307: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
10308: }else{
10309: printf(" 0.");
10310: fprintf(ficparo," 0.");
10311: }
10312: }
10313: }
10314: } /* end lk */
10315: } /* end lj */
10316: } /* end li */
10317: printf("\n");
10318: fprintf(ficparo,"\n");
10319: numlinepar++;
10320: } /* end k*/
10321: } /*end j */
10322: } /* end i */
10323: } /* end itimes */
10324:
10325: } /* end of prwizard */
10326: /******************* Gompertz Likelihood ******************************/
10327: double gompertz(double x[])
10328: {
1.302 brouard 10329: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 10330: int i,n=0; /* n is the size of the sample */
10331:
1.220 brouard 10332: for (i=1;i<=imx ; i++) {
1.126 brouard 10333: sump=sump+weight[i];
10334: /* sump=sump+1;*/
10335: num=num+1;
10336: }
1.302 brouard 10337: L=0.0;
10338: /* agegomp=AGEGOMP; */
1.126 brouard 10339: /* for (i=0; i<=imx; i++)
10340: if (wav[i]>0) printf("i=%d ageex=%lf agecens=%lf agedc=%lf cens=%d %d\n" ,i,ageexmed[i],agecens[i],agedc[i],cens[i],wav[i]);*/
10341:
1.302 brouard 10342: for (i=1;i<=imx ; i++) {
10343: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
10344: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
10345: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
10346: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
10347: * +
10348: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
10349: */
10350: if (wav[i] > 1 || agedc[i] < AGESUP) {
10351: if (cens[i] == 1){
10352: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
10353: } else if (cens[i] == 0){
1.126 brouard 10354: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 10355: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
10356: } else
10357: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 10358: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 10359: L=L+A*weight[i];
1.126 brouard 10360: /* printf("\ni=%d A=%f L=%lf x[1]=%lf x[2]=%lf ageex=%lf agecens=%lf cens=%d agedc=%lf weight=%lf\n",i,A,L,x[1],x[2],ageexmed[i]*12,agecens[i]*12,cens[i],agedc[i]*12,weight[i]);*/
1.302 brouard 10361: }
10362: }
1.126 brouard 10363:
1.302 brouard 10364: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 10365:
10366: return -2*L*num/sump;
10367: }
10368:
1.136 brouard 10369: #ifdef GSL
10370: /******************* Gompertz_f Likelihood ******************************/
10371: double gompertz_f(const gsl_vector *v, void *params)
10372: {
1.302 brouard 10373: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 10374: double *x= (double *) v->data;
10375: int i,n=0; /* n is the size of the sample */
10376:
10377: for (i=0;i<=imx-1 ; i++) {
10378: sump=sump+weight[i];
10379: /* sump=sump+1;*/
10380: num=num+1;
10381: }
10382:
10383:
10384: /* for (i=0; i<=imx; i++)
10385: if (wav[i]>0) printf("i=%d ageex=%lf agecens=%lf agedc=%lf cens=%d %d\n" ,i,ageexmed[i],agecens[i],agedc[i],cens[i],wav[i]);*/
10386: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
10387: for (i=1;i<=imx ; i++)
10388: {
10389: if (cens[i] == 1 && wav[i]>1)
10390: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
10391:
10392: if (cens[i] == 0 && wav[i]>1)
10393: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
10394: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
10395:
10396: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
10397: if (wav[i] > 1 ) { /* ??? */
10398: LL=LL+A*weight[i];
10399: /* printf("\ni=%d A=%f L=%lf x[1]=%lf x[2]=%lf ageex=%lf agecens=%lf cens=%d agedc=%lf weight=%lf\n",i,A,L,x[1],x[2],ageexmed[i]*12,agecens[i]*12,cens[i],agedc[i]*12,weight[i]);*/
10400: }
10401: }
10402:
10403: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
10404: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
10405:
10406: return -2*LL*num/sump;
10407: }
10408: #endif
10409:
1.126 brouard 10410: /******************* Printing html file ***********/
1.201 brouard 10411: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 10412: int lastpass, int stepm, int weightopt, char model[],\
10413: int imx, double p[],double **matcov,double agemortsup){
10414: int i,k;
10415:
10416: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
10417: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
10418: for (i=1;i<=2;i++)
10419: fprintf(fichtm," p[%d] = %lf [%f ; %f]<br>\n",i,p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.199 brouard 10420: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 10421: fprintf(fichtm,"</ul>");
10422:
10423: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
10424:
10425: fprintf(fichtm,"\nAge l<inf>x</inf> q<inf>x</inf> d(x,x+1) L<inf>x</inf> T<inf>x</inf> e<infx</inf><br>");
10426:
10427: for (k=agegomp;k<(agemortsup-2);k++)
10428: fprintf(fichtm,"%d %.0lf %lf %.0lf %.0lf %.0lf %lf<br>\n",k,lsurv[k],p[1]*exp(p[2]*(k-agegomp)),(p[1]*exp(p[2]*(k-agegomp)))*lsurv[k],lpop[k],tpop[k],tpop[k]/lsurv[k]);
10429:
10430:
10431: fflush(fichtm);
10432: }
10433:
10434: /******************* Gnuplot file **************/
1.201 brouard 10435: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 10436:
10437: char dirfileres[132],optfileres[132];
1.164 brouard 10438:
1.126 brouard 10439: int ng;
10440:
10441:
10442: /*#ifdef windows */
10443: fprintf(ficgp,"cd \"%s\" \n",pathc);
10444: /*#endif */
10445:
10446:
10447: strcpy(dirfileres,optionfilefiname);
10448: strcpy(optfileres,"vpl");
1.199 brouard 10449: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 10450: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 10451: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 10452: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 10453: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
10454:
10455: }
10456:
1.136 brouard 10457: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
10458: {
1.126 brouard 10459:
1.136 brouard 10460: /*-------- data file ----------*/
10461: FILE *fic;
10462: char dummy[]=" ";
1.240 brouard 10463: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 10464: int lstra;
1.136 brouard 10465: int linei, month, year,iout;
1.302 brouard 10466: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 10467: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 10468: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 10469: char *stratrunc;
1.223 brouard 10470:
1.240 brouard 10471: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
10472: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 10473: for(v=1;v<NCOVMAX;v++){
10474: DummyV[v]=0;
10475: FixedV[v]=0;
10476: }
1.126 brouard 10477:
1.240 brouard 10478: for(v=1; v <=ncovcol;v++){
10479: DummyV[v]=0;
10480: FixedV[v]=0;
10481: }
10482: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
10483: DummyV[v]=1;
10484: FixedV[v]=0;
10485: }
10486: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
10487: DummyV[v]=0;
10488: FixedV[v]=1;
10489: }
10490: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
10491: DummyV[v]=1;
10492: FixedV[v]=1;
10493: }
10494: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
10495: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10496: fprintf(ficlog,"Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10497: }
1.339 brouard 10498:
10499: ncovcolt=ncovcol+nqv+ntv+nqtv; /* total of covariates in the data, not in the model equation */
10500:
1.136 brouard 10501: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 10502: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10503: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 10504: }
1.126 brouard 10505:
1.302 brouard 10506: /* Is it a BOM UTF-8 Windows file? */
10507: /* First data line */
10508: linei=0;
10509: while(fgets(line, MAXLINE, fic)) {
10510: noffset=0;
10511: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10512: {
10513: noffset=noffset+3;
10514: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
10515: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
10516: fflush(ficlog); return 1;
10517: }
10518: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
10519: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
10520: {
10521: noffset=noffset+2;
1.304 brouard 10522: printf("# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);fflush(stdout);
10523: fprintf(ficlog,"# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
1.302 brouard 10524: fflush(ficlog); return 1;
10525: }
10526: else if( line[0] == 0 && line[1] == 0)
10527: {
10528: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10529: noffset=noffset+4;
1.304 brouard 10530: printf("# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);fflush(stdout);
10531: fprintf(ficlog,"# Error Data file '%s' is a huge UTF16BE BOM file, please convert to UTF8 or ascii file (for example with dos2unix) and rerun.\n",datafile);
1.302 brouard 10532: fflush(ficlog); return 1;
10533: }
10534: } else{
10535: ;/*printf(" Not a BOM file\n");*/
10536: }
10537: /* If line starts with a # it is a comment */
10538: if (line[noffset] == '#') {
10539: linei=linei+1;
10540: break;
10541: }else{
10542: break;
10543: }
10544: }
10545: fclose(fic);
10546: if((fic=fopen(datafile,"r"))==NULL) {
10547: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10548: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
10549: }
10550: /* Not a Bom file */
10551:
1.136 brouard 10552: i=1;
10553: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
10554: linei=linei+1;
10555: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
10556: if(line[j] == '\t')
10557: line[j] = ' ';
10558: }
10559: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
10560: ;
10561: };
10562: line[j+1]=0; /* Trims blanks at end of line */
10563: if(line[0]=='#'){
10564: fprintf(ficlog,"Comment line\n%s\n",line);
10565: printf("Comment line\n%s\n",line);
10566: continue;
10567: }
10568: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 10569: strcpy(line, linetmp);
1.223 brouard 10570:
10571: /* Loops on waves */
10572: for (j=maxwav;j>=1;j--){
10573: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 10574: cutv(stra, strb, line, ' ');
10575: if(strb[0]=='.') { /* Missing value */
10576: lval=-1;
10577: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
1.341 brouard 10578: cotvar[j][ncovcol+nqv+ntv+iv][i]=-1; /* For performance reasons */
1.238 brouard 10579: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
10580: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. If missing, you should remove this individual or impute a value. Exiting.\n", strb, linei,i,line,iv, nqtv, j);
10581: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. If missing, you should remove this individual or impute a value. Exiting.\n", strb, linei,i,line,iv, nqtv, j);fflush(ficlog);
10582: return 1;
10583: }
10584: }else{
10585: errno=0;
10586: /* what_kind_of_number(strb); */
10587: dval=strtod(strb,&endptr);
10588: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
10589: /* if(strb != endptr && *endptr == '\0') */
10590: /* dval=dlval; */
10591: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10592: if( strb[0]=='\0' || (*endptr != '\0')){
10593: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,iv, nqtv, j,maxwav);
10594: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value out of %d measured at wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line, iv, nqtv, j,maxwav);fflush(ficlog);
10595: return 1;
10596: }
10597: cotqvar[j][iv][i]=dval;
1.341 brouard 10598: cotvar[j][ncovcol+nqv+ntv+iv][i]=dval; /* because cotvar starts now at first ntv */
1.238 brouard 10599: }
10600: strcpy(line,stra);
1.223 brouard 10601: }/* end loop ntqv */
1.225 brouard 10602:
1.223 brouard 10603: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 10604: cutv(stra, strb, line, ' ');
10605: if(strb[0]=='.') { /* Missing value */
10606: lval=-1;
10607: }else{
10608: errno=0;
10609: lval=strtol(strb,&endptr,10);
10610: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10611: if( strb[0]=='\0' || (*endptr != '\0')){
10612: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th dummy covariate out of %d measured at wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,iv, ntv, j,maxwav);
10613: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d dummy covariate out of %d measured wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,iv, ntv,j,maxwav);fflush(ficlog);
10614: return 1;
10615: }
10616: }
10617: if(lval <-1 || lval >1){
10618: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10619: Should be a value of %d(nth) covariate of wave %d (0 should be the value for the reference and 1\n \
1.223 brouard 10620: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10621: For example, for multinomial values like 1, 2 and 3,\n \
10622: build V1=0 V2=0 for the reference value (1),\n \
10623: V1=1 V2=0 for (2) \n \
1.223 brouard 10624: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10625: output of IMaCh is often meaningless.\n \
1.319 brouard 10626: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 10627: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10628: Should be a value of %d(nth) covariate of wave %d (0 should be the value for the reference and 1\n \
1.223 brouard 10629: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10630: For example, for multinomial values like 1, 2 and 3,\n \
10631: build V1=0 V2=0 for the reference value (1),\n \
10632: V1=1 V2=0 for (2) \n \
1.223 brouard 10633: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10634: output of IMaCh is often meaningless.\n \
1.319 brouard 10635: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 10636: return 1;
10637: }
1.341 brouard 10638: cotvar[j][ncovcol+nqv+iv][i]=(double)(lval);
1.238 brouard 10639: strcpy(line,stra);
1.223 brouard 10640: }/* end loop ntv */
1.225 brouard 10641:
1.223 brouard 10642: /* Statuses at wave */
1.137 brouard 10643: cutv(stra, strb, line, ' ');
1.223 brouard 10644: if(strb[0]=='.') { /* Missing value */
1.238 brouard 10645: lval=-1;
1.136 brouard 10646: }else{
1.238 brouard 10647: errno=0;
10648: lval=strtol(strb,&endptr,10);
10649: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10650: if( strb[0]=='\0' || (*endptr != '\0')){
10651: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a status of wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,j,maxwav);
10652: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a status of wave %d. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line,j,maxwav);fflush(ficlog);
10653: return 1;
10654: }
1.136 brouard 10655: }
1.225 brouard 10656:
1.136 brouard 10657: s[j][i]=lval;
1.225 brouard 10658:
1.223 brouard 10659: /* Date of Interview */
1.136 brouard 10660: strcpy(line,stra);
10661: cutv(stra, strb,line,' ');
1.169 brouard 10662: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10663: }
1.169 brouard 10664: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 10665: month=99;
10666: year=9999;
1.136 brouard 10667: }else{
1.225 brouard 10668: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of interview (mm/yyyy or .) at wave %d. Exiting.\n",strb, linei,i, line,j);
10669: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of interview (mm/yyyy or .) at wave %d. Exiting.\n",strb, linei,i, line,j);fflush(ficlog);
10670: return 1;
1.136 brouard 10671: }
10672: anint[j][i]= (double) year;
1.302 brouard 10673: mint[j][i]= (double)month;
10674: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
10675: /* printf("Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, mint[j][i],anint[j][i], moisnais[i],annais[i]); */
10676: /* fprintf(ficlog,"Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, mint[j][i],anint[j][i], moisnais[i],annais[i]); */
10677: /* } */
1.136 brouard 10678: strcpy(line,stra);
1.223 brouard 10679: } /* End loop on waves */
1.225 brouard 10680:
1.223 brouard 10681: /* Date of death */
1.136 brouard 10682: cutv(stra, strb,line,' ');
1.169 brouard 10683: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10684: }
1.169 brouard 10685: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 10686: month=99;
10687: year=9999;
10688: }else{
1.141 brouard 10689: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of death (mm/yyyy or .). Exiting.\n",strb, linei,i,line);
1.225 brouard 10690: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of death (mm/yyyy or .). Exiting.\n",strb, linei,i,line);fflush(ficlog);
10691: return 1;
1.136 brouard 10692: }
10693: andc[i]=(double) year;
10694: moisdc[i]=(double) month;
10695: strcpy(line,stra);
10696:
1.223 brouard 10697: /* Date of birth */
1.136 brouard 10698: cutv(stra, strb,line,' ');
1.169 brouard 10699: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10700: }
1.169 brouard 10701: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 10702: month=99;
10703: year=9999;
10704: }else{
1.141 brouard 10705: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy or .). Exiting.\n",strb, linei,i,line);
10706: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy or .). Exiting.\n",strb, linei,i,line);fflush(ficlog);
1.225 brouard 10707: return 1;
1.136 brouard 10708: }
10709: if (year==9999) {
1.141 brouard 10710: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy) but at least the year of birth should be given. Exiting.\n",strb, linei,i,line);
10711: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be a date of birth (mm/yyyy) but at least the year of birth should be given. Exiting.\n",strb, linei,i,line);fflush(ficlog);
1.225 brouard 10712: return 1;
10713:
1.136 brouard 10714: }
10715: annais[i]=(double)(year);
1.302 brouard 10716: moisnais[i]=(double)(month);
10717: for (j=1;j<=maxwav;j++){
10718: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
10719: printf("Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, (int)mint[j][i],(int)anint[j][i], j,(int)moisnais[i],(int)annais[i]);
10720: fprintf(ficlog,"Warning reading data around '%s' at line number %d for individual %d, '%s'\nThe date of interview (%2d/%4d) at wave %d occurred before the date of birth (%2d/%4d).\n",strb, linei,i, line, (int)mint[j][i],(int)anint[j][i], j, (int)moisnais[i],(int)annais[i]);
10721: }
10722: }
10723:
1.136 brouard 10724: strcpy(line,stra);
1.225 brouard 10725:
1.223 brouard 10726: /* Sample weight */
1.136 brouard 10727: cutv(stra, strb,line,' ');
10728: errno=0;
10729: dval=strtod(strb,&endptr);
10730: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 10731: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
10732: fprintf(ficlog,"Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
1.136 brouard 10733: fflush(ficlog);
10734: return 1;
10735: }
10736: weight[i]=dval;
10737: strcpy(line,stra);
1.225 brouard 10738:
1.223 brouard 10739: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10740: cutv(stra, strb, line, ' ');
10741: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10742: lval=-1;
1.311 brouard 10743: coqvar[iv][i]=NAN;
10744: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10745: }else{
1.225 brouard 10746: errno=0;
10747: /* what_kind_of_number(strb); */
10748: dval=strtod(strb,&endptr);
10749: /* if(strb != endptr && *endptr == '\0') */
10750: /* dval=dlval; */
10751: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10752: if( strb[0]=='\0' || (*endptr != '\0')){
10753: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value (out of %d) constant for all waves. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line, iv, nqv, maxwav);
10754: fprintf(ficlog,"Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th quantitative value (out of %d) constant for all waves. Setting maxwav=%d might be wrong. Exiting.\n", strb, linei,i,line, iv, nqv, maxwav);fflush(ficlog);
10755: return 1;
10756: }
10757: coqvar[iv][i]=dval;
1.226 brouard 10758: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10759: }
10760: strcpy(line,stra);
10761: }/* end loop nqv */
1.136 brouard 10762:
1.223 brouard 10763: /* Covariate values */
1.136 brouard 10764: for (j=ncovcol;j>=1;j--){
10765: cutv(stra, strb,line,' ');
1.223 brouard 10766: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10767: lval=-1;
1.136 brouard 10768: }else{
1.225 brouard 10769: errno=0;
10770: lval=strtol(strb,&endptr,10);
10771: if( strb[0]=='\0' || (*endptr != '\0')){
10772: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\nShould be a covariate value (=0 for the reference or 1 for alternative). Exiting.\n",lval, linei,i, line);
10773: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\nShould be a covariate value (=0 for the reference or 1 for alternative). Exiting.\n",lval, linei,i, line);fflush(ficlog);
10774: return 1;
10775: }
1.136 brouard 10776: }
10777: if(lval <-1 || lval >1){
1.225 brouard 10778: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10779: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10780: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10781: For example, for multinomial values like 1, 2 and 3,\n \
10782: build V1=0 V2=0 for the reference value (1),\n \
10783: V1=1 V2=0 for (2) \n \
1.136 brouard 10784: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10785: output of IMaCh is often meaningless.\n \
1.136 brouard 10786: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10787: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10788: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10789: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10790: For example, for multinomial values like 1, 2 and 3,\n \
10791: build V1=0 V2=0 for the reference value (1),\n \
10792: V1=1 V2=0 for (2) \n \
1.136 brouard 10793: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10794: output of IMaCh is often meaningless.\n \
1.136 brouard 10795: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10796: return 1;
1.136 brouard 10797: }
10798: covar[j][i]=(double)(lval);
10799: strcpy(line,stra);
10800: }
10801: lstra=strlen(stra);
1.225 brouard 10802:
1.136 brouard 10803: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10804: stratrunc = &(stra[lstra-9]);
10805: num[i]=atol(stratrunc);
10806: }
10807: else
10808: num[i]=atol(stra);
10809: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10810: printf("%ld %.lf %.lf %.lf %.lf/%.lf %.lf/%.lf %.lf/%.lf %d %.lf/%.lf %d %.lf/%.lf %d %.lf/%.lf %d\n",num[i],(covar[1][i]), (covar[2][i]),weight[i], (moisnais[i]), (annais[i]), (moisdc[i]), (andc[i]), (mint[1][i]), (anint[1][i]), (s[1][i]), (mint[2][i]), (anint[2][i]), (s[2][i]), (mint[3][i]), (anint[3][i]), (s[3][i]), (mint[4][i]), (anint[4][i]), (s[4][i])); ij=ij+1;}*/
10811:
10812: i=i+1;
10813: } /* End loop reading data */
1.225 brouard 10814:
1.136 brouard 10815: *imax=i-1; /* Number of individuals */
10816: fclose(fic);
1.225 brouard 10817:
1.136 brouard 10818: return (0);
1.164 brouard 10819: /* endread: */
1.225 brouard 10820: printf("Exiting readdata: ");
10821: fclose(fic);
10822: return (1);
1.223 brouard 10823: }
1.126 brouard 10824:
1.234 brouard 10825: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10826: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10827: while (*p2 == ' ')
1.234 brouard 10828: p2++;
10829: /* while ((*p1++ = *p2++) !=0) */
10830: /* ; */
10831: /* do */
10832: /* while (*p2 == ' ') */
10833: /* p2++; */
10834: /* while (*p1++ == *p2++); */
10835: *stri=p2;
1.145 brouard 10836: }
10837:
1.330 brouard 10838: int decoderesult( char resultline[], int nres)
1.230 brouard 10839: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10840: {
1.235 brouard 10841: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10842: char resultsav[MAXLINE];
1.330 brouard 10843: /* int resultmodel[MAXLINE]; */
1.334 brouard 10844: /* int modelresult[MAXLINE]; */
1.230 brouard 10845: char stra[80], strb[80], strc[80], strd[80],stre[80];
10846:
1.234 brouard 10847: removefirstspace(&resultline);
1.332 brouard 10848: printf("decoderesult:%s\n",resultline);
1.230 brouard 10849:
1.332 brouard 10850: strcpy(resultsav,resultline);
1.342 brouard 10851: /* printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline); */
1.230 brouard 10852: if (strlen(resultsav) >1){
1.334 brouard 10853: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 10854: }
1.253 brouard 10855: if(j == 0){ /* Resultline but no = */
10856: TKresult[nres]=0; /* Combination for the nresult and the model */
10857: return (0);
10858: }
1.234 brouard 10859: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.334 brouard 10860: printf("ERROR: the number of variables in the resultline which is %d, differs from the number %d of single variables used in the model line, %s.\n",j, cptcovs, model);
10861: fprintf(ficlog,"ERROR: the number of variables in the resultline which is %d, differs from the number %d of single variables used in the model line, %s.\n",j, cptcovs, model);
1.332 brouard 10862: /* return 1;*/
1.234 brouard 10863: }
1.334 brouard 10864: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 10865: if(nbocc(resultsav,'=') >1){
1.318 brouard 10866: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' ' (stra is the rest of the resultline to be analyzed in the next loop *//* resultsav= "V4=1 V5=25.1 V3=0" stra= "V5=25.1 V3=0" strb= "V4=1" */
1.332 brouard 10867: /* If resultsav= "V4= 1 V5=25.1 V3=0" with a blank then strb="V4=" and stra="1 V5=25.1 V3=0" */
1.318 brouard 10868: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 10869: /* If a blank, then strc="V4=" and strd='\0' */
10870: if(strc[0]=='\0'){
10871: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
10872: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
10873: return 1;
10874: }
1.234 brouard 10875: }else
10876: cutl(strc,strd,resultsav,'=');
1.318 brouard 10877: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10878:
1.230 brouard 10879: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10880: Tvarsel[k]=atoi(strc); /* 4 */ /* Tvarsel is the id of the kth covariate in the result line Tvarsel[1] in "V4=1.." is 4.*/
1.230 brouard 10881: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10882: /* cptcovsel++; */
10883: if (nbocc(stra,'=') >0)
10884: strcpy(resultsav,stra); /* and analyzes it */
10885: }
1.235 brouard 10886: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10887: /* Feeds resultmodel[nres][k1]=k2 for k1th product covariate with age in the model equation fed by the index k2 of the resutline*/
1.334 brouard 10888: for(k1=1; k1<= cptcovt ;k1++){ /* Loop on MODEL LINE V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.332 brouard 10889: if(Typevar[k1]==0){ /* Single covariate in model */
10890: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 10891: match=0;
1.318 brouard 10892: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10893: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10894: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10895: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10896: break;
10897: }
10898: }
10899: if(match == 0){
1.338 brouard 10900: printf("Error in result line (Dummy single): V%d is missing in result: %s according to model=1+age+%s. Tvar[k1=%d]=%d is different from Tvarsel[k2=%d]=%d.\n",Tvar[k1], resultline, model,k1, Tvar[k1], k2, Tvarsel[k2]);
10901: fprintf(ficlog,"Error in result line (Dummy single): V%d is missing in result: %s according to model=1+age+%s\n",Tvar[k1], resultline, model);
1.310 brouard 10902: return 1;
1.234 brouard 10903: }
1.332 brouard 10904: }else if(Typevar[k1]==1){ /* Product with age We want to get the position k2 in the resultline of the product k1 in the model line*/
10905: /* We feed resultmodel[k1]=k2; */
10906: match=0;
10907: for(k2=1; k2 <=j;k2++){/* Loop on resultline. jth occurence of = signs in the result line. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10908: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10909: modelresult[nres][k2]=k1;/* we found a Vn=1 corrresponding to Vn*age in the model modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.332 brouard 10910: resultmodel[nres][k1]=k2; /* Added here */
1.342 brouard 10911: /* printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]); */
1.332 brouard 10912: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10913: break;
10914: }
10915: }
10916: if(match == 0){
1.338 brouard 10917: printf("Error in result line (Product with age): V%d is missing in result: %s according to model=1+age+%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
10918: fprintf(ficlog,"Error in result line (Product with age): V%d is missing in result: %s according to model=1+age+%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
1.332 brouard 10919: return 1;
10920: }
10921: }else if(Typevar[k1]==2){ /* Product No age We want to get the position in the resultline of the product in the model line*/
10922: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
10923: match=0;
1.342 brouard 10924: /* printf("Decoderesult very first Product Tvardk[k1=%d][1]=%d Tvardk[k1=%d][2]=%d V%d * V%d\n",k1,Tvardk[k1][1],k1,Tvardk[k1][2],Tvardk[k1][1],Tvardk[k1][2]); */
1.332 brouard 10925: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10926: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10927: /* modelresult[k2]=k1; */
1.342 brouard 10928: /* printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]); */
1.332 brouard 10929: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10930: }
10931: }
10932: if(match == 0){
1.338 brouard 10933: printf("Error in result line (Product without age first variable): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][1], resultline, model);
10934: fprintf(ficlog,"Error in result line (Product without age first variable): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][1], resultline, model);
1.332 brouard 10935: return 1;
10936: }
10937: match=0;
10938: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10939: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10940: /* modelresult[k2]=k1;*/
1.342 brouard 10941: /* printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]); */
1.332 brouard 10942: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10943: break;
10944: }
10945: }
10946: if(match == 0){
1.338 brouard 10947: printf("Error in result line (Product without age second variable): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][2], resultline, model);
10948: fprintf(ficlog,"Error in result line (Product without age second variable): V%d is missing in result : %s according to model=1+age+%s\n",Tvardk[k1][2], resultline, model);
1.332 brouard 10949: return 1;
10950: }
10951: }/* End of testing */
1.333 brouard 10952: }/* End loop cptcovt */
1.235 brouard 10953: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10954: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 10955: for(k2=1; k2 <=j;k2++){ /* j or cptcovs is the number of single covariates used either in the model line as well as in the result line (dummy or quantitative)
10956: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 10957: match=0;
1.318 brouard 10958: for(k1=1; k1<= cptcovt ;k1++){ /* loop on model: model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.332 brouard 10959: if(Typevar[k1]==0){ /* Single only */
1.237 brouard 10960: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 brouard 10961: resultmodel[nres][k1]=k2; /* k1th position in the model equation corresponds to k2th position in the result line. resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.334 brouard 10962: modelresult[nres][k2]=k1; /* k1th position in the model equation corresponds to k2th position in the result line. modelresult[1]=2 modelresult[2]=1 modelresult[3]=3 remodelresult[4]=6 modelresult[5]=9 */
1.234 brouard 10963: ++match;
10964: }
10965: }
10966: }
10967: if(match == 0){
1.338 brouard 10968: printf("Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
10969: fprintf(ficlog,"Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
1.310 brouard 10970: return 1;
1.234 brouard 10971: }else if(match > 1){
1.338 brouard 10972: printf("Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
10973: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
1.310 brouard 10974: return 1;
1.234 brouard 10975: }
10976: }
1.334 brouard 10977: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 10978: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10979: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 10980: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10981: /* should correspond to the combination 6 of dummy: V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 1*1 + 0*2 + 1*4 = 5 + (1offset) = 6*/
10982: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 10983: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10984: /* 1 0 0 0 */
10985: /* 2 1 0 0 */
10986: /* 3 0 1 0 */
1.330 brouard 10987: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 10988: /* 5 0 0 1 */
1.330 brouard 10989: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 10990: /* 7 0 1 1 */
10991: /* 8 1 1 1 */
1.237 brouard 10992: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10993: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10994: /* V5*age V5 known which value for nres? */
10995: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 10996: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* cptcovt number of covariates (excluding 1 and age or age*age) in the MODEL equation.
10997: * loop on position k1 in the MODEL LINE */
1.331 brouard 10998: /* k counting number of combination of single dummies in the equation model */
10999: /* k4 counting single dummies in the equation model */
11000: /* k4q counting single quantitatives in the equation model */
1.344 brouard 11001: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Dummy and Single, fixed or timevarying, k1 is sorting according to MODEL, but k3 to resultline */
1.334 brouard 11002: /* k4+1= (not always if quant in model) position in the resultline V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) */
1.332 brouard 11003: /* modelresult[k3]=k1: k3th position in the result line corresponds to the k1 position in the model line (doesn't work with products)*/
1.330 brouard 11004: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 11005: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
11006: /* k3 is the position in the nres result line of the k1th variable of the model equation */
11007: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
11008: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
11009: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 11010: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 11011: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 11012: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 11013: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
11014: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
11015: k2=(int)Tvarsel[k3]; /* from position k3 in resultline get name k2: nres=1 k1=2=>k3=1 Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 (V4); k1=3=>k3=2 Tvarsel[2]=3 (V3)*/
1.330 brouard 11016: k+=Tvalsel[k3]*pow(2,k4); /* nres=1 k1=2 Tvalsel[1]=1 (V4=1); k1=3 k3=2 Tvalsel[2]=0 (V3=0) */
1.334 brouard 11017: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 11018: /* Tinvresult[nres][4]=1 */
1.334 brouard 11019: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
11020: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
11021: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
11022: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 11023: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 11024: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.342 brouard 11025: /* printf("Decoderesult Dummy k=%d, k1=%d precov[nres=%d][k1=%d]=%.f V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k1, nres, k1,precov[nres][k1], k2, k3, (int)Tvalsel[k3], k4); */
1.235 brouard 11026: k4++;;
1.331 brouard 11027: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 11028: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 11029: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 11030: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 11031: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
11032: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
11033: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 11034: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
11035: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
11036: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
11037: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
11038: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
11039: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 11040: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 11041: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 11042: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 11043: /* printf("Decoderesult Quantitative nres=%d,precov[nres=%d][k1=%d]=%.f V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, nres, k1,precov[nres][k1], k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]); */
1.235 brouard 11044: k4q++;;
1.331 brouard 11045: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
11046: /* Tvar[k1]; */ /* Age variable */
1.332 brouard 11047: /* Wrong we want the value of variable name Tvar[k1] */
11048:
11049: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
1.331 brouard 11050: k2=(int)Tvarsel[k3]; /* nres=1 k1=2=>k3=1 Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 (V4); k1=3=>k3=2 Tvarsel[2]=3 (V3)*/
1.334 brouard 11051: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
1.332 brouard 11052: precov[nres][k1]=Tvalsel[k3];
1.342 brouard 11053: /* printf("Decoderesult Dummy with age k=%d, k1=%d precov[nres=%d][k1=%d]=%.f Tvar[%d]=V%d k2=Tvarsel[%d]=%d Tvalsel[%d]=%d\n",k, k1, nres, k1,precov[nres][k1], k1, Tvar[k1], k3,(int)Tvarsel[k3], k3, (int)Tvalsel[k3]); */
1.331 brouard 11054: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.332 brouard 11055: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.331 brouard 11056: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 11057: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
1.332 brouard 11058: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 11059: /* printf("Decoderesult Quantitative with age nres=%d, k1=%d, precov[nres=%d][k1=%d]=%f Tvar[%d]=V%d V(k2q=%d)= Tvarsel[%d]=%d, Tvalsel[%d]=%f\n",nres, k1, nres, k1,precov[nres][k1], k1, Tvar[k1], k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]); */
1.331 brouard 11060: }else if(Typevar[k1]==2 ){ /* For product quant or dummy (not with age) */
1.332 brouard 11061: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
1.342 brouard 11062: /* printf("Decoderesult Quantitative or Dummy (not with age) nres=%d k1=%d precov[nres=%d][k1=%d]=%.f V%d(=%.f) * V%d(=%.f) \n",nres, k1, nres, k1,precov[nres][k1], Tvardk[k1][1], TinvDoQresult[nres][Tvardk[k1][1]], Tvardk[k1][2], TinvDoQresult[nres][Tvardk[k1][2]]); */
1.330 brouard 11063: }else{
1.332 brouard 11064: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
11065: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 11066: }
11067: }
1.234 brouard 11068:
1.334 brouard 11069: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 11070: return (0);
11071: }
1.235 brouard 11072:
1.230 brouard 11073: int decodemodel( char model[], int lastobs)
11074: /**< This routine decodes the model and returns:
1.224 brouard 11075: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
11076: * - nagesqr = 1 if age*age in the model, otherwise 0.
11077: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
11078: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
11079: * - cptcovage number of covariates with age*products =2
11080: * - cptcovs number of simple covariates
1.339 brouard 11081: * ncovcolt=ncovcol+nqv+ntv+nqtv total of covariates in the data, not in the model equation
1.224 brouard 11082: * - Tvar[k] is the id of the kth covariate Tvar[1]@12 {1, 2, 3, 8, 10, 11, 8, 3, 7, 8, 5, 6}, thus Tvar[5=V7*V8]=10
1.339 brouard 11083: * which is a new column after the 9 (ncovcol+nqv+ntv+nqtv) variables.
1.319 brouard 11084: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 11085: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
11086: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
11087: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
11088: */
1.319 brouard 11089: /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 */
1.136 brouard 11090: {
1.238 brouard 11091: int i, j, k, ks, v;
1.227 brouard 11092: int j1, k1, k2, k3, k4;
1.136 brouard 11093: char modelsav[80];
1.145 brouard 11094: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 11095: char *strpt;
1.136 brouard 11096:
1.145 brouard 11097: /*removespace(model);*/
1.136 brouard 11098: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 11099: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 11100: if (strstr(model,"AGE") !=0){
1.192 brouard 11101: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
11102: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 11103: return 1;
11104: }
1.141 brouard 11105: if (strstr(model,"v") !=0){
1.338 brouard 11106: printf("Error. 'v' must be in upper case 'V' model=1+age+%s ",model);
11107: fprintf(ficlog,"Error. 'v' must be in upper case model=1+age+%s ",model);fflush(ficlog);
1.141 brouard 11108: return 1;
11109: }
1.187 brouard 11110: strcpy(modelsav,model);
11111: if ((strpt=strstr(model,"age*age")) !=0){
1.338 brouard 11112: printf(" strpt=%s, model=1+age+%s\n",strpt, model);
1.187 brouard 11113: if(strpt != model){
1.338 brouard 11114: printf("Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 11115: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 11116: corresponding column of parameters.\n",model);
1.338 brouard 11117: fprintf(ficlog,"Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 11118: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 11119: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 11120: return 1;
1.225 brouard 11121: }
1.187 brouard 11122: nagesqr=1;
11123: if (strstr(model,"+age*age") !=0)
1.234 brouard 11124: substrchaine(modelsav, model, "+age*age");
1.187 brouard 11125: else if (strstr(model,"age*age+") !=0)
1.234 brouard 11126: substrchaine(modelsav, model, "age*age+");
1.187 brouard 11127: else
1.234 brouard 11128: substrchaine(modelsav, model, "age*age");
1.187 brouard 11129: }else
11130: nagesqr=0;
11131: if (strlen(modelsav) >1){
11132: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
11133: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 11134: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 11135: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 11136: * cst, age and age*age
11137: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
11138: /* including age products which are counted in cptcovage.
11139: * but the covariates which are products must be treated
11140: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 11141: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
11142: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 11143:
11144:
1.187 brouard 11145: /* Design
11146: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
11147: * < ncovcol=8 >
11148: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
11149: * k= 1 2 3 4 5 6 7 8
11150: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
1.345 brouard 11151: * covar[k,i], are for fixed covariates, value of kth covariate if not including age for individual i:
1.224 brouard 11152: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
11153: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 11154: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
11155: * Tage[++cptcovage]=k
1.345 brouard 11156: * if products, new covar are created after ncovcol + nqv (quanti fixed) with k1
1.187 brouard 11157: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
11158: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
11159: * Tvard[k1][1]=m Tvard[k1][2]=m; Tvard[1][1]=5 (V5) Tvard[1][2]=6 Tvard[2][1]=7 (V7) Tvard[2][2]=8
11160: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
11161: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
11162: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
1.345 brouard 11163: * < ncovcol=8 8 fixed covariate. Additional starts at 9 (V5*V6) and 10(V7*V8) >
1.187 brouard 11164: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
11165: * k= 1 2 3 4 5 6 7 8 9 10 11 12
1.345 brouard 11166: * Tvard[k]= 2 1 3 3 10 11 8 8 5 6 7 8
11167: * p Tvar[1]@12={2, 1, 3, 3, 9, 10, 8, 8}
1.187 brouard 11168: * p Tprod[1]@2={ 6, 5}
11169: *p Tvard[1][1]@4= {7, 8, 5, 6}
11170: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
11171: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 11172: *How to reorganize? Tvars(orted)
1.187 brouard 11173: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
11174: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
11175: * {2, 1, 4, 8, 5, 6, 3, 7}
11176: * Struct []
11177: */
1.225 brouard 11178:
1.187 brouard 11179: /* This loop fills the array Tvar from the string 'model'.*/
11180: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
11181: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
11182: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
11183: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
11184: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
11185: /* k=1 Tvar[1]=2 (from V2) */
11186: /* k=5 Tvar[5] */
11187: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 11188: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 11189: /* } */
1.198 brouard 11190: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 11191: /*
11192: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 11193: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
11194: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
11195: }
1.187 brouard 11196: cptcovage=0;
1.319 brouard 11197: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
11198: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
11199: modelsav==V2+V1+V5*age+V4+V3*age strb=V3*age stra=V2+V1V5*age+V4 */ /* <model> "V5+V4+V3+V4*V3+V5*age+V1*age+V1" strb="V5" stra="V4+V3+V4*V3+V5*age+V1*age+V1" */
11200: if (nbocc(modelsav,'+')==0)
11201: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 11202: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
11203: /*scanf("%d",i);*/
1.319 brouard 11204: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
11205: cutl(strc,strd,strb,'*'); /**< k=1 strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
1.234 brouard 11206: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
11207: /* covar is not filled and then is empty */
11208: cptcovprod--;
11209: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 11210: Tvar[k]=atoi(stre); /* V2+V1+V5*age+V4+V3*age Tvar[5]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
1.234 brouard 11211: Typevar[k]=1; /* 1 for age product */
1.319 brouard 11212: cptcovage++; /* Counts the number of covariates which include age as a product */
11213: Tage[cptcovage]=k; /* V2+V1+V4+V3*age Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
1.234 brouard 11214: /*printf("stre=%s ", stre);*/
11215: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
11216: cptcovprod--;
11217: cutl(stre,strb,strc,'V');
11218: Tvar[k]=atoi(stre);
11219: Typevar[k]=1; /* 1 for age product */
11220: cptcovage++;
11221: Tage[cptcovage]=k;
11222: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
11223: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
11224: cptcovn++;
11225: cptcovprodnoage++;k1++;
11226: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
1.339 brouard 11227: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* ncovcolt+k1; For model-covariate k tells which data-covariate to use but
1.234 brouard 11228: because this model-covariate is a construction we invent a new column
11229: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.335 brouard 11230: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
1.319 brouard 11231: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
1.339 brouard 11232: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=3 etc */
1.335 brouard 11233: /* Please remark that the new variables are model dependent */
11234: /* If we have 4 variable but the model uses only 3, like in
11235: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
11236: * k= 1 2 3 4 5 6 7 8
11237: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
11238: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
11239: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
11240: */
1.339 brouard 11241: Typevar[k]=2; /* 2 for product */
1.234 brouard 11242: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
11243: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 11244: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 11245: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 brouard 11246: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 11247: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 brouard 11248: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 11249: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
11250: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
11251: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 11252: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 11253: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
1.339 brouard 11254: if( FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){ /* If the product is a fixed covariate then we feed the new column with Vn*Vm */
11255: for (i=1; i<=lastobs;i++){/* For fixed product */
1.234 brouard 11256: /* Computes the new covariate which is a product of
11257: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
1.339 brouard 11258: covar[ncovcolt+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
11259: }
11260: } /*End of FixedV */
1.234 brouard 11261: } /* End age is not in the model */
11262: } /* End if model includes a product */
1.319 brouard 11263: else { /* not a product */
1.234 brouard 11264: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
11265: /* scanf("%d",i);*/
11266: cutl(strd,strc,strb,'V');
11267: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
11268: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
11269: Tvar[k]=atoi(strd);
11270: Typevar[k]=0; /* 0 for simple covariates */
11271: }
11272: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 11273: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 11274: scanf("%d",i);*/
1.187 brouard 11275: } /* end of loop + on total covariates */
11276: } /* end if strlen(modelsave == 0) age*age might exist */
11277: } /* end if strlen(model == 0) */
1.136 brouard 11278:
11279: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
11280: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 11281:
1.136 brouard 11282: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 11283: printf("cptcovprod=%d ", cptcovprod);
11284: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
11285: scanf("%d ",i);*/
11286:
11287:
1.230 brouard 11288: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
11289: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 11290: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
11291: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
11292: k = 1 2 3 4 5 6 7 8 9
11293: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 11294: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 11295: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
11296: Dummy[k] 1 0 0 0 3 1 1 2 3
11297: Tmodelind[combination of covar]=k;
1.225 brouard 11298: */
11299: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 11300: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 11301: /* Tvar[k] is the value n of Vn with n varying for 1 to nvcol, or p Vp=Vn*Vm for product */
1.226 brouard 11302: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 11303: printf("Model=1+age+%s\n\
1.227 brouard 11304: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11305: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11306: Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product\n",model);
1.318 brouard 11307: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 11308: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11309: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11310: Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product\n",model);
1.342 brouard 11311: for(k=-1;k<=NCOVMAX; k++){ Fixed[k]=0; Dummy[k]=0;}
11312: for(k=1;k<=NCOVMAX; k++){TvarFind[k]=0; TvarVind[k]=0;}
1.343 brouard 11313: for(k=1, ncovf=0, nsd=0, nsq=0, ncovv=0, ncova=0, ncoveff=0, nqfveff=0, ntveff=0, nqtveff=0, ncovvt=0;k<=cptcovt; k++){ /* or cptocvt loop on k from model */
1.234 brouard 11314: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 11315: Fixed[k]= 0;
11316: Dummy[k]= 0;
1.225 brouard 11317: ncoveff++;
1.232 brouard 11318: ncovf++;
1.234 brouard 11319: nsd++;
11320: modell[k].maintype= FTYPE;
11321: TvarsD[nsd]=Tvar[k];
11322: TvarsDind[nsd]=k;
1.330 brouard 11323: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 11324: TvarF[ncovf]=Tvar[k];
11325: TvarFind[ncovf]=k;
11326: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
11327: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.339 brouard 11328: /* }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /\* Product of fixed dummy (<=ncovcol) covariates For a fixed product k is higher than ncovcol *\/ */
11329: }else if( Tposprod[k]>0 && Typevar[k]==2 && FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){ /* Needs a fixed product Product of fixed dummy (<=ncovcol) covariates For a fixed product k is higher than ncovcol */
1.234 brouard 11330: Fixed[k]= 0;
11331: Dummy[k]= 0;
11332: ncoveff++;
11333: ncovf++;
11334: modell[k].maintype= FTYPE;
11335: TvarF[ncovf]=Tvar[k];
1.330 brouard 11336: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 11337: TvarFind[ncovf]=k;
1.230 brouard 11338: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 11339: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 11340: }else if( Tvar[k] <=ncovcol+nqv && Typevar[k]==0){/* Remind that product Vn*Vm are added in k Only simple fixed quantitative variable */
1.227 brouard 11341: Fixed[k]= 0;
11342: Dummy[k]= 1;
1.230 brouard 11343: nqfveff++;
1.234 brouard 11344: modell[k].maintype= FTYPE;
11345: modell[k].subtype= FQ;
11346: nsq++;
1.334 brouard 11347: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
11348: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 11349: ncovf++;
1.234 brouard 11350: TvarF[ncovf]=Tvar[k];
11351: TvarFind[ncovf]=k;
1.231 brouard 11352: TvarFQ[nqfveff]=Tvar[k]-ncovcol; /* TvarFQ[1]=V2-1=1st in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1.230 brouard 11353: TvarFQind[nqfveff]=k; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1.242 brouard 11354: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.339 brouard 11355: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11356: /* model V1+V3+age*V1+age*V3+V1*V3 */
11357: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11358: ncovvt++;
11359: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11360: TvarVVind[ncovvt]=k; /* TvarVVind[1]=2 (second position in the model equation */
11361:
1.227 brouard 11362: Fixed[k]= 1;
11363: Dummy[k]= 0;
1.225 brouard 11364: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 11365: modell[k].maintype= VTYPE;
11366: modell[k].subtype= VD;
11367: nsd++;
11368: TvarsD[nsd]=Tvar[k];
11369: TvarsDind[nsd]=k;
1.330 brouard 11370: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 11371: ncovv++; /* Only simple time varying variables */
11372: TvarV[ncovv]=Tvar[k];
1.242 brouard 11373: TvarVind[ncovv]=k; /* TvarVind[2]=2 TvarVind[3]=3 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Any time varying singele */
1.231 brouard 11374: TvarVD[ntveff]=Tvar[k]; /* TvarVD[1]=V4 TvarVD[2]=V3 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying dummy variable */
11375: TvarVDind[ntveff]=k; /* TvarVDind[1]=2 TvarVDind[2]=3 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying dummy variable */
1.228 brouard 11376: printf("Quasi Tmodelind[%d]=%d,Tvar[Tmodelind[%d]]=V%d, ncovcol=%d, nqv=%d,Tvar[k]- ncovcol-nqv=%d\n",ntveff,k,ntveff,Tvar[k], ncovcol, nqv,Tvar[k]- ncovcol-nqv);
11377: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 11378: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.339 brouard 11379: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11380: /* model V1+V3+age*V1+age*V3+V1*V3 */
11381: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11382: ncovvt++;
11383: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11384: TvarVVind[ncovvt]=k; /* TvarVV[1]=V3 (first time varying in the model equation */
11385:
1.234 brouard 11386: Fixed[k]= 1;
11387: Dummy[k]= 1;
11388: nqtveff++;
11389: modell[k].maintype= VTYPE;
11390: modell[k].subtype= VQ;
11391: ncovv++; /* Only simple time varying variables */
11392: nsq++;
1.334 brouard 11393: TvarsQ[nsq]=Tvar[k]; /* k=1 Tvar=5 nsq=1 TvarsQ[1]=5 */ /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary here) */
11394: TvarsQind[nsq]=k; /* For single quantitative covariate gives the model position of each single quantitative covariate *//* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.234 brouard 11395: TvarV[ncovv]=Tvar[k];
1.242 brouard 11396: TvarVind[ncovv]=k; /* TvarVind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Any time varying singele */
1.231 brouard 11397: TvarVQ[nqtveff]=Tvar[k]; /* TvarVQ[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying quantitative variable */
11398: TvarVQind[nqtveff]=k; /* TvarVQind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying quantitative variable */
1.234 brouard 11399: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
11400: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
1.342 brouard 11401: /* printf("Quasi TmodelQind[%d]=%d,Tvar[TmodelQind[%d]]=V%d, ncovcol=%d, nqv=%d, ntv=%d,Tvar[k]- ncovcol-nqv-ntv=%d\n",nqtveff,k,nqtveff,Tvar[k], ncovcol, nqv, ntv, Tvar[k]- ncovcol-nqv-ntv); */
11402: /* printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv); */
1.227 brouard 11403: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 11404: ncova++;
11405: TvarA[ncova]=Tvar[k];
11406: TvarAind[ncova]=k;
1.231 brouard 11407: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 11408: Fixed[k]= 2;
11409: Dummy[k]= 2;
11410: modell[k].maintype= ATYPE;
11411: modell[k].subtype= APFD;
11412: /* ncoveff++; */
1.227 brouard 11413: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 11414: Fixed[k]= 2;
11415: Dummy[k]= 3;
11416: modell[k].maintype= ATYPE;
11417: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
11418: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 11419: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 11420: Fixed[k]= 3;
11421: Dummy[k]= 2;
11422: modell[k].maintype= ATYPE;
11423: modell[k].subtype= APVD; /* Product age * varying dummy */
11424: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 11425: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 11426: Fixed[k]= 3;
11427: Dummy[k]= 3;
11428: modell[k].maintype= ATYPE;
11429: modell[k].subtype= APVQ; /* Product age * varying quantitative */
11430: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 11431: }
1.339 brouard 11432: }else if (Typevar[k] == 2) { /* product Vn * Vm without age, V1+V3+age*V1+age*V3+V1*V3 looking at V1*V3, Typevar={0, 0, 1, 1, 2}, k=5, V1 is fixed, V3 is timevary, V5 is a product */
11433: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11434: /* model V1+V3+age*V1+age*V3+V1*V3 */
11435: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11436: k1=Tposprod[k]; /* Position in the products of product k, Tposprod={0, 0, 0, 0, 1} k1=1 first product but second time varying because of V3 */
11437: ncovvt++;
11438: TvarVV[ncovvt]=Tvard[k1][1]; /* TvarVV[2]=V1 (because TvarVV[1] was V3, first time varying covariates */
11439: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11440: ncovvt++;
11441: TvarVV[ncovvt]=Tvard[k1][2]; /* TvarVV[3]=V3 */
11442: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11443:
11444:
11445: if(Tvard[k1][1] <=ncovcol){ /* Vn is dummy fixed, (Tvard[1][1]=V1), (Tvard[1][1]=V3 time varying) */
11446: if(Tvard[k1][2] <=ncovcol){ /* Vm is dummy fixed */
1.240 brouard 11447: Fixed[k]= 1;
11448: Dummy[k]= 0;
11449: modell[k].maintype= FTYPE;
11450: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
11451: ncovf++; /* Fixed variables without age */
11452: TvarF[ncovf]=Tvar[k];
11453: TvarFind[ncovf]=k;
1.339 brouard 11454: }else if(Tvard[k1][2] <=ncovcol+nqv){ /* Vm is quanti fixed */
11455: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11456: Dummy[k]= 1;
11457: modell[k].maintype= FTYPE;
11458: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
11459: ncovf++; /* Varying variables without age */
11460: TvarF[ncovf]=Tvar[k];
11461: TvarFind[ncovf]=k;
1.339 brouard 11462: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is a time varying dummy covariate */
1.240 brouard 11463: Fixed[k]= 1;
11464: Dummy[k]= 0;
11465: modell[k].maintype= VTYPE;
11466: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
11467: ncovv++; /* Varying variables without age */
1.339 brouard 11468: TvarV[ncovv]=Tvar[k]; /* TvarV[1]=Tvar[5]=5 because there is a V4 */
11469: TvarVind[ncovv]=k;/* TvarVind[1]=5 */
11470: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is a time varying quantitative covariate */
1.240 brouard 11471: Fixed[k]= 1;
11472: Dummy[k]= 1;
11473: modell[k].maintype= VTYPE;
11474: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
11475: ncovv++; /* Varying variables without age */
11476: TvarV[ncovv]=Tvar[k];
11477: TvarVind[ncovv]=k;
11478: }
1.339 brouard 11479: }else if(Tvard[k1][1] <=ncovcol+nqv){ /* Vn is fixed quanti */
11480: if(Tvard[k1][2] <=ncovcol){ /* Vm is fixed dummy */
11481: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11482: Dummy[k]= 1;
11483: modell[k].maintype= FTYPE;
11484: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
11485: ncovf++; /* Fixed variables without age */
11486: TvarF[ncovf]=Tvar[k];
11487: TvarFind[ncovf]=k;
1.339 brouard 11488: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is time varying */
1.240 brouard 11489: Fixed[k]= 1;
11490: Dummy[k]= 1;
11491: modell[k].maintype= VTYPE;
11492: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
11493: ncovv++; /* Varying variables without age */
11494: TvarV[ncovv]=Tvar[k];
11495: TvarVind[ncovv]=k;
1.339 brouard 11496: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is time varying quanti */
1.240 brouard 11497: Fixed[k]= 1;
11498: Dummy[k]= 1;
11499: modell[k].maintype= VTYPE;
11500: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
11501: ncovv++; /* Varying variables without age */
11502: TvarV[ncovv]=Tvar[k];
11503: TvarVind[ncovv]=k;
11504: ncovv++; /* Varying variables without age */
11505: TvarV[ncovv]=Tvar[k];
11506: TvarVind[ncovv]=k;
11507: }
1.339 brouard 11508: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){ /* Vn is time varying dummy */
1.240 brouard 11509: if(Tvard[k1][2] <=ncovcol){
11510: Fixed[k]= 1;
11511: Dummy[k]= 1;
11512: modell[k].maintype= VTYPE;
11513: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
11514: ncovv++; /* Varying variables without age */
11515: TvarV[ncovv]=Tvar[k];
11516: TvarVind[ncovv]=k;
11517: }else if(Tvard[k1][2] <=ncovcol+nqv){
11518: Fixed[k]= 1;
11519: Dummy[k]= 1;
11520: modell[k].maintype= VTYPE;
11521: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
11522: ncovv++; /* Varying variables without age */
11523: TvarV[ncovv]=Tvar[k];
11524: TvarVind[ncovv]=k;
11525: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11526: Fixed[k]= 1;
11527: Dummy[k]= 0;
11528: modell[k].maintype= VTYPE;
11529: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
11530: ncovv++; /* Varying variables without age */
11531: TvarV[ncovv]=Tvar[k];
11532: TvarVind[ncovv]=k;
11533: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11534: Fixed[k]= 1;
11535: Dummy[k]= 1;
11536: modell[k].maintype= VTYPE;
11537: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
11538: ncovv++; /* Varying variables without age */
11539: TvarV[ncovv]=Tvar[k];
11540: TvarVind[ncovv]=k;
11541: }
1.339 brouard 11542: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){ /* Vn is time varying quanti */
1.240 brouard 11543: if(Tvard[k1][2] <=ncovcol){
11544: Fixed[k]= 1;
11545: Dummy[k]= 1;
11546: modell[k].maintype= VTYPE;
11547: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
11548: ncovv++; /* Varying variables without age */
11549: TvarV[ncovv]=Tvar[k];
11550: TvarVind[ncovv]=k;
11551: }else if(Tvard[k1][2] <=ncovcol+nqv){
11552: Fixed[k]= 1;
11553: Dummy[k]= 1;
11554: modell[k].maintype= VTYPE;
11555: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
11556: ncovv++; /* Varying variables without age */
11557: TvarV[ncovv]=Tvar[k];
11558: TvarVind[ncovv]=k;
11559: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11560: Fixed[k]= 1;
11561: Dummy[k]= 1;
11562: modell[k].maintype= VTYPE;
11563: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
11564: ncovv++; /* Varying variables without age */
11565: TvarV[ncovv]=Tvar[k];
11566: TvarVind[ncovv]=k;
11567: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11568: Fixed[k]= 1;
11569: Dummy[k]= 1;
11570: modell[k].maintype= VTYPE;
11571: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
11572: ncovv++; /* Varying variables without age */
11573: TvarV[ncovv]=Tvar[k];
11574: TvarVind[ncovv]=k;
11575: }
1.227 brouard 11576: }else{
1.240 brouard 11577: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11578: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11579: } /*end k1*/
1.225 brouard 11580: }else{
1.226 brouard 11581: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
11582: fprintf(ficlog,"Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
1.225 brouard 11583: }
1.342 brouard 11584: /* printf("Decodemodel, k=%d, Tvar[%d]=V%d,Typevar=%d, Fixed=%d, Dummy=%d\n",k, k,Tvar[k],Typevar[k],Fixed[k],Dummy[k]); */
11585: /* printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype); */
1.227 brouard 11586: fprintf(ficlog,"Decodemodel, k=%d, Tvar[%d]=V%d,Typevar=%d, Fixed=%d, Dummy=%d\n",k, k,Tvar[k],Typevar[k],Fixed[k],Dummy[k]);
11587: }
11588: /* Searching for doublons in the model */
11589: for(k1=1; k1<= cptcovt;k1++){
11590: for(k2=1; k2 <k1;k2++){
1.285 brouard 11591: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
11592: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 11593: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
11594: if(Tvar[k1]==Tvar[k2]){
1.338 brouard 11595: printf("Error duplication in the model=1+age+%s at positions (+) %d and %d, Tvar[%d]=V%d, Tvar[%d]=V%d, Typevar=%d, Fixed=%d, Dummy=%d\n", model, k1,k2, k1, Tvar[k1], k2, Tvar[k2],Typevar[k1],Fixed[k1],Dummy[k1]);
11596: fprintf(ficlog,"Error duplication in the model=1+age+%s at positions (+) %d and %d, Tvar[%d]=V%d, Tvar[%d]=V%d, Typevar=%d, Fixed=%d, Dummy=%d\n", model, k1,k2, k1, Tvar[k1], k2, Tvar[k2],Typevar[k1],Fixed[k1],Dummy[k1]); fflush(ficlog);
1.234 brouard 11597: return(1);
11598: }
11599: }else if (Typevar[k1] ==2){
11600: k3=Tposprod[k1];
11601: k4=Tposprod[k2];
11602: if( ((Tvard[k3][1]== Tvard[k4][1])&&(Tvard[k3][2]== Tvard[k4][2])) || ((Tvard[k3][1]== Tvard[k4][2])&&(Tvard[k3][2]== Tvard[k4][1])) ){
1.338 brouard 11603: printf("Error duplication in the model=1+age+%s at positions (+) %d and %d, V%d*V%d, Typevar=%d, Fixed=%d, Dummy=%d\n",model, k1,k2, Tvard[k3][1], Tvard[k3][2],Typevar[k1],Fixed[Tvar[k1]],Dummy[Tvar[k1]]);
11604: fprintf(ficlog,"Error duplication in the model=1+age+%s at positions (+) %d and %d, V%d*V%d, Typevar=%d, Fixed=%d, Dummy=%d\n",model, k1,k2, Tvard[k3][1], Tvard[k3][2],Typevar[k1],Fixed[Tvar[k1]],Dummy[Tvar[k1]]); fflush(ficlog);
1.234 brouard 11605: return(1);
11606: }
11607: }
1.227 brouard 11608: }
11609: }
1.225 brouard 11610: }
11611: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
11612: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 11613: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
11614: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 11615: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 11616: /*endread:*/
1.225 brouard 11617: printf("Exiting decodemodel: ");
11618: return (1);
1.136 brouard 11619: }
11620:
1.169 brouard 11621: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 11622: {/* Check ages at death */
1.136 brouard 11623: int i, m;
1.218 brouard 11624: int firstone=0;
11625:
1.136 brouard 11626: for (i=1; i<=imx; i++) {
11627: for(m=2; (m<= maxwav); m++) {
11628: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
11629: anint[m][i]=9999;
1.216 brouard 11630: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
11631: s[m][i]=-1;
1.136 brouard 11632: }
11633: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 11634: *nberr = *nberr + 1;
1.218 brouard 11635: if(firstone == 0){
11636: firstone=1;
1.260 brouard 11637: printf("Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\nOther similar cases in log file\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.218 brouard 11638: }
1.262 brouard 11639: fprintf(ficlog,"Warning (#%d)! Date of death (month %2d and year %4d) of individual %ld on line %d was unknown but status is a death state %d at wave %d. If you don't know the vital status, please enter -2. If he/she is still alive but don't know the state, please code with '-1 or '.'. Here, we do not believe in a death, skipped.\n", *nberr,(int)moisdc[i],(int)andc[i],num[i],i,s[m][i],m);
1.260 brouard 11640: s[m][i]=-1; /* Droping the death status */
1.136 brouard 11641: }
11642: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 11643: (*nberr)++;
1.259 brouard 11644: printf("Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\nOther similar cases in log file\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
1.262 brouard 11645: fprintf(ficlog,"Error (#%d)! Month of death of individual %ld on line %d was unknown (%2d) (year of death is %4d) and status is a death state %d at wave %d. Please impute an arbitrary (or not) month and rerun. Currently this transition to death will be skipped (status is set to -2).\n", *nberr, num[i],i,(int)moisdc[i],(int)andc[i],s[m][i],m);
1.259 brouard 11646: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 11647: }
11648: }
11649: }
11650:
11651: for (i=1; i<=imx; i++) {
11652: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
11653: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 11654: if(s[m][i] >0 || s[m][i]==-1 || s[m][i]==-2 || s[m][i]==-4 || s[m][i]==-5){ /* What if s[m][i]=-1 */
1.136 brouard 11655: if (s[m][i] >= nlstate+1) {
1.169 brouard 11656: if(agedc[i]>0){
11657: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 11658: agev[m][i]=agedc[i];
1.214 brouard 11659: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 11660: }else {
1.136 brouard 11661: if ((int)andc[i]!=9999){
11662: nbwarn++;
11663: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
11664: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
11665: agev[m][i]=-1;
11666: }
11667: }
1.169 brouard 11668: } /* agedc > 0 */
1.214 brouard 11669: } /* end if */
1.136 brouard 11670: else if(s[m][i] !=9){ /* Standard case, age in fractional
11671: years but with the precision of a month */
11672: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
11673: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
11674: agev[m][i]=1;
11675: else if(agev[m][i] < *agemin){
11676: *agemin=agev[m][i];
11677: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
11678: }
11679: else if(agev[m][i] >*agemax){
11680: *agemax=agev[m][i];
1.156 brouard 11681: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 11682: }
11683: /*agev[m][i]=anint[m][i]-annais[i];*/
11684: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 11685: } /* en if 9*/
1.136 brouard 11686: else { /* =9 */
1.214 brouard 11687: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 11688: agev[m][i]=1;
11689: s[m][i]=-1;
11690: }
11691: }
1.214 brouard 11692: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 11693: agev[m][i]=1;
1.214 brouard 11694: else{
11695: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11696: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11697: agev[m][i]=0;
11698: }
11699: } /* End for lastpass */
11700: }
1.136 brouard 11701:
11702: for (i=1; i<=imx; i++) {
11703: for(m=firstpass; (m<=lastpass); m++){
11704: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 11705: (*nberr)++;
1.136 brouard 11706: printf("Error: on wave %d of individual %d status %d > (nlstate+ndeath)=(%d+%d)=%d\n",m,i,s[m][i],nlstate, ndeath, nlstate+ndeath);
11707: fprintf(ficlog,"Error: on wave %d of individual %d status %d > (nlstate+ndeath)=(%d+%d)=%d\n",m,i,s[m][i],nlstate, ndeath, nlstate+ndeath);
11708: return 1;
11709: }
11710: }
11711: }
11712:
11713: /*for (i=1; i<=imx; i++){
11714: for (m=firstpass; (m<lastpass); m++){
11715: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
11716: }
11717:
11718: }*/
11719:
11720:
1.139 brouard 11721: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
11722: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 11723:
11724: return (0);
1.164 brouard 11725: /* endread:*/
1.136 brouard 11726: printf("Exiting calandcheckages: ");
11727: return (1);
11728: }
11729:
1.172 brouard 11730: #if defined(_MSC_VER)
11731: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11732: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11733: //#include "stdafx.h"
11734: //#include <stdio.h>
11735: //#include <tchar.h>
11736: //#include <windows.h>
11737: //#include <iostream>
11738: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
11739:
11740: LPFN_ISWOW64PROCESS fnIsWow64Process;
11741:
11742: BOOL IsWow64()
11743: {
11744: BOOL bIsWow64 = FALSE;
11745:
11746: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
11747: // (HANDLE, PBOOL);
11748:
11749: //LPFN_ISWOW64PROCESS fnIsWow64Process;
11750:
11751: HMODULE module = GetModuleHandle(_T("kernel32"));
11752: const char funcName[] = "IsWow64Process";
11753: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
11754: GetProcAddress(module, funcName);
11755:
11756: if (NULL != fnIsWow64Process)
11757: {
11758: if (!fnIsWow64Process(GetCurrentProcess(),
11759: &bIsWow64))
11760: //throw std::exception("Unknown error");
11761: printf("Unknown error\n");
11762: }
11763: return bIsWow64 != FALSE;
11764: }
11765: #endif
1.177 brouard 11766:
1.191 brouard 11767: void syscompilerinfo(int logged)
1.292 brouard 11768: {
11769: #include <stdint.h>
11770:
11771: /* #include "syscompilerinfo.h"*/
1.185 brouard 11772: /* command line Intel compiler 32bit windows, XP compatible:*/
11773: /* /GS /W3 /Gy
11774: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
11775: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
11776: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 11777: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
11778: */
11779: /* 64 bits */
1.185 brouard 11780: /*
11781: /GS /W3 /Gy
11782: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
11783: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
11784: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
11785: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
11786: /* Optimization are useless and O3 is slower than O2 */
11787: /*
11788: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
11789: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
11790: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
11791: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
11792: */
1.186 brouard 11793: /* Link is */ /* /OUT:"visual studio
1.185 brouard 11794: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
11795: /PDB:"visual studio
11796: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
11797: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
11798: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
11799: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
11800: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
11801: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
11802: uiAccess='false'"
11803: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
11804: /NOLOGO /TLBID:1
11805: */
1.292 brouard 11806:
11807:
1.177 brouard 11808: #if defined __INTEL_COMPILER
1.178 brouard 11809: #if defined(__GNUC__)
11810: struct utsname sysInfo; /* For Intel on Linux and OS/X */
11811: #endif
1.177 brouard 11812: #elif defined(__GNUC__)
1.179 brouard 11813: #ifndef __APPLE__
1.174 brouard 11814: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 11815: #endif
1.177 brouard 11816: struct utsname sysInfo;
1.178 brouard 11817: int cross = CROSS;
11818: if (cross){
11819: printf("Cross-");
1.191 brouard 11820: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 11821: }
1.174 brouard 11822: #endif
11823:
1.191 brouard 11824: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 11825: #if defined(__clang__)
1.191 brouard 11826: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 11827: #endif
11828: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 11829: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 11830: #endif
11831: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 11832: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 11833: #endif
11834: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 11835: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 11836: #endif
11837: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 11838: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 11839: #endif
11840: #if defined(_MSC_VER)
1.191 brouard 11841: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 11842: #endif
11843: #if defined(__PGI)
1.191 brouard 11844: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 11845: #endif
11846: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 11847: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 11848: #endif
1.191 brouard 11849: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 11850:
1.167 brouard 11851: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
11852: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
11853: // Windows (x64 and x86)
1.191 brouard 11854: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 11855: #elif __unix__ // all unices, not all compilers
11856: // Unix
1.191 brouard 11857: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 11858: #elif __linux__
11859: // linux
1.191 brouard 11860: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 11861: #elif __APPLE__
1.174 brouard 11862: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 11863: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11864: #endif
11865:
11866: /* __MINGW32__ */
11867: /* __CYGWIN__ */
11868: /* __MINGW64__ */
11869: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11870: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11871: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11872: /* _WIN64 // Defined for applications for Win64. */
11873: /* _M_X64 // Defined for compilations that target x64 processors. */
11874: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11875:
1.167 brouard 11876: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11877: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11878: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11879: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11880: #else
1.191 brouard 11881: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11882: #endif
11883:
1.169 brouard 11884: #if defined(__GNUC__)
11885: # if defined(__GNUC_PATCHLEVEL__)
11886: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11887: + __GNUC_MINOR__ * 100 \
11888: + __GNUC_PATCHLEVEL__)
11889: # else
11890: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11891: + __GNUC_MINOR__ * 100)
11892: # endif
1.174 brouard 11893: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11894: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11895:
11896: if (uname(&sysInfo) != -1) {
11897: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11898: if(logged) fprintf(ficlog,"Running on: %s %s %s %s %s\n ",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.176 brouard 11899: }
11900: else
11901: perror("uname() error");
1.179 brouard 11902: //#ifndef __INTEL_COMPILER
11903: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11904: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11905: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11906: #endif
1.169 brouard 11907: #endif
1.172 brouard 11908:
1.286 brouard 11909: // void main ()
1.172 brouard 11910: // {
1.169 brouard 11911: #if defined(_MSC_VER)
1.174 brouard 11912: if (IsWow64()){
1.191 brouard 11913: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11914: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11915: }
11916: else{
1.191 brouard 11917: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11918: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11919: }
1.172 brouard 11920: // printf("\nPress Enter to continue...");
11921: // getchar();
11922: // }
11923:
1.169 brouard 11924: #endif
11925:
1.167 brouard 11926:
1.219 brouard 11927: }
1.136 brouard 11928:
1.219 brouard 11929: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11930: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 11931: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 11932: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11933: /* double ftolpl = 1.e-10; */
1.180 brouard 11934: double age, agebase, agelim;
1.203 brouard 11935: double tot;
1.180 brouard 11936:
1.202 brouard 11937: strcpy(filerespl,"PL_");
11938: strcat(filerespl,fileresu);
11939: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11940: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11941: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11942: }
1.288 brouard 11943: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11944: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11945: pstamp(ficrespl);
1.288 brouard 11946: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11947: fprintf(ficrespl,"#Age ");
11948: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11949: fprintf(ficrespl,"\n");
1.180 brouard 11950:
1.219 brouard 11951: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11952:
1.219 brouard 11953: agebase=ageminpar;
11954: agelim=agemaxpar;
1.180 brouard 11955:
1.227 brouard 11956: /* i1=pow(2,ncoveff); */
1.234 brouard 11957: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11958: if (cptcovn < 1){i1=1;}
1.180 brouard 11959:
1.337 brouard 11960: /* for(k=1; k<=i1;k++){ /\* For each combination k of dummy covariates in the model *\/ */
1.238 brouard 11961: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 11962: k=TKresult[nres];
1.338 brouard 11963: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 11964: /* if(i1 != 1 && TKresult[nres]!= k) /\* We found the combination k corresponding to the resultline value of dummies *\/ */
11965: /* continue; */
1.235 brouard 11966:
1.238 brouard 11967: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11968: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11969: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11970: /* k=k+1; */
11971: /* to clean */
1.332 brouard 11972: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11973: fprintf(ficrespl,"#******");
11974: printf("#******");
11975: fprintf(ficlog,"#******");
1.337 brouard 11976: for(j=1;j<=cptcovs ;j++) {/**< cptcovs number of SIMPLE covariates in the model or resultline V2+V1 =2 (dummy or quantit or time varying) */
1.332 brouard 11977: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
1.337 brouard 11978: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11979: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11980: fprintf(ficrespl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11981: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11982: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11983: }
11984: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
11985: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11986: /* fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11987: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11988: /* } */
1.238 brouard 11989: fprintf(ficrespl,"******\n");
11990: printf("******\n");
11991: fprintf(ficlog,"******\n");
11992: if(invalidvarcomb[k]){
11993: printf("\nCombination (%d) ignored because no case \n",k);
11994: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
11995: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
11996: continue;
11997: }
1.219 brouard 11998:
1.238 brouard 11999: fprintf(ficrespl,"#Age ");
1.337 brouard 12000: /* for(j=1;j<=cptcoveff;j++) { */
12001: /* fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12002: /* } */
12003: for(j=1;j<=cptcovs;j++) { /* New the quanti variable is added */
12004: fprintf(ficrespl,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12005: }
12006: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
12007: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 12008:
1.238 brouard 12009: for (age=agebase; age<=agelim; age++){
12010: /* for (age=agebase; age<=agebase; age++){ */
1.337 brouard 12011: /**< Computes the prevalence limit in each live state at age x and for covariate combination (k and) nres */
12012: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres); /* Nicely done */
1.238 brouard 12013: fprintf(ficrespl,"%.0f ",age );
1.337 brouard 12014: /* for(j=1;j<=cptcoveff;j++) */
12015: /* fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12016: for(j=1;j<=cptcovs;j++)
12017: fprintf(ficrespl,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12018: tot=0.;
12019: for(i=1; i<=nlstate;i++){
12020: tot += prlim[i][i];
12021: fprintf(ficrespl," %.5f", prlim[i][i]);
12022: }
12023: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
12024: } /* Age */
12025: /* was end of cptcod */
1.337 brouard 12026: } /* nres */
12027: /* } /\* for each combination *\/ */
1.219 brouard 12028: return 0;
1.180 brouard 12029: }
12030:
1.218 brouard 12031: int back_prevalence_limit(double *p, double **bprlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp, double dateprev1,double dateprev2, int firstpass, int lastpass, int mobilavproj){
1.288 brouard 12032: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 12033:
12034: /* Computes the back prevalence limit for any combination of covariate values
12035: * at any age between ageminpar and agemaxpar
12036: */
1.235 brouard 12037: int i, j, k, i1, nres=0 ;
1.217 brouard 12038: /* double ftolpl = 1.e-10; */
12039: double age, agebase, agelim;
12040: double tot;
1.218 brouard 12041: /* double ***mobaverage; */
12042: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 12043:
12044: strcpy(fileresplb,"PLB_");
12045: strcat(fileresplb,fileresu);
12046: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 12047: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
12048: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 12049: }
1.288 brouard 12050: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
12051: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 12052: pstamp(ficresplb);
1.288 brouard 12053: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 12054: fprintf(ficresplb,"#Age ");
12055: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
12056: fprintf(ficresplb,"\n");
12057:
1.218 brouard 12058:
12059: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
12060:
12061: agebase=ageminpar;
12062: agelim=agemaxpar;
12063:
12064:
1.227 brouard 12065: i1=pow(2,cptcoveff);
1.218 brouard 12066: if (cptcovn < 1){i1=1;}
1.227 brouard 12067:
1.238 brouard 12068: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.338 brouard 12069: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
12070: k=TKresult[nres];
12071: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
12072: /* if(i1 != 1 && TKresult[nres]!= k) */
12073: /* continue; */
12074: /* /\*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*\/ */
1.238 brouard 12075: fprintf(ficresplb,"#******");
12076: printf("#******");
12077: fprintf(ficlog,"#******");
1.338 brouard 12078: for(j=1;j<=cptcovs ;j++) {/**< cptcovs number of SIMPLE covariates in the model or resultline V2+V1 =2 (dummy or quantit or time varying) */
12079: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12080: fprintf(ficresplb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12081: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12082: }
1.338 brouard 12083: /* for(j=1;j<=cptcoveff ;j++) {/\* all covariates *\/ */
12084: /* fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12085: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12086: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12087: /* } */
12088: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12089: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12090: /* fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12091: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12092: /* } */
1.238 brouard 12093: fprintf(ficresplb,"******\n");
12094: printf("******\n");
12095: fprintf(ficlog,"******\n");
12096: if(invalidvarcomb[k]){
12097: printf("\nCombination (%d) ignored because no cases \n",k);
12098: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
12099: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
12100: continue;
12101: }
1.218 brouard 12102:
1.238 brouard 12103: fprintf(ficresplb,"#Age ");
1.338 brouard 12104: for(j=1;j<=cptcovs;j++) {
12105: fprintf(ficresplb,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12106: }
12107: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
12108: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 12109:
12110:
1.238 brouard 12111: for (age=agebase; age<=agelim; age++){
12112: /* for (age=agebase; age<=agebase; age++){ */
12113: if(mobilavproj > 0){
12114: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
12115: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 12116: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 12117: }else if (mobilavproj == 0){
12118: printf("There is no chance to get back prevalence limit if data aren't non zero and summing to 1, please try a non null mobil_average(=%d) parameter or mobil_average=-1 if you want to try at your own risk.\n",mobilavproj);
12119: fprintf(ficlog,"There is no chance to get back prevalence limit if data aren't non zero and summing to 1, please try a non null mobil_average(=%d) parameter or mobil_average=-1 if you want to try at your own risk.\n",mobilavproj);
12120: exit(1);
12121: }else{
12122: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 12123: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 12124: /* printf("TOTOT\n"); */
12125: /* exit(1); */
1.238 brouard 12126: }
12127: fprintf(ficresplb,"%.0f ",age );
1.338 brouard 12128: for(j=1;j<=cptcovs;j++)
12129: fprintf(ficresplb,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12130: tot=0.;
12131: for(i=1; i<=nlstate;i++){
12132: tot += bprlim[i][i];
12133: fprintf(ficresplb," %.5f", bprlim[i][i]);
12134: }
12135: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
12136: } /* Age */
12137: /* was end of cptcod */
1.255 brouard 12138: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.338 brouard 12139: /* } /\* end of any combination *\/ */
1.238 brouard 12140: } /* end of nres */
1.218 brouard 12141: /* hBijx(p, bage, fage); */
12142: /* fclose(ficrespijb); */
12143:
12144: return 0;
1.217 brouard 12145: }
1.218 brouard 12146:
1.180 brouard 12147: int hPijx(double *p, int bage, int fage){
12148: /*------------- h Pij x at various ages ------------*/
1.336 brouard 12149: /* to be optimized with precov */
1.180 brouard 12150: int stepsize;
12151: int agelim;
12152: int hstepm;
12153: int nhstepm;
1.235 brouard 12154: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 12155:
12156: double agedeb;
12157: double ***p3mat;
12158:
1.337 brouard 12159: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
12160: if((ficrespij=fopen(filerespij,"w"))==NULL) {
12161: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
12162: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
12163: }
12164: printf("Computing pij: result on file '%s' \n", filerespij);
12165: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
12166:
12167: stepsize=(int) (stepm+YEARM-1)/YEARM;
12168: /*if (stepm<=24) stepsize=2;*/
12169:
12170: agelim=AGESUP;
12171: hstepm=stepsize*YEARM; /* Every year of age */
12172: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
12173:
12174: /* hstepm=1; aff par mois*/
12175: pstamp(ficrespij);
12176: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
12177: i1= pow(2,cptcoveff);
12178: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12179: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12180: /* k=k+1; */
12181: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
12182: k=TKresult[nres];
1.338 brouard 12183: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12184: /* for(k=1; k<=i1;k++){ */
12185: /* if(i1 != 1 && TKresult[nres]!= k) */
12186: /* continue; */
12187: fprintf(ficrespij,"\n#****** ");
12188: for(j=1;j<=cptcovs;j++){
12189: fprintf(ficrespij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12190: /* fprintf(ficrespij,"@wV%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12191: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
12192: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12193: /* fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12194: }
12195: fprintf(ficrespij,"******\n");
12196:
12197: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
12198: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
12199: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
12200:
12201: /* nhstepm=nhstepm*YEARM; aff par mois*/
12202:
12203: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12204: oldm=oldms;savm=savms;
12205: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
12206: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
12207: for(i=1; i<=nlstate;i++)
12208: for(j=1; j<=nlstate+ndeath;j++)
12209: fprintf(ficrespij," %1d-%1d",i,j);
12210: fprintf(ficrespij,"\n");
12211: for (h=0; h<=nhstepm; h++){
12212: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12213: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.183 brouard 12214: for(i=1; i<=nlstate;i++)
12215: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12216: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.183 brouard 12217: fprintf(ficrespij,"\n");
12218: }
1.337 brouard 12219: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12220: fprintf(ficrespij,"\n");
1.180 brouard 12221: }
1.337 brouard 12222: }
12223: /*}*/
12224: return 0;
1.180 brouard 12225: }
1.218 brouard 12226:
12227: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 12228: /*------------- h Bij x at various ages ------------*/
1.336 brouard 12229: /* To be optimized with precov */
1.217 brouard 12230: int stepsize;
1.218 brouard 12231: /* int agelim; */
12232: int ageminl;
1.217 brouard 12233: int hstepm;
12234: int nhstepm;
1.238 brouard 12235: int h, i, i1, j, k, nres;
1.218 brouard 12236:
1.217 brouard 12237: double agedeb;
12238: double ***p3mat;
1.218 brouard 12239:
12240: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
12241: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
12242: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12243: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12244: }
12245: printf("Computing pij back: result on file '%s' \n", filerespijb);
12246: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
12247:
12248: stepsize=(int) (stepm+YEARM-1)/YEARM;
12249: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 12250:
1.218 brouard 12251: /* agelim=AGESUP; */
1.289 brouard 12252: ageminl=AGEINF; /* was 30 */
1.218 brouard 12253: hstepm=stepsize*YEARM; /* Every year of age */
12254: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
12255:
12256: /* hstepm=1; aff par mois*/
12257: pstamp(ficrespijb);
1.255 brouard 12258: fprintf(ficrespijb,"#****** h Bij x Back probability to be in state i at age x-h being in j at x: B1j+B2j+...=1 ");
1.227 brouard 12259: i1= pow(2,cptcoveff);
1.218 brouard 12260: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12261: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12262: /* k=k+1; */
1.238 brouard 12263: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 12264: k=TKresult[nres];
1.338 brouard 12265: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12266: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
12267: /* if(i1 != 1 && TKresult[nres]!= k) */
12268: /* continue; */
12269: fprintf(ficrespijb,"\n#****** ");
12270: for(j=1;j<=cptcovs;j++){
1.338 brouard 12271: fprintf(ficrespijb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.337 brouard 12272: /* for(j=1;j<=cptcoveff;j++) */
12273: /* fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12274: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12275: /* fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12276: }
12277: fprintf(ficrespijb,"******\n");
12278: if(invalidvarcomb[k]){ /* Is it necessary here? */
12279: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
12280: continue;
12281: }
12282:
12283: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
12284: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
12285: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
12286: nhstepm=(int) rint((agedeb-ageminl)*YEARM/stepm+0.1)-1; /* Typically 20 years = 20*12/6=40 or 55*12/24=27.5-1.1=>27 */
12287: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
12288:
12289: /* nhstepm=nhstepm*YEARM; aff par mois*/
12290:
12291: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
12292: /* and memory limitations if stepm is small */
12293:
12294: /* oldm=oldms;savm=savms; */
12295: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
12296: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
12297: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
12298: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
12299: for(i=1; i<=nlstate;i++)
12300: for(j=1; j<=nlstate+ndeath;j++)
12301: fprintf(ficrespijb," %1d-%1d",i,j);
12302: fprintf(ficrespijb,"\n");
12303: for (h=0; h<=nhstepm; h++){
12304: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12305: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
12306: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
1.217 brouard 12307: for(i=1; i<=nlstate;i++)
12308: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12309: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.217 brouard 12310: fprintf(ficrespijb,"\n");
1.337 brouard 12311: }
12312: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12313: fprintf(ficrespijb,"\n");
12314: } /* end age deb */
12315: /* } /\* end combination *\/ */
1.238 brouard 12316: } /* end nres */
1.218 brouard 12317: return 0;
12318: } /* hBijx */
1.217 brouard 12319:
1.180 brouard 12320:
1.136 brouard 12321: /***********************************************/
12322: /**************** Main Program *****************/
12323: /***********************************************/
12324:
12325: int main(int argc, char *argv[])
12326: {
12327: #ifdef GSL
12328: const gsl_multimin_fminimizer_type *T;
12329: size_t iteri = 0, it;
12330: int rval = GSL_CONTINUE;
12331: int status = GSL_SUCCESS;
12332: double ssval;
12333: #endif
12334: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 12335: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
12336: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 12337: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 12338: int jj, ll, li, lj, lk;
1.136 brouard 12339: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 12340: int num_filled;
1.136 brouard 12341: int itimes;
12342: int NDIM=2;
12343: int vpopbased=0;
1.235 brouard 12344: int nres=0;
1.258 brouard 12345: int endishere=0;
1.277 brouard 12346: int noffset=0;
1.274 brouard 12347: int ncurrv=0; /* Temporary variable */
12348:
1.164 brouard 12349: char ca[32], cb[32];
1.136 brouard 12350: /* FILE *fichtm; *//* Html File */
12351: /* FILE *ficgp;*/ /*Gnuplot File */
12352: struct stat info;
1.191 brouard 12353: double agedeb=0.;
1.194 brouard 12354:
12355: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 12356: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 12357:
1.165 brouard 12358: double fret;
1.191 brouard 12359: double dum=0.; /* Dummy variable */
1.136 brouard 12360: double ***p3mat;
1.218 brouard 12361: /* double ***mobaverage; */
1.319 brouard 12362: double wald;
1.164 brouard 12363:
12364: char line[MAXLINE];
1.197 brouard 12365: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
12366:
1.234 brouard 12367: char modeltemp[MAXLINE];
1.332 brouard 12368: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 12369:
1.136 brouard 12370: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 12371: char *tok, *val; /* pathtot */
1.334 brouard 12372: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.195 brouard 12373: int c, h , cpt, c2;
1.191 brouard 12374: int jl=0;
12375: int i1, j1, jk, stepsize=0;
1.194 brouard 12376: int count=0;
12377:
1.164 brouard 12378: int *tab;
1.136 brouard 12379: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 12380: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
12381: /* double anprojf, mprojf, jprojf; */
12382: /* double jintmean,mintmean,aintmean; */
12383: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12384: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12385: double yrfproj= 10.0; /* Number of years of forward projections */
12386: double yrbproj= 10.0; /* Number of years of backward projections */
12387: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 12388: int mobilav=0,popforecast=0;
1.191 brouard 12389: int hstepm=0, nhstepm=0;
1.136 brouard 12390: int agemortsup;
12391: float sumlpop=0.;
12392: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
12393: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
12394:
1.191 brouard 12395: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 12396: double ftolpl=FTOL;
12397: double **prlim;
1.217 brouard 12398: double **bprlim;
1.317 brouard 12399: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
12400: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 12401: double ***paramstart; /* Matrix of starting parameter values */
12402: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 12403: double **matcov; /* Matrix of covariance */
1.203 brouard 12404: double **hess; /* Hessian matrix */
1.136 brouard 12405: double ***delti3; /* Scale */
12406: double *delti; /* Scale */
12407: double ***eij, ***vareij;
12408: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 12409:
1.136 brouard 12410: double *epj, vepp;
1.164 brouard 12411:
1.273 brouard 12412: double dateprev1, dateprev2;
1.296 brouard 12413: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
12414: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
12415:
1.217 brouard 12416:
1.136 brouard 12417: double **ximort;
1.145 brouard 12418: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 12419: int *dcwave;
12420:
1.164 brouard 12421: char z[1]="c";
1.136 brouard 12422:
12423: /*char *strt;*/
12424: char strtend[80];
1.126 brouard 12425:
1.164 brouard 12426:
1.126 brouard 12427: /* setlocale (LC_ALL, ""); */
12428: /* bindtextdomain (PACKAGE, LOCALEDIR); */
12429: /* textdomain (PACKAGE); */
12430: /* setlocale (LC_CTYPE, ""); */
12431: /* setlocale (LC_MESSAGES, ""); */
12432:
12433: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 12434: rstart_time = time(NULL);
12435: /* (void) gettimeofday(&start_time,&tzp);*/
12436: start_time = *localtime(&rstart_time);
1.126 brouard 12437: curr_time=start_time;
1.157 brouard 12438: /*tml = *localtime(&start_time.tm_sec);*/
12439: /* strcpy(strstart,asctime(&tml)); */
12440: strcpy(strstart,asctime(&start_time));
1.126 brouard 12441:
12442: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 12443: /* tp.tm_sec = tp.tm_sec +86400; */
12444: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 12445: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
12446: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
12447: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 12448: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 12449: /* strt=asctime(&tmg); */
12450: /* printf("Time(after) =%s",strstart); */
12451: /* (void) time (&time_value);
12452: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
12453: * tm = *localtime(&time_value);
12454: * strstart=asctime(&tm);
12455: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
12456: */
12457:
12458: nberr=0; /* Number of errors and warnings */
12459: nbwarn=0;
1.184 brouard 12460: #ifdef WIN32
12461: _getcwd(pathcd, size);
12462: #else
1.126 brouard 12463: getcwd(pathcd, size);
1.184 brouard 12464: #endif
1.191 brouard 12465: syscompilerinfo(0);
1.196 brouard 12466: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 12467: if(argc <=1){
12468: printf("\nEnter the parameter file name: ");
1.205 brouard 12469: if(!fgets(pathr,FILENAMELENGTH,stdin)){
12470: printf("ERROR Empty parameter file name\n");
12471: goto end;
12472: }
1.126 brouard 12473: i=strlen(pathr);
12474: if(pathr[i-1]=='\n')
12475: pathr[i-1]='\0';
1.156 brouard 12476: i=strlen(pathr);
1.205 brouard 12477: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 12478: pathr[i-1]='\0';
1.205 brouard 12479: }
12480: i=strlen(pathr);
12481: if( i==0 ){
12482: printf("ERROR Empty parameter file name\n");
12483: goto end;
12484: }
12485: for (tok = pathr; tok != NULL; ){
1.126 brouard 12486: printf("Pathr |%s|\n",pathr);
12487: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
12488: printf("val= |%s| pathr=%s\n",val,pathr);
12489: strcpy (pathtot, val);
12490: if(pathr[0] == '\0') break; /* Dirty */
12491: }
12492: }
1.281 brouard 12493: else if (argc<=2){
12494: strcpy(pathtot,argv[1]);
12495: }
1.126 brouard 12496: else{
12497: strcpy(pathtot,argv[1]);
1.281 brouard 12498: strcpy(z,argv[2]);
12499: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 12500: }
12501: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
12502: /*cygwin_split_path(pathtot,path,optionfile);
12503: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
12504: /* cutv(path,optionfile,pathtot,'\\');*/
12505:
12506: /* Split argv[0], imach program to get pathimach */
12507: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
12508: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12509: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12510: /* strcpy(pathimach,argv[0]); */
12511: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
12512: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
12513: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 12514: #ifdef WIN32
12515: _chdir(path); /* Can be a relative path */
12516: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
12517: #else
1.126 brouard 12518: chdir(path); /* Can be a relative path */
1.184 brouard 12519: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
12520: #endif
12521: printf("Current directory %s!\n",pathcd);
1.126 brouard 12522: strcpy(command,"mkdir ");
12523: strcat(command,optionfilefiname);
12524: if((outcmd=system(command)) != 0){
1.169 brouard 12525: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 12526: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
12527: /* fclose(ficlog); */
12528: /* exit(1); */
12529: }
12530: /* if((imk=mkdir(optionfilefiname))<0){ */
12531: /* perror("mkdir"); */
12532: /* } */
12533:
12534: /*-------- arguments in the command line --------*/
12535:
1.186 brouard 12536: /* Main Log file */
1.126 brouard 12537: strcat(filelog, optionfilefiname);
12538: strcat(filelog,".log"); /* */
12539: if((ficlog=fopen(filelog,"w"))==NULL) {
12540: printf("Problem with logfile %s\n",filelog);
12541: goto end;
12542: }
12543: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 12544: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 12545: fprintf(ficlog,"\nEnter the parameter file name: \n");
12546: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
12547: path=%s \n\
12548: optionfile=%s\n\
12549: optionfilext=%s\n\
1.156 brouard 12550: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 12551:
1.197 brouard 12552: syscompilerinfo(1);
1.167 brouard 12553:
1.126 brouard 12554: printf("Local time (at start):%s",strstart);
12555: fprintf(ficlog,"Local time (at start): %s",strstart);
12556: fflush(ficlog);
12557: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 12558: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 12559:
12560: /* */
12561: strcpy(fileres,"r");
12562: strcat(fileres, optionfilefiname);
1.201 brouard 12563: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 12564: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 12565: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 12566:
1.186 brouard 12567: /* Main ---------arguments file --------*/
1.126 brouard 12568:
12569: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 12570: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
12571: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 12572: fflush(ficlog);
1.149 brouard 12573: /* goto end; */
12574: exit(70);
1.126 brouard 12575: }
12576:
12577: strcpy(filereso,"o");
1.201 brouard 12578: strcat(filereso,fileresu);
1.126 brouard 12579: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
12580: printf("Problem with Output resultfile: %s\n", filereso);
12581: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
12582: fflush(ficlog);
12583: goto end;
12584: }
1.278 brouard 12585: /*-------- Rewriting parameter file ----------*/
12586: strcpy(rfileres,"r"); /* "Rparameterfile */
12587: strcat(rfileres,optionfilefiname); /* Parameter file first name */
12588: strcat(rfileres,"."); /* */
12589: strcat(rfileres,optionfilext); /* Other files have txt extension */
12590: if((ficres =fopen(rfileres,"w"))==NULL) {
12591: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
12592: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
12593: fflush(ficlog);
12594: goto end;
12595: }
12596: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 12597:
1.278 brouard 12598:
1.126 brouard 12599: /* Reads comments: lines beginning with '#' */
12600: numlinepar=0;
1.277 brouard 12601: /* Is it a BOM UTF-8 Windows file? */
12602: /* First parameter line */
1.197 brouard 12603: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 12604: noffset=0;
12605: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12606: {
12607: noffset=noffset+3;
12608: printf("# File is an UTF8 Bom.\n"); // 0xBF
12609: }
1.302 brouard 12610: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12611: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 12612: {
12613: noffset=noffset+2;
12614: printf("# File is an UTF16BE BOM file\n");
12615: }
12616: else if( line[0] == 0 && line[1] == 0)
12617: {
12618: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12619: noffset=noffset+4;
12620: printf("# File is an UTF16BE BOM file\n");
12621: }
12622: } else{
12623: ;/*printf(" Not a BOM file\n");*/
12624: }
12625:
1.197 brouard 12626: /* If line starts with a # it is a comment */
1.277 brouard 12627: if (line[noffset] == '#') {
1.197 brouard 12628: numlinepar++;
12629: fputs(line,stdout);
12630: fputs(line,ficparo);
1.278 brouard 12631: fputs(line,ficres);
1.197 brouard 12632: fputs(line,ficlog);
12633: continue;
12634: }else
12635: break;
12636: }
12637: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
12638: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
12639: if (num_filled != 5) {
12640: printf("Should be 5 parameters\n");
1.283 brouard 12641: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 12642: }
1.126 brouard 12643: numlinepar++;
1.197 brouard 12644: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 12645: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12646: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12647: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 12648: }
12649: /* Second parameter line */
12650: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 12651: /* while(fscanf(ficpar,"%[^\n]", line)) { */
12652: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 12653: if (line[0] == '#') {
12654: numlinepar++;
1.283 brouard 12655: printf("%s",line);
12656: fprintf(ficres,"%s",line);
12657: fprintf(ficparo,"%s",line);
12658: fprintf(ficlog,"%s",line);
1.197 brouard 12659: continue;
12660: }else
12661: break;
12662: }
1.223 brouard 12663: if((num_filled=sscanf(line,"ftol=%lf stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n", \
12664: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
12665: if (num_filled != 11) {
12666: printf("Not 11 parameters, for example:ftol=1.e-8 stepm=12 ncovcol=2 nqv=1 ntv=2 nqtv=1 nlstate=2 ndeath=1 maxwav=3 mle=1 weight=1\n");
1.209 brouard 12667: printf("but line=%s\n",line);
1.283 brouard 12668: fprintf(ficlog,"Not 11 parameters, for example:ftol=1.e-8 stepm=12 ncovcol=2 nqv=1 ntv=2 nqtv=1 nlstate=2 ndeath=1 maxwav=3 mle=1 weight=1\n");
12669: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 12670: }
1.286 brouard 12671: if( lastpass > maxwav){
12672: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12673: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12674: fflush(ficlog);
12675: goto end;
12676: }
12677: printf("ftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n",ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, mle, weightopt);
1.283 brouard 12678: fprintf(ficparo,"ftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n",ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, mle, weightopt);
1.286 brouard 12679: fprintf(ficres,"ftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n",ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, 0, weightopt);
1.283 brouard 12680: fprintf(ficlog,"ftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\n",ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, mle, weightopt);
1.126 brouard 12681: }
1.203 brouard 12682: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 12683: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 12684: /* Third parameter line */
12685: while(fgets(line, MAXLINE, ficpar)) {
12686: /* If line starts with a # it is a comment */
12687: if (line[0] == '#') {
12688: numlinepar++;
1.283 brouard 12689: printf("%s",line);
12690: fprintf(ficres,"%s",line);
12691: fprintf(ficparo,"%s",line);
12692: fprintf(ficlog,"%s",line);
1.197 brouard 12693: continue;
12694: }else
12695: break;
12696: }
1.201 brouard 12697: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 12698: if (num_filled != 1){
1.302 brouard 12699: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
12700: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 12701: model[0]='\0';
12702: goto end;
12703: }
12704: else{
12705: if (model[0]=='+'){
12706: for(i=1; i<=strlen(model);i++)
12707: modeltemp[i-1]=model[i];
1.201 brouard 12708: strcpy(model,modeltemp);
1.197 brouard 12709: }
12710: }
1.338 brouard 12711: /* printf(" model=1+age%s modeltemp= %s, model=1+age+%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 12712: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 12713: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
12714: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
12715: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 12716: }
12717: /* fscanf(ficpar,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%lf stepm=%d ncovcol=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d model=1+age+%s\n",title, datafile, &lastobs, &firstpass,&lastpass,&ftol, &stepm, &ncovcol, &nlstate,&ndeath, &maxwav, &mle, &weightopt,model); */
12718: /* numlinepar=numlinepar+3; /\* In general *\/ */
12719: /* printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%e stepm=%d ncovcol=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\nmodel=1+age+%s\n", title, datafile, lastobs, firstpass,lastpass,ftol, stepm, ncovcol, nlstate,ndeath, maxwav, mle, weightopt,model); */
1.283 brouard 12720: /* fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\nmodel=1+age+%s.\n", title, datafile, lastobs, firstpass,lastpass,ftol,stepm,ncovcol, nqv, ntv, nqtv, nlstate,ndeath,maxwav, mle, weightopt,model); */
12721: /* fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle=%d weight=%d\nmodel=1+age+%s.\n", title, datafile, lastobs, firstpass,lastpass,ftol,stepm,ncovcol, nqv, ntv, nqtv, nlstate,ndeath,maxwav, mle, weightopt,model); */
1.126 brouard 12722: fflush(ficlog);
1.190 brouard 12723: /* if(model[0]=='#'|| model[0]== '\0'){ */
12724: if(model[0]=='#'){
1.279 brouard 12725: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
12726: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
12727: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 12728: if(mle != -1){
1.279 brouard 12729: printf("Fix the model line and run imach with mle=-1 to get a correct template of the parameter vectors and subdiagonal covariance matrix.\n");
1.187 brouard 12730: exit(1);
12731: }
12732: }
1.126 brouard 12733: while((c=getc(ficpar))=='#' && c!= EOF){
12734: ungetc(c,ficpar);
12735: fgets(line, MAXLINE, ficpar);
12736: numlinepar++;
1.195 brouard 12737: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
12738: z[0]=line[1];
1.342 brouard 12739: }else if(line[1]=='d'){ /* For debugging individual values of covariates in ficresilk */
1.343 brouard 12740: debugILK=1;printf("DebugILK\n");
1.195 brouard 12741: }
12742: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 12743: fputs(line, stdout);
12744: //puts(line);
1.126 brouard 12745: fputs(line,ficparo);
12746: fputs(line,ficlog);
12747: }
12748: ungetc(c,ficpar);
12749:
12750:
1.290 brouard 12751: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
12752: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
12753: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
1.341 brouard 12754: /* if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /\**< Time varying covariate (dummy and quantitative)*\/ */
12755: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs); /**< Might be better */
1.136 brouard 12756: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
12757: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
12758: v1+v2*age+v2*v3 makes cptcovn = 3
12759: */
12760: if (strlen(model)>1)
1.187 brouard 12761: ncovmodel=2+nbocc(model,'+')+1; /*Number of variables including intercept and age = cptcovn + intercept + age : v1+v2+v3+v2*v4+v5*age makes 5+2=7,age*age makes 3*/
1.145 brouard 12762: else
1.187 brouard 12763: ncovmodel=2; /* Constant and age */
1.133 brouard 12764: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
12765: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 12766: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
12767: printf("Too complex model for current IMaCh: npar=(nlstate+ndeath-1)*nlstate*ncovmodel=%d >= %d(MAXPARM) or nlstate=%d >= %d(NLSTATEMAX) or ndeath=%d >= %d(NDEATHMAX) or ncovmodel=(k+age+#of+signs)=%d(NCOVMAX) >= %d\n",npar, MAXPARM, nlstate, NLSTATEMAX, ndeath, NDEATHMAX, ncovmodel, NCOVMAX);
12768: fprintf(ficlog,"Too complex model for current IMaCh: %d >=%d(MAXPARM) or %d >=%d(NLSTATEMAX) or %d >=%d(NDEATHMAX) or %d(NCOVMAX) >=%d\n",npar, MAXPARM, nlstate, NLSTATEMAX, ndeath, NDEATHMAX, ncovmodel, NCOVMAX);
12769: fflush(stdout);
12770: fclose (ficlog);
12771: goto end;
12772: }
1.126 brouard 12773: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12774: delti=delti3[1][1];
12775: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
12776: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 12777: /* We could also provide initial parameters values giving by simple logistic regression
12778: * only one way, that is without matrix product. We will have nlstate maximizations */
12779: /* for(i=1;i<nlstate;i++){ */
12780: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12781: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12782: /* } */
1.126 brouard 12783: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 12784: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
12785: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12786: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12787: fclose (ficparo);
12788: fclose (ficlog);
12789: goto end;
12790: exit(0);
1.220 brouard 12791: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 12792: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 12793: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
12794: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12795: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12796: matcov=matrix(1,npar,1,npar);
1.203 brouard 12797: hess=matrix(1,npar,1,npar);
1.220 brouard 12798: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 12799: /* Read guessed parameters */
1.126 brouard 12800: /* Reads comments: lines beginning with '#' */
12801: while((c=getc(ficpar))=='#' && c!= EOF){
12802: ungetc(c,ficpar);
12803: fgets(line, MAXLINE, ficpar);
12804: numlinepar++;
1.141 brouard 12805: fputs(line,stdout);
1.126 brouard 12806: fputs(line,ficparo);
12807: fputs(line,ficlog);
12808: }
12809: ungetc(c,ficpar);
12810:
12811: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 12812: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 12813: for(i=1; i <=nlstate; i++){
1.234 brouard 12814: j=0;
1.126 brouard 12815: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 12816: if(jj==i) continue;
12817: j++;
1.292 brouard 12818: while((c=getc(ficpar))=='#' && c!= EOF){
12819: ungetc(c,ficpar);
12820: fgets(line, MAXLINE, ficpar);
12821: numlinepar++;
12822: fputs(line,stdout);
12823: fputs(line,ficparo);
12824: fputs(line,ficlog);
12825: }
12826: ungetc(c,ficpar);
1.234 brouard 12827: fscanf(ficpar,"%1d%1d",&i1,&j1);
12828: if ((i1 != i) || (j1 != jj)){
12829: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 12830: It might be a problem of design; if ncovcol and the model are correct\n \
12831: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 12832: exit(1);
12833: }
12834: fprintf(ficparo,"%1d%1d",i1,j1);
12835: if(mle==1)
12836: printf("%1d%1d",i,jj);
12837: fprintf(ficlog,"%1d%1d",i,jj);
12838: for(k=1; k<=ncovmodel;k++){
12839: fscanf(ficpar," %lf",¶m[i][j][k]);
12840: if(mle==1){
12841: printf(" %lf",param[i][j][k]);
12842: fprintf(ficlog," %lf",param[i][j][k]);
12843: }
12844: else
12845: fprintf(ficlog," %lf",param[i][j][k]);
12846: fprintf(ficparo," %lf",param[i][j][k]);
12847: }
12848: fscanf(ficpar,"\n");
12849: numlinepar++;
12850: if(mle==1)
12851: printf("\n");
12852: fprintf(ficlog,"\n");
12853: fprintf(ficparo,"\n");
1.126 brouard 12854: }
12855: }
12856: fflush(ficlog);
1.234 brouard 12857:
1.251 brouard 12858: /* Reads parameters values */
1.126 brouard 12859: p=param[1][1];
1.251 brouard 12860: pstart=paramstart[1][1];
1.126 brouard 12861:
12862: /* Reads comments: lines beginning with '#' */
12863: while((c=getc(ficpar))=='#' && c!= EOF){
12864: ungetc(c,ficpar);
12865: fgets(line, MAXLINE, ficpar);
12866: numlinepar++;
1.141 brouard 12867: fputs(line,stdout);
1.126 brouard 12868: fputs(line,ficparo);
12869: fputs(line,ficlog);
12870: }
12871: ungetc(c,ficpar);
12872:
12873: for(i=1; i <=nlstate; i++){
12874: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 12875: fscanf(ficpar,"%1d%1d",&i1,&j1);
12876: if ( (i1-i) * (j1-j) != 0){
12877: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
12878: exit(1);
12879: }
12880: printf("%1d%1d",i,j);
12881: fprintf(ficparo,"%1d%1d",i1,j1);
12882: fprintf(ficlog,"%1d%1d",i1,j1);
12883: for(k=1; k<=ncovmodel;k++){
12884: fscanf(ficpar,"%le",&delti3[i][j][k]);
12885: printf(" %le",delti3[i][j][k]);
12886: fprintf(ficparo," %le",delti3[i][j][k]);
12887: fprintf(ficlog," %le",delti3[i][j][k]);
12888: }
12889: fscanf(ficpar,"\n");
12890: numlinepar++;
12891: printf("\n");
12892: fprintf(ficparo,"\n");
12893: fprintf(ficlog,"\n");
1.126 brouard 12894: }
12895: }
12896: fflush(ficlog);
1.234 brouard 12897:
1.145 brouard 12898: /* Reads covariance matrix */
1.126 brouard 12899: delti=delti3[1][1];
1.220 brouard 12900:
12901:
1.126 brouard 12902: /* free_ma3x(delti3,1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */ /* Hasn't to to freed here otherwise delti is no more allocated */
1.220 brouard 12903:
1.126 brouard 12904: /* Reads comments: lines beginning with '#' */
12905: while((c=getc(ficpar))=='#' && c!= EOF){
12906: ungetc(c,ficpar);
12907: fgets(line, MAXLINE, ficpar);
12908: numlinepar++;
1.141 brouard 12909: fputs(line,stdout);
1.126 brouard 12910: fputs(line,ficparo);
12911: fputs(line,ficlog);
12912: }
12913: ungetc(c,ficpar);
1.220 brouard 12914:
1.126 brouard 12915: matcov=matrix(1,npar,1,npar);
1.203 brouard 12916: hess=matrix(1,npar,1,npar);
1.131 brouard 12917: for(i=1; i <=npar; i++)
12918: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12919:
1.194 brouard 12920: /* Scans npar lines */
1.126 brouard 12921: for(i=1; i <=npar; i++){
1.226 brouard 12922: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12923: if(count != 3){
1.226 brouard 12924: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12925: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12926: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12927: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12928: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12929: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12930: exit(1);
1.220 brouard 12931: }else{
1.226 brouard 12932: if(mle==1)
12933: printf("%1d%1d%d",i1,j1,jk);
12934: }
12935: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12936: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12937: for(j=1; j <=i; j++){
1.226 brouard 12938: fscanf(ficpar," %le",&matcov[i][j]);
12939: if(mle==1){
12940: printf(" %.5le",matcov[i][j]);
12941: }
12942: fprintf(ficlog," %.5le",matcov[i][j]);
12943: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12944: }
12945: fscanf(ficpar,"\n");
12946: numlinepar++;
12947: if(mle==1)
1.220 brouard 12948: printf("\n");
1.126 brouard 12949: fprintf(ficlog,"\n");
12950: fprintf(ficparo,"\n");
12951: }
1.194 brouard 12952: /* End of read covariance matrix npar lines */
1.126 brouard 12953: for(i=1; i <=npar; i++)
12954: for(j=i+1;j<=npar;j++)
1.226 brouard 12955: matcov[i][j]=matcov[j][i];
1.126 brouard 12956:
12957: if(mle==1)
12958: printf("\n");
12959: fprintf(ficlog,"\n");
12960:
12961: fflush(ficlog);
12962:
12963: } /* End of mle != -3 */
1.218 brouard 12964:
1.186 brouard 12965: /* Main data
12966: */
1.290 brouard 12967: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12968: /* num=lvector(1,n); */
12969: /* moisnais=vector(1,n); */
12970: /* annais=vector(1,n); */
12971: /* moisdc=vector(1,n); */
12972: /* andc=vector(1,n); */
12973: /* weight=vector(1,n); */
12974: /* agedc=vector(1,n); */
12975: /* cod=ivector(1,n); */
12976: /* for(i=1;i<=n;i++){ */
12977: num=lvector(firstobs,lastobs);
12978: moisnais=vector(firstobs,lastobs);
12979: annais=vector(firstobs,lastobs);
12980: moisdc=vector(firstobs,lastobs);
12981: andc=vector(firstobs,lastobs);
12982: weight=vector(firstobs,lastobs);
12983: agedc=vector(firstobs,lastobs);
12984: cod=ivector(firstobs,lastobs);
12985: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 12986: num[i]=0;
12987: moisnais[i]=0;
12988: annais[i]=0;
12989: moisdc[i]=0;
12990: andc[i]=0;
12991: agedc[i]=0;
12992: cod[i]=0;
12993: weight[i]=1.0; /* Equal weights, 1 by default */
12994: }
1.290 brouard 12995: mint=matrix(1,maxwav,firstobs,lastobs);
12996: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 12997: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.336 brouard 12998: /* printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel)); */
1.126 brouard 12999: tab=ivector(1,NCOVMAX);
1.144 brouard 13000: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 13001: ncodemaxwundef=ivector(1,NCOVMAX); /* Number of code per covariate; if - 1 O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.126 brouard 13002:
1.136 brouard 13003: /* Reads data from file datafile */
13004: if (readdata(datafile, firstobs, lastobs, &imx)==1)
13005: goto end;
13006:
13007: /* Calculation of the number of parameters from char model */
1.234 brouard 13008: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 13009: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
13010: k=3 V4 Tvar[k=3]= 4 (from V4)
13011: k=2 V1 Tvar[k=2]= 1 (from V1)
13012: k=1 Tvar[1]=2 (from V2)
1.234 brouard 13013: */
13014:
13015: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
13016: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 13017: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 brouard 13018: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 13019: TvarsD=ivector(1,NCOVMAX); /* */
13020: TvarsQind=ivector(1,NCOVMAX); /* */
13021: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 13022: TvarF=ivector(1,NCOVMAX); /* */
13023: TvarFind=ivector(1,NCOVMAX); /* */
13024: TvarV=ivector(1,NCOVMAX); /* */
13025: TvarVind=ivector(1,NCOVMAX); /* */
13026: TvarA=ivector(1,NCOVMAX); /* */
13027: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 13028: TvarFD=ivector(1,NCOVMAX); /* */
13029: TvarFDind=ivector(1,NCOVMAX); /* */
13030: TvarFQ=ivector(1,NCOVMAX); /* */
13031: TvarFQind=ivector(1,NCOVMAX); /* */
13032: TvarVD=ivector(1,NCOVMAX); /* */
13033: TvarVDind=ivector(1,NCOVMAX); /* */
13034: TvarVQ=ivector(1,NCOVMAX); /* */
13035: TvarVQind=ivector(1,NCOVMAX); /* */
1.339 brouard 13036: TvarVV=ivector(1,NCOVMAX); /* */
13037: TvarVVind=ivector(1,NCOVMAX); /* */
1.231 brouard 13038:
1.230 brouard 13039: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 13040: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 13041: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
13042: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
13043: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 13044: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
13045: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
13046: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
13047: */
13048: /* For model-covariate k tells which data-covariate to use but
13049: because this model-covariate is a construction we invent a new column
13050: ncovcol + k1
13051: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
13052: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 13053: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
13054: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 13055: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
13056: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 13057: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 13058: */
1.145 brouard 13059: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
13060: Tvard=imatrix(1,NCOVMAX,1,2); /* n=Tvard[k1][1] and m=Tvard[k1][2] gives the couple n,m of the k1 th product Vn*Vm
1.141 brouard 13061: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
13062: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 brouard 13063: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 13064: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 13065: 4 covariates (3 plus signs)
13066: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 13067: */
13068: for(i=1;i<NCOVMAX;i++)
13069: Tage[i]=0;
1.230 brouard 13070: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 13071: * individual dummy, fixed or varying:
13072: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
13073: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 13074: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
13075: * V1 df, V2 qf, V3 & V4 dv, V5 qv
13076: * Tmodelind[1]@9={9,0,3,2,}*/
13077: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
13078: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 13079: * individual quantitative, fixed or varying:
13080: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
13081: * 3, 1, 0, 0, 0, 0, 0, 0},
13082: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 13083: /* Main decodemodel */
13084:
1.187 brouard 13085:
1.223 brouard 13086: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 13087: goto end;
13088:
1.137 brouard 13089: if((double)(lastobs-imx)/(double)imx > 1.10){
13090: nbwarn++;
13091: printf("Warning: The value of parameter lastobs=%d is big compared to the \n effective number of cases imx=%d, please adjust, \n otherwise you are allocating more memory than necessary.\n",lastobs, imx);
13092: fprintf(ficlog,"Warning: The value of parameter lastobs=%d is big compared to the \n effective number of cases imx=%d, please adjust, \n otherwise you are allocating more memory than necessary.\n",lastobs, imx);
13093: }
1.136 brouard 13094: /* if(mle==1){*/
1.137 brouard 13095: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
13096: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 13097: }
13098:
13099: /*-calculation of age at interview from date of interview and age at death -*/
13100: agev=matrix(1,maxwav,1,imx);
13101:
13102: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
13103: goto end;
13104:
1.126 brouard 13105:
1.136 brouard 13106: agegomp=(int)agemin;
1.290 brouard 13107: free_vector(moisnais,firstobs,lastobs);
13108: free_vector(annais,firstobs,lastobs);
1.126 brouard 13109: /* free_matrix(mint,1,maxwav,1,n);
13110: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 13111: /* free_vector(moisdc,1,n); */
13112: /* free_vector(andc,1,n); */
1.145 brouard 13113: /* */
13114:
1.126 brouard 13115: wav=ivector(1,imx);
1.214 brouard 13116: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
13117: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
13118: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
13119: dh=imatrix(1,lastpass-firstpass+2,1,imx); /* We are adding a wave if status is unknown at last wave but death occurs after last wave.*/
13120: bh=imatrix(1,lastpass-firstpass+2,1,imx);
13121: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 13122:
13123: /* Concatenates waves */
1.214 brouard 13124: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
13125: Death is a valid wave (if date is known).
13126: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
13127: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
13128: and mw[mi+1][i]. dh depends on stepm.
13129: */
13130:
1.126 brouard 13131: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 13132: /* Concatenates waves */
1.145 brouard 13133:
1.290 brouard 13134: free_vector(moisdc,firstobs,lastobs);
13135: free_vector(andc,firstobs,lastobs);
1.215 brouard 13136:
1.126 brouard 13137: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
13138: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
13139: ncodemax[1]=1;
1.145 brouard 13140: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 13141: cptcoveff=0;
1.220 brouard 13142: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 brouard 13143: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; as well as calculate cptcoveff or number of total effective dummy covariates*/
1.227 brouard 13144: }
13145:
13146: ncovcombmax=pow(2,cptcoveff);
1.338 brouard 13147: invalidvarcomb=ivector(0, ncovcombmax);
13148: for(i=0;i<ncovcombmax;i++)
1.227 brouard 13149: invalidvarcomb[i]=0;
13150:
1.211 brouard 13151: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 13152: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 13153: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 13154:
1.200 brouard 13155: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 13156: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 13157: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 13158: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
13159: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
13160: * (currently 0 or 1) in the data.
13161: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
13162: * corresponding modality (h,j).
13163: */
13164:
1.145 brouard 13165: h=0;
13166: /*if (cptcovn > 0) */
1.126 brouard 13167: m=pow(2,cptcoveff);
13168:
1.144 brouard 13169: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 13170: * For k=4 covariates, h goes from 1 to m=2**k
13171: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
13172: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 13173: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
13174: *______________________________ *______________________
13175: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
13176: * 2 2 1 1 1 * 1 0 0 0 1
13177: * 3 i=2 1 2 1 1 * 2 0 0 1 0
13178: * 4 2 2 1 1 * 3 0 0 1 1
13179: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
13180: * 6 2 1 2 1 * 5 0 1 0 1
13181: * 7 i=4 1 2 2 1 * 6 0 1 1 0
13182: * 8 2 2 2 1 * 7 0 1 1 1
13183: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
13184: * 10 2 1 1 2 * 9 1 0 0 1
13185: * 11 i=6 1 2 1 2 * 10 1 0 1 0
13186: * 12 2 2 1 2 * 11 1 0 1 1
13187: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
13188: * 14 2 1 2 2 * 13 1 1 0 1
13189: * 15 i=8 1 2 2 2 * 14 1 1 1 0
13190: * 16 2 2 2 2 * 15 1 1 1 1
13191: */
1.212 brouard 13192: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 13193: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
13194: * and the value of each covariate?
13195: * V1=1, V2=1, V3=2, V4=1 ?
13196: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
13197: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
13198: * In order to get the real value in the data, we use nbcode
13199: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
13200: * We are keeping this crazy system in order to be able (in the future?)
13201: * to have more than 2 values (0 or 1) for a covariate.
13202: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
13203: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
13204: * bbbbbbbb
13205: * 76543210
13206: * h-1 00000101 (6-1=5)
1.219 brouard 13207: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 13208: * &
13209: * 1 00000001 (1)
1.219 brouard 13210: * 00000000 = 1 & ((h-1) >> (k-1))
13211: * +1= 00000001 =1
1.211 brouard 13212: *
13213: * h=14, k=3 => h'=h-1=13, k'=k-1=2
13214: * h' 1101 =2^3+2^2+0x2^1+2^0
13215: * >>k' 11
13216: * & 00000001
13217: * = 00000001
13218: * +1 = 00000010=2 = codtabm(14,3)
13219: * Reverse h=6 and m=16?
13220: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
13221: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
13222: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
13223: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
13224: * V3=decodtabm(14,3,2**4)=2
13225: * h'=13 1101 =2^3+2^2+0x2^1+2^0
13226: *(h-1) >> (j-1) 0011 =13 >> 2
13227: * &1 000000001
13228: * = 000000001
13229: * +1= 000000010 =2
13230: * 2211
13231: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
13232: * V3=2
1.220 brouard 13233: * codtabm and decodtabm are identical
1.211 brouard 13234: */
13235:
1.145 brouard 13236:
13237: free_ivector(Ndum,-1,NCOVMAX);
13238:
13239:
1.126 brouard 13240:
1.186 brouard 13241: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 13242: strcpy(optionfilegnuplot,optionfilefiname);
13243: if(mle==-3)
1.201 brouard 13244: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 13245: strcat(optionfilegnuplot,".gp");
13246:
13247: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
13248: printf("Problem with file %s",optionfilegnuplot);
13249: }
13250: else{
1.204 brouard 13251: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 13252: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 13253: //fprintf(ficgp,"set missing 'NaNq'\n");
13254: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 13255: }
13256: /* fclose(ficgp);*/
1.186 brouard 13257:
13258:
13259: /* Initialisation of --------- index.htm --------*/
1.126 brouard 13260:
13261: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
13262: if(mle==-3)
1.201 brouard 13263: strcat(optionfilehtm,"-MORT_");
1.126 brouard 13264: strcat(optionfilehtm,".htm");
13265: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 13266: printf("Problem with %s \n",optionfilehtm);
13267: exit(0);
1.126 brouard 13268: }
13269:
13270: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
13271: strcat(optionfilehtmcov,"-cov.htm");
13272: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
13273: printf("Problem with %s \n",optionfilehtmcov), exit(0);
13274: }
13275: else{
13276: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
13277: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13278: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 13279: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
13280: }
13281:
1.335 brouard 13282: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
13283: <title>IMaCh %s</title></head>\n\
13284: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
13285: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
13286: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
13287: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
13288: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
13289:
13290: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13291: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 13292: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.337 brouard 13293: This file: <a href=\"%s\">%s</a></br>Title=%s <br>Datafile=<a href=\"%s\">%s</a> Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 13294: \n\
13295: <hr size=\"2\" color=\"#EC5E5E\">\
13296: <ul><li><h4>Parameter files</h4>\n\
13297: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
13298: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
13299: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
13300: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
13301: - Date and time at start: %s</ul>\n",\
1.335 brouard 13302: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 13303: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
13304: fileres,fileres,\
13305: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
13306: fflush(fichtm);
13307:
13308: strcpy(pathr,path);
13309: strcat(pathr,optionfilefiname);
1.184 brouard 13310: #ifdef WIN32
13311: _chdir(optionfilefiname); /* Move to directory named optionfile */
13312: #else
1.126 brouard 13313: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 13314: #endif
13315:
1.126 brouard 13316:
1.220 brouard 13317: /* Calculates basic frequencies. Computes observed prevalence at single age
13318: and for any valid combination of covariates
1.126 brouard 13319: and prints on file fileres'p'. */
1.251 brouard 13320: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 13321: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 13322:
13323: fprintf(fichtm,"\n");
1.286 brouard 13324: fprintf(fichtm,"<h4>Parameter line 2</h4><ul><li>Tolerance for the convergence of the likelihood: ftol=%g \n<li>Interval for the elementary matrix (in month): stepm=%d",\
1.274 brouard 13325: ftol, stepm);
13326: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
13327: ncurrv=1;
13328: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
13329: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
13330: ncurrv=i;
13331: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13332: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 13333: ncurrv=i;
13334: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13335: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 13336: ncurrv=i;
13337: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
13338: fprintf(fichtm,"\n<li>Weights column \n<br>Number of alive states: nlstate=%d <br>Number of death states (not really implemented): ndeath=%d \n<li>Number of waves: maxwav=%d \n<li>Parameter for maximization (1), using parameter values (0), for design of parameters and variance-covariance matrix: mle=%d \n<li>Does the weight column be taken into account (1), or not (0): weight=%d</ul>\n", \
13339: nlstate, ndeath, maxwav, mle, weightopt);
13340:
13341: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
13342: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
13343:
13344:
1.317 brouard 13345: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 13346: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
13347: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 13348: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 13349: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 13350: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13351: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13352: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13353: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 13354:
1.126 brouard 13355: /* For Powell, parameters are in a vector p[] starting at p[1]
13356: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
13357: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
13358:
13359: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 13360: /* For mortality only */
1.126 brouard 13361: if (mle==-3){
1.136 brouard 13362: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 13363: for(i=1;i<=NDIM;i++)
13364: for(j=1;j<=NDIM;j++)
13365: ximort[i][j]=0.;
1.186 brouard 13366: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 13367: cens=ivector(firstobs,lastobs);
13368: ageexmed=vector(firstobs,lastobs);
13369: agecens=vector(firstobs,lastobs);
13370: dcwave=ivector(firstobs,lastobs);
1.223 brouard 13371:
1.126 brouard 13372: for (i=1; i<=imx; i++){
13373: dcwave[i]=-1;
13374: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 13375: if (s[m][i]>nlstate) {
13376: dcwave[i]=m;
13377: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
13378: break;
13379: }
1.126 brouard 13380: }
1.226 brouard 13381:
1.126 brouard 13382: for (i=1; i<=imx; i++) {
13383: if (wav[i]>0){
1.226 brouard 13384: ageexmed[i]=agev[mw[1][i]][i];
13385: j=wav[i];
13386: agecens[i]=1.;
13387:
13388: if (ageexmed[i]> 1 && wav[i] > 0){
13389: agecens[i]=agev[mw[j][i]][i];
13390: cens[i]= 1;
13391: }else if (ageexmed[i]< 1)
13392: cens[i]= -1;
13393: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
13394: cens[i]=0 ;
1.126 brouard 13395: }
13396: else cens[i]=-1;
13397: }
13398:
13399: for (i=1;i<=NDIM;i++) {
13400: for (j=1;j<=NDIM;j++)
1.226 brouard 13401: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 13402: }
13403:
1.302 brouard 13404: p[1]=0.0268; p[NDIM]=0.083;
13405: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 13406:
13407:
1.136 brouard 13408: #ifdef GSL
13409: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 13410: #else
1.126 brouard 13411: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 13412: #endif
1.201 brouard 13413: strcpy(filerespow,"POW-MORT_");
13414: strcat(filerespow,fileresu);
1.126 brouard 13415: if((ficrespow=fopen(filerespow,"w"))==NULL) {
13416: printf("Problem with resultfile: %s\n", filerespow);
13417: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
13418: }
1.136 brouard 13419: #ifdef GSL
13420: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 13421: #else
1.126 brouard 13422: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 13423: #endif
1.126 brouard 13424: /* for (i=1;i<=nlstate;i++)
13425: for(j=1;j<=nlstate+ndeath;j++)
13426: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
13427: */
13428: fprintf(ficrespow,"\n");
1.136 brouard 13429: #ifdef GSL
13430: /* gsl starts here */
13431: T = gsl_multimin_fminimizer_nmsimplex;
13432: gsl_multimin_fminimizer *sfm = NULL;
13433: gsl_vector *ss, *x;
13434: gsl_multimin_function minex_func;
13435:
13436: /* Initial vertex size vector */
13437: ss = gsl_vector_alloc (NDIM);
13438:
13439: if (ss == NULL){
13440: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
13441: }
13442: /* Set all step sizes to 1 */
13443: gsl_vector_set_all (ss, 0.001);
13444:
13445: /* Starting point */
1.126 brouard 13446:
1.136 brouard 13447: x = gsl_vector_alloc (NDIM);
13448:
13449: if (x == NULL){
13450: gsl_vector_free(ss);
13451: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
13452: }
13453:
13454: /* Initialize method and iterate */
13455: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 13456: /* gsl_vector_set(x, 0, 0.0268); */
13457: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 13458: gsl_vector_set(x, 0, p[1]);
13459: gsl_vector_set(x, 1, p[2]);
13460:
13461: minex_func.f = &gompertz_f;
13462: minex_func.n = NDIM;
13463: minex_func.params = (void *)&p; /* ??? */
13464:
13465: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
13466: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
13467:
13468: printf("Iterations beginning .....\n\n");
13469: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
13470:
13471: iteri=0;
13472: while (rval == GSL_CONTINUE){
13473: iteri++;
13474: status = gsl_multimin_fminimizer_iterate(sfm);
13475:
13476: if (status) printf("error: %s\n", gsl_strerror (status));
13477: fflush(0);
13478:
13479: if (status)
13480: break;
13481:
13482: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
13483: ssval = gsl_multimin_fminimizer_size (sfm);
13484:
13485: if (rval == GSL_SUCCESS)
13486: printf ("converged to a local maximum at\n");
13487:
13488: printf("%5d ", iteri);
13489: for (it = 0; it < NDIM; it++){
13490: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
13491: }
13492: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
13493: }
13494:
13495: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
13496:
13497: gsl_vector_free(x); /* initial values */
13498: gsl_vector_free(ss); /* inital step size */
13499: for (it=0; it<NDIM; it++){
13500: p[it+1]=gsl_vector_get(sfm->x,it);
13501: fprintf(ficrespow," %.12lf", p[it]);
13502: }
13503: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
13504: #endif
13505: #ifdef POWELL
13506: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
13507: #endif
1.126 brouard 13508: fclose(ficrespow);
13509:
1.203 brouard 13510: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 13511:
13512: for(i=1; i <=NDIM; i++)
13513: for(j=i+1;j<=NDIM;j++)
1.220 brouard 13514: matcov[i][j]=matcov[j][i];
1.126 brouard 13515:
13516: printf("\nCovariance matrix\n ");
1.203 brouard 13517: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 13518: for(i=1; i <=NDIM; i++) {
13519: for(j=1;j<=NDIM;j++){
1.220 brouard 13520: printf("%f ",matcov[i][j]);
13521: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 13522: }
1.203 brouard 13523: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 13524: }
13525:
13526: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 13527: for (i=1;i<=NDIM;i++) {
1.126 brouard 13528: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 13529: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
13530: }
1.302 brouard 13531: lsurv=vector(agegomp,AGESUP);
13532: lpop=vector(agegomp,AGESUP);
13533: tpop=vector(agegomp,AGESUP);
1.126 brouard 13534: lsurv[agegomp]=100000;
13535:
13536: for (k=agegomp;k<=AGESUP;k++) {
13537: agemortsup=k;
13538: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
13539: }
13540:
13541: for (k=agegomp;k<agemortsup;k++)
13542: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
13543:
13544: for (k=agegomp;k<agemortsup;k++){
13545: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
13546: sumlpop=sumlpop+lpop[k];
13547: }
13548:
13549: tpop[agegomp]=sumlpop;
13550: for (k=agegomp;k<(agemortsup-3);k++){
13551: /* tpop[k+1]=2;*/
13552: tpop[k+1]=tpop[k]-lpop[k];
13553: }
13554:
13555:
13556: printf("\nAge lx qx dx Lx Tx e(x)\n");
13557: for (k=agegomp;k<(agemortsup-2);k++)
13558: printf("%d %.0lf %lf %.0lf %.0lf %.0lf %lf\n",k,lsurv[k],p[1]*exp(p[2]*(k-agegomp)),(p[1]*exp(p[2]*(k-agegomp)))*lsurv[k],lpop[k],tpop[k],tpop[k]/lsurv[k]);
13559:
13560:
13561: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 13562: ageminpar=50;
13563: agemaxpar=100;
1.194 brouard 13564: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
13565: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13566: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13567: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
13568: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13569: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13570: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13571: }else{
13572: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
13573: fprintf(ficlog,"Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
1.201 brouard 13574: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 13575: }
1.201 brouard 13576: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 13577: stepm, weightopt,\
13578: model,imx,p,matcov,agemortsup);
13579:
1.302 brouard 13580: free_vector(lsurv,agegomp,AGESUP);
13581: free_vector(lpop,agegomp,AGESUP);
13582: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 13583: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 13584: free_ivector(dcwave,firstobs,lastobs);
13585: free_vector(agecens,firstobs,lastobs);
13586: free_vector(ageexmed,firstobs,lastobs);
13587: free_ivector(cens,firstobs,lastobs);
1.220 brouard 13588: #ifdef GSL
1.136 brouard 13589: #endif
1.186 brouard 13590: } /* Endof if mle==-3 mortality only */
1.205 brouard 13591: /* Standard */
13592: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
13593: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13594: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 13595: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 13596: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
13597: for (k=1; k<=npar;k++)
13598: printf(" %d %8.5f",k,p[k]);
13599: printf("\n");
1.205 brouard 13600: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
13601: /* mlikeli uses func not funcone */
1.247 brouard 13602: /* for(i=1;i<nlstate;i++){ */
13603: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
13604: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
13605: /* } */
1.205 brouard 13606: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
13607: }
13608: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
13609: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13610: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
13611: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13612: }
13613: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 13614: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13615: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 brouard 13616: /* exit(0); */
1.126 brouard 13617: for (k=1; k<=npar;k++)
13618: printf(" %d %8.5f",k,p[k]);
13619: printf("\n");
13620:
13621: /*--------- results files --------------*/
1.283 brouard 13622: /* fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\nftol=%e stepm=%d ncovcol=%d nqv=%d ntv=%d nqtv=%d nlstate=%d ndeath=%d maxwav=%d mle= 0 weight=%d\nmodel=1+age+%s.\n", title, datafile, lastobs, firstpass,lastpass,ftol, stepm, ncovcol, nqv, ntv, nqtv, nlstate, ndeath, maxwav, weightopt,model); */
1.126 brouard 13623:
13624:
13625: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13626: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 13627: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13628:
13629: printf("#model= 1 + age ");
13630: fprintf(ficres,"#model= 1 + age ");
13631: fprintf(ficlog,"#model= 1 + age ");
13632: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
13633: </ul>", model);
13634:
13635: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
13636: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
13637: if(nagesqr==1){
13638: printf(" + age*age ");
13639: fprintf(ficres," + age*age ");
13640: fprintf(ficlog," + age*age ");
13641: fprintf(fichtm, "<th>+ age*age</th>");
13642: }
13643: for(j=1;j <=ncovmodel-2;j++){
13644: if(Typevar[j]==0) {
13645: printf(" + V%d ",Tvar[j]);
13646: fprintf(ficres," + V%d ",Tvar[j]);
13647: fprintf(ficlog," + V%d ",Tvar[j]);
13648: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13649: }else if(Typevar[j]==1) {
13650: printf(" + V%d*age ",Tvar[j]);
13651: fprintf(ficres," + V%d*age ",Tvar[j]);
13652: fprintf(ficlog," + V%d*age ",Tvar[j]);
13653: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13654: }else if(Typevar[j]==2) {
13655: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13656: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13657: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13658: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13659: }
13660: }
13661: printf("\n");
13662: fprintf(ficres,"\n");
13663: fprintf(ficlog,"\n");
13664: fprintf(fichtm, "</tr>");
13665: fprintf(fichtm, "\n");
13666:
13667:
1.126 brouard 13668: for(i=1,jk=1; i <=nlstate; i++){
13669: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 13670: if (k != i) {
1.319 brouard 13671: fprintf(fichtm, "<tr>");
1.225 brouard 13672: printf("%d%d ",i,k);
13673: fprintf(ficlog,"%d%d ",i,k);
13674: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 13675: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13676: for(j=1; j <=ncovmodel; j++){
13677: printf("%12.7f ",p[jk]);
13678: fprintf(ficlog,"%12.7f ",p[jk]);
13679: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 13680: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 13681: jk++;
13682: }
13683: printf("\n");
13684: fprintf(ficlog,"\n");
13685: fprintf(ficres,"\n");
1.319 brouard 13686: fprintf(fichtm, "</tr>\n");
1.225 brouard 13687: }
1.126 brouard 13688: }
13689: }
1.319 brouard 13690: /* fprintf(fichtm,"</tr>\n"); */
13691: fprintf(fichtm,"</table>\n");
13692: fprintf(fichtm, "\n");
13693:
1.203 brouard 13694: if(mle != 0){
13695: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 13696: ftolhess=ftol; /* Usually correct */
1.203 brouard 13697: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
13698: printf("Parameters and 95%% confidence intervals\n W is simply the result of the division of the parameter by the square root of covariance of the parameter.\n And Wald-based confidence intervals plus and minus 1.96 * W .\n But be careful that parameters are highly correlated because incidence of disability is highly correlated to incidence of recovery.\n It might be better to visualize the covariance matrix. See the page 'Matrix of variance-covariance of one-step probabilities' and its graphs.\n");
13699: fprintf(ficlog, "Parameters, Wald tests and Wald-based confidence intervals\n W is simply the result of the division of the parameter by the square root of covariance of the parameter.\n And Wald-based confidence intervals plus and minus 1.96 * W \n It might be better to visualize the covariance matrix. See the page 'Matrix of variance-covariance of one-step probabilities' and its graphs.\n");
1.322 brouard 13700: fprintf(fichtm, "\n<p>The Wald test results are output only if the maximimzation of the Likelihood is performed (mle=1)\n</br>Parameters, Wald tests and Wald-based confidence intervals\n</br> W is simply the result of the division of the parameter by the square root of covariance of the parameter.\n</br> And Wald-based confidence intervals plus and minus 1.96 * W \n </br> It might be better to visualize the covariance matrix. See the page '<a href=\"%s\">Matrix of variance-covariance of one-step probabilities and its graphs</a>'.\n</br>",optionfilehtmcov);
1.319 brouard 13701: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
13702: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
13703: if(nagesqr==1){
13704: printf(" + age*age ");
13705: fprintf(ficres," + age*age ");
13706: fprintf(ficlog," + age*age ");
13707: fprintf(fichtm, "<th>+ age*age</th>");
13708: }
13709: for(j=1;j <=ncovmodel-2;j++){
13710: if(Typevar[j]==0) {
13711: printf(" + V%d ",Tvar[j]);
13712: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13713: }else if(Typevar[j]==1) {
13714: printf(" + V%d*age ",Tvar[j]);
13715: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13716: }else if(Typevar[j]==2) {
13717: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13718: }
13719: }
13720: fprintf(fichtm, "</tr>\n");
13721:
1.203 brouard 13722: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 13723: for(k=1; k <=(nlstate+ndeath); k++){
13724: if (k != i) {
1.319 brouard 13725: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 13726: printf("%d%d ",i,k);
13727: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 13728: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13729: for(j=1; j <=ncovmodel; j++){
1.319 brouard 13730: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 13731: printf("%12.7f(%12.7f) W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk],sqrt(matcov[jk][jk]), p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
13732: fprintf(ficlog,"%12.7f(%12.7f) W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk],sqrt(matcov[jk][jk]), p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
1.319 brouard 13733: if(fabs(wald) > 1.96){
1.321 brouard 13734: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 13735: }else{
13736: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
13737: }
1.324 brouard 13738: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 13739: fprintf(fichtm,"[%12.7f;%12.7f]</br></td>", p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
1.225 brouard 13740: jk++;
13741: }
13742: printf("\n");
13743: fprintf(ficlog,"\n");
1.319 brouard 13744: fprintf(fichtm, "</tr>\n");
1.225 brouard 13745: }
13746: }
1.193 brouard 13747: }
1.203 brouard 13748: } /* end of hesscov and Wald tests */
1.319 brouard 13749: fprintf(fichtm,"</table>\n");
1.225 brouard 13750:
1.203 brouard 13751: /* */
1.126 brouard 13752: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
13753: printf("# Scales (for hessian or gradient estimation)\n");
13754: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
13755: for(i=1,jk=1; i <=nlstate; i++){
13756: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 13757: if (j!=i) {
13758: fprintf(ficres,"%1d%1d",i,j);
13759: printf("%1d%1d",i,j);
13760: fprintf(ficlog,"%1d%1d",i,j);
13761: for(k=1; k<=ncovmodel;k++){
13762: printf(" %.5e",delti[jk]);
13763: fprintf(ficlog," %.5e",delti[jk]);
13764: fprintf(ficres," %.5e",delti[jk]);
13765: jk++;
13766: }
13767: printf("\n");
13768: fprintf(ficlog,"\n");
13769: fprintf(ficres,"\n");
13770: }
1.126 brouard 13771: }
13772: }
13773:
13774: fprintf(ficres,"# Covariance matrix \n# 121 Var(a12)\n# 122 Cov(b12,a12) Var(b12)\n# ...\n# 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n");
1.203 brouard 13775: if(mle >= 1) /* To big for the screen */
1.126 brouard 13776: printf("# Covariance matrix \n# 121 Var(a12)\n# 122 Cov(b12,a12) Var(b12)\n# ...\n# 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n");
13777: fprintf(ficlog,"# Covariance matrix \n# 121 Var(a12)\n# 122 Cov(b12,a12) Var(b12)\n# ...\n# 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n");
13778: /* # 121 Var(a12)\n\ */
13779: /* # 122 Cov(b12,a12) Var(b12)\n\ */
13780: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
13781: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
13782: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
13783: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
13784: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
13785: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
13786:
13787:
13788: /* Just to have a covariance matrix which will be more understandable
13789: even is we still don't want to manage dictionary of variables
13790: */
13791: for(itimes=1;itimes<=2;itimes++){
13792: jj=0;
13793: for(i=1; i <=nlstate; i++){
1.225 brouard 13794: for(j=1; j <=nlstate+ndeath; j++){
13795: if(j==i) continue;
13796: for(k=1; k<=ncovmodel;k++){
13797: jj++;
13798: ca[0]= k+'a'-1;ca[1]='\0';
13799: if(itimes==1){
13800: if(mle>=1)
13801: printf("#%1d%1d%d",i,j,k);
13802: fprintf(ficlog,"#%1d%1d%d",i,j,k);
13803: fprintf(ficres,"#%1d%1d%d",i,j,k);
13804: }else{
13805: if(mle>=1)
13806: printf("%1d%1d%d",i,j,k);
13807: fprintf(ficlog,"%1d%1d%d",i,j,k);
13808: fprintf(ficres,"%1d%1d%d",i,j,k);
13809: }
13810: ll=0;
13811: for(li=1;li <=nlstate; li++){
13812: for(lj=1;lj <=nlstate+ndeath; lj++){
13813: if(lj==li) continue;
13814: for(lk=1;lk<=ncovmodel;lk++){
13815: ll++;
13816: if(ll<=jj){
13817: cb[0]= lk +'a'-1;cb[1]='\0';
13818: if(ll<jj){
13819: if(itimes==1){
13820: if(mle>=1)
13821: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13822: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13823: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13824: }else{
13825: if(mle>=1)
13826: printf(" %.5e",matcov[jj][ll]);
13827: fprintf(ficlog," %.5e",matcov[jj][ll]);
13828: fprintf(ficres," %.5e",matcov[jj][ll]);
13829: }
13830: }else{
13831: if(itimes==1){
13832: if(mle>=1)
13833: printf(" Var(%s%1d%1d)",ca,i,j);
13834: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
13835: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
13836: }else{
13837: if(mle>=1)
13838: printf(" %.7e",matcov[jj][ll]);
13839: fprintf(ficlog," %.7e",matcov[jj][ll]);
13840: fprintf(ficres," %.7e",matcov[jj][ll]);
13841: }
13842: }
13843: }
13844: } /* end lk */
13845: } /* end lj */
13846: } /* end li */
13847: if(mle>=1)
13848: printf("\n");
13849: fprintf(ficlog,"\n");
13850: fprintf(ficres,"\n");
13851: numlinepar++;
13852: } /* end k*/
13853: } /*end j */
1.126 brouard 13854: } /* end i */
13855: } /* end itimes */
13856:
13857: fflush(ficlog);
13858: fflush(ficres);
1.225 brouard 13859: while(fgets(line, MAXLINE, ficpar)) {
13860: /* If line starts with a # it is a comment */
13861: if (line[0] == '#') {
13862: numlinepar++;
13863: fputs(line,stdout);
13864: fputs(line,ficparo);
13865: fputs(line,ficlog);
1.299 brouard 13866: fputs(line,ficres);
1.225 brouard 13867: continue;
13868: }else
13869: break;
13870: }
13871:
1.209 brouard 13872: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
13873: /* ungetc(c,ficpar); */
13874: /* fgets(line, MAXLINE, ficpar); */
13875: /* fputs(line,stdout); */
13876: /* fputs(line,ficparo); */
13877: /* } */
13878: /* ungetc(c,ficpar); */
1.126 brouard 13879:
13880: estepm=0;
1.209 brouard 13881: if((num_filled=sscanf(line,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm, &ftolpl)) !=EOF){
1.225 brouard 13882:
13883: if (num_filled != 6) {
13884: printf("Error: Not 6 parameters in line, for example:agemin=60 agemax=95 bage=55 fage=95 estepm=24 ftolpl=6e-4\n, your line=%s . Probably you are running an older format.\n",line);
13885: fprintf(ficlog,"Error: Not 6 parameters in line, for example:agemin=60 agemax=95 bage=55 fage=95 estepm=24 ftolpl=6e-4\n, your line=%s . Probably you are running an older format.\n",line);
13886: goto end;
13887: }
13888: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
13889: }
13890: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
13891: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
13892:
1.209 brouard 13893: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 13894: if (estepm==0 || estepm < stepm) estepm=stepm;
13895: if (fage <= 2) {
13896: bage = ageminpar;
13897: fage = agemaxpar;
13898: }
13899:
13900: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 13901: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
13902: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 13903:
1.186 brouard 13904: /* Other stuffs, more or less useful */
1.254 brouard 13905: while(fgets(line, MAXLINE, ficpar)) {
13906: /* If line starts with a # it is a comment */
13907: if (line[0] == '#') {
13908: numlinepar++;
13909: fputs(line,stdout);
13910: fputs(line,ficparo);
13911: fputs(line,ficlog);
1.299 brouard 13912: fputs(line,ficres);
1.254 brouard 13913: continue;
13914: }else
13915: break;
13916: }
13917:
13918: if((num_filled=sscanf(line,"begin-prev-date=%lf/%lf/%lf end-prev-date=%lf/%lf/%lf mov_average=%d\n",&jprev1, &mprev1,&anprev1,&jprev2, &mprev2,&anprev2,&mobilav)) !=EOF){
13919:
13920: if (num_filled != 7) {
13921: printf("Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
13922: fprintf(ficlog,"Error: Not 7 (data)parameters in line but %d, for example:begin-prev-date=1/1/1990 end-prev-date=1/6/2004 mov_average=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
13923: goto end;
13924: }
13925: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13926: fprintf(ficparo,"begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13927: fprintf(ficres,"begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13928: fprintf(ficlog,"begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
1.126 brouard 13929: }
1.254 brouard 13930:
13931: while(fgets(line, MAXLINE, ficpar)) {
13932: /* If line starts with a # it is a comment */
13933: if (line[0] == '#') {
13934: numlinepar++;
13935: fputs(line,stdout);
13936: fputs(line,ficparo);
13937: fputs(line,ficlog);
1.299 brouard 13938: fputs(line,ficres);
1.254 brouard 13939: continue;
13940: }else
13941: break;
1.126 brouard 13942: }
13943:
13944:
13945: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13946: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13947:
1.254 brouard 13948: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13949: if (num_filled != 1) {
13950: printf("Error: Not 1 (data)parameters in line but %d, for example:pop_based=0\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
13951: fprintf(ficlog,"Error: Not 1 (data)parameters in line but %d, for example: pop_based=1\n, your line=%s . Probably you are running an older format.\n",num_filled,line);
13952: goto end;
13953: }
13954: printf("pop_based=%d\n",popbased);
13955: fprintf(ficlog,"pop_based=%d\n",popbased);
13956: fprintf(ficparo,"pop_based=%d\n",popbased);
13957: fprintf(ficres,"pop_based=%d\n",popbased);
13958: }
13959:
1.258 brouard 13960: /* Results */
1.332 brouard 13961: /* Value of covariate in each resultine will be compututed (if product) and sorted according to model rank */
13962: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
13963: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 13964: endishere=0;
1.258 brouard 13965: nresult=0;
1.308 brouard 13966: parameterline=0;
1.258 brouard 13967: do{
13968: if(!fgets(line, MAXLINE, ficpar)){
13969: endishere=1;
1.308 brouard 13970: parameterline=15;
1.258 brouard 13971: }else if (line[0] == '#') {
13972: /* If line starts with a # it is a comment */
1.254 brouard 13973: numlinepar++;
13974: fputs(line,stdout);
13975: fputs(line,ficparo);
13976: fputs(line,ficlog);
1.299 brouard 13977: fputs(line,ficres);
1.254 brouard 13978: continue;
1.258 brouard 13979: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13980: parameterline=11;
1.296 brouard 13981: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 13982: parameterline=12;
1.307 brouard 13983: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 13984: parameterline=13;
1.307 brouard 13985: }
1.258 brouard 13986: else{
13987: parameterline=14;
1.254 brouard 13988: }
1.308 brouard 13989: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 13990: case 11:
1.296 brouard 13991: if((num_filled=sscanf(line,"prevforecast=%d starting-proj-date=%lf/%lf/%lf final-proj-date=%lf/%lf/%lf mobil_average=%d\n",&prevfcast,&jproj1,&mproj1,&anproj1,&jproj2,&mproj2,&anproj2,&mobilavproj)) !=EOF && (num_filled == 8)){
13992: fprintf(ficparo,"prevforecast=%d starting-proj-date=%.lf/%.lf/%.lf final-proj-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevfcast,jproj1,mproj1,anproj1,jproj2,mproj2,anproj2,mobilavproj);
1.258 brouard 13993: printf("prevforecast=%d starting-proj-date=%.lf/%.lf/%.lf final-proj-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevfcast,jproj1,mproj1,anproj1,jproj2,mproj2,anproj2,mobilavproj);
13994: fprintf(ficlog,"prevforecast=%d starting-proj-date=%.lf/%.lf/%.lf final-proj-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevfcast,jproj1,mproj1,anproj1,jproj2,mproj2,anproj2,mobilavproj);
13995: fprintf(ficres,"prevforecast=%d starting-proj-date=%.lf/%.lf/%.lf final-proj-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevfcast,jproj1,mproj1,anproj1,jproj2,mproj2,anproj2,mobilavproj);
13996: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 13997: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
13998: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 13999: prvforecast = 1;
14000: }
14001: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 14002: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
14003: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
14004: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 14005: prvforecast = 2;
14006: }
14007: else {
14008: printf("Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevforecast=1 yearsfproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
14009: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevforecast=1 starting-proj-date=1/1/1990 final-proj-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevforecast=1 yearproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
14010: goto end;
1.258 brouard 14011: }
1.254 brouard 14012: break;
1.258 brouard 14013: case 12:
1.296 brouard 14014: if((num_filled=sscanf(line,"prevbackcast=%d starting-back-date=%lf/%lf/%lf final-back-date=%lf/%lf/%lf mobil_average=%d\n",&prevbcast,&jback1,&mback1,&anback1,&jback2,&mback2,&anback2,&mobilavproj)) !=EOF && (num_filled == 8)){
14015: fprintf(ficparo,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
14016: printf("prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
14017: fprintf(ficlog,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
14018: fprintf(ficres,"prevbackcast=%d starting-back-date=%.lf/%.lf/%.lf final-back-date=%.lf/%.lf/%.lf mobil_average=%d\n",prevbcast,jback1,mback1,anback1,jback2,mback2,anback2,mobilavproj);
14019: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 14020: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
14021: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 14022: prvbackcast = 1;
14023: }
14024: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 14025: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
14026: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
14027: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 14028: prvbackcast = 2;
14029: }
14030: else {
14031: printf("Error: Not 8 (data)parameters in line but %d, for example:prevbackcast=1 starting-back-date=1/1/1990 final-back-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevbackcast=1 yearsbproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
14032: fprintf(ficlog,"Error: Not 8 (data)parameters in line but %d, for example:prevbackcast=1 starting-back-date=1/1/1990 final-back-date=1/1/2000 mobil_average=0\nnor 3 (data)parameters, for example:prevbackcast=1 yearbproj=10 mobil_average=0. Your line=%s . You are running probably an older format.\n, ",num_filled,line);
14033: goto end;
1.258 brouard 14034: }
1.230 brouard 14035: break;
1.258 brouard 14036: case 13:
1.332 brouard 14037: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 14038: nresult++; /* Sum of resultlines */
1.342 brouard 14039: /* printf("Result %d: result:%s\n",nresult, resultlineori); */
1.332 brouard 14040: /* removefirstspace(&resultlineori); */
14041:
14042: if(strstr(resultlineori,"v") !=0){
14043: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
14044: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
14045: return 1;
14046: }
14047: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
1.342 brouard 14048: /* printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori); */
1.318 brouard 14049: if(nresult > MAXRESULTLINESPONE-1){
14050: printf("ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\nYou can use the 'r' parameter file '%s' which uses option mle=0 to get other results. ",MAXRESULTLINESPONE-1,nresult,rfileres);
14051: fprintf(ficlog,"ERROR: Current version of IMaCh limits the number of resultlines to %d, you used %d\nYou can use the 'r' parameter file '%s' which uses option mle=0 to get other results. ",MAXRESULTLINESPONE-1,nresult,rfileres);
1.307 brouard 14052: goto end;
14053: }
1.332 brouard 14054:
1.310 brouard 14055: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 14056: fprintf(ficparo,"result: %s\n",resultline);
14057: fprintf(ficres,"result: %s\n",resultline);
14058: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 14059: } else
14060: goto end;
1.307 brouard 14061: break;
14062: case 14:
14063: printf("Error: Unknown command '%s'\n",line);
14064: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 14065: if(line[0] == ' ' || line[0] == '\n'){
14066: printf("It should not be an empty line '%s'\n",line);
14067: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
14068: }
1.307 brouard 14069: if(ncovmodel >=2 && nresult==0 ){
14070: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
14071: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 14072: }
1.307 brouard 14073: /* goto end; */
14074: break;
1.308 brouard 14075: case 15:
14076: printf("End of resultlines.\n");
14077: fprintf(ficlog,"End of resultlines.\n");
14078: break;
14079: default: /* parameterline =0 */
1.307 brouard 14080: nresult=1;
14081: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 14082: } /* End switch parameterline */
14083: }while(endishere==0); /* End do */
1.126 brouard 14084:
1.230 brouard 14085: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 14086: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 14087:
14088: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 14089: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 14090: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 14091: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
14092: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 14093: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 14094: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
14095: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 14096: }else{
1.270 brouard 14097: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 14098: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
14099: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
14100: if(prvforecast==1){
14101: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
14102: jprojd=jproj1;
14103: mprojd=mproj1;
14104: anprojd=anproj1;
14105: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
14106: jprojf=jproj2;
14107: mprojf=mproj2;
14108: anprojf=anproj2;
14109: } else if(prvforecast == 2){
14110: dateprojd=dateintmean;
14111: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
14112: dateprojf=dateintmean+yrfproj;
14113: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
14114: }
14115: if(prvbackcast==1){
14116: datebackd=(jback1+12*mback1+365*anback1)/365;
14117: jbackd=jback1;
14118: mbackd=mback1;
14119: anbackd=anback1;
14120: datebackf=(jback2+12*mback2+365*anback2)/365;
14121: jbackf=jback2;
14122: mbackf=mback2;
14123: anbackf=anback2;
14124: } else if(prvbackcast == 2){
14125: datebackd=dateintmean;
14126: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
14127: datebackf=dateintmean-yrbproj;
14128: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
14129: }
14130:
14131: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 14132: }
14133: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 14134: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
14135: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 14136:
1.225 brouard 14137: /*------------ free_vector -------------*/
14138: /* chdir(path); */
1.220 brouard 14139:
1.215 brouard 14140: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
14141: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
14142: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
14143: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 14144: free_lvector(num,firstobs,lastobs);
14145: free_vector(agedc,firstobs,lastobs);
1.126 brouard 14146: /*free_matrix(covar,0,NCOVMAX,1,n);*/
14147: /*free_matrix(covar,1,NCOVMAX,1,n);*/
14148: fclose(ficparo);
14149: fclose(ficres);
1.220 brouard 14150:
14151:
1.186 brouard 14152: /* Other results (useful)*/
1.220 brouard 14153:
14154:
1.126 brouard 14155: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 14156: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
14157: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 14158: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 14159: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 14160: fclose(ficrespl);
14161:
14162: /*------------- h Pij x at various ages ------------*/
1.180 brouard 14163: /*#include "hpijx.h"*/
1.332 brouard 14164: /** h Pij x Probability to be in state j at age x+h being in i at x, for each combination k of dummies in the model line or to nres?*/
14165: /* calls hpxij with combination k */
1.180 brouard 14166: hPijx(p, bage, fage);
1.145 brouard 14167: fclose(ficrespij);
1.227 brouard 14168:
1.220 brouard 14169: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 14170: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 14171: k=1;
1.126 brouard 14172: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 14173:
1.269 brouard 14174: /* Prevalence for each covariate combination in probs[age][status][cov] */
14175: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14176: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 14177: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 14178: for(k=1;k<=ncovcombmax;k++)
14179: probs[i][j][k]=0.;
1.269 brouard 14180: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
14181: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 14182: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 14183: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14184: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 14185: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 14186: for(k=1;k<=ncovcombmax;k++)
14187: mobaverages[i][j][k]=0.;
1.219 brouard 14188: mobaverage=mobaverages;
14189: if (mobilav!=0) {
1.235 brouard 14190: printf("Movingaveraging observed prevalence\n");
1.258 brouard 14191: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 14192: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
14193: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
14194: printf(" Error in movingaverage mobilav=%d\n",mobilav);
14195: }
1.269 brouard 14196: } else if (mobilavproj !=0) {
1.235 brouard 14197: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 14198: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 14199: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
14200: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
14201: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
14202: }
1.269 brouard 14203: }else{
14204: printf("Internal error moving average\n");
14205: fflush(stdout);
14206: exit(1);
1.219 brouard 14207: }
14208: }/* end if moving average */
1.227 brouard 14209:
1.126 brouard 14210: /*---------- Forecasting ------------------*/
1.296 brouard 14211: if(prevfcast==1){
14212: /* /\* if(stepm ==1){*\/ */
14213: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14214: /*This done previously after freqsummary.*/
14215: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
14216: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
14217:
14218: /* } else if (prvforecast==2){ */
14219: /* /\* if(stepm ==1){*\/ */
14220: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14221: /* } */
14222: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
14223: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 14224: }
1.269 brouard 14225:
1.296 brouard 14226: /* Prevbcasting */
14227: if(prevbcast==1){
1.219 brouard 14228: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14229: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14230: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14231:
14232: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
14233:
14234: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 14235:
1.219 brouard 14236: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
14237: fclose(ficresplb);
14238:
1.222 brouard 14239: hBijx(p, bage, fage, mobaverage);
14240: fclose(ficrespijb);
1.219 brouard 14241:
1.296 brouard 14242: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
14243: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
14244: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
14245: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
14246: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
14247: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
14248:
14249:
1.269 brouard 14250: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14251:
14252:
1.269 brouard 14253: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 14254: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14255: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14256: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 14257: } /* end Prevbcasting */
1.268 brouard 14258:
1.186 brouard 14259:
14260: /* ------ Other prevalence ratios------------ */
1.126 brouard 14261:
1.215 brouard 14262: free_ivector(wav,1,imx);
14263: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
14264: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
14265: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 14266:
14267:
1.127 brouard 14268: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 14269:
1.201 brouard 14270: strcpy(filerese,"E_");
14271: strcat(filerese,fileresu);
1.126 brouard 14272: if((ficreseij=fopen(filerese,"w"))==NULL) {
14273: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14274: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14275: }
1.208 brouard 14276: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
14277: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 14278:
14279: pstamp(ficreseij);
1.219 brouard 14280:
1.235 brouard 14281: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14282: if (cptcovn < 1){i1=1;}
14283:
14284: for(nres=1; nres <= nresult; nres++) /* For each resultline */
14285: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 14286: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 14287: continue;
1.219 brouard 14288: fprintf(ficreseij,"\n#****** ");
1.235 brouard 14289: printf("\n#****** ");
1.225 brouard 14290: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 14291: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
14292: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 14293: }
14294: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.337 brouard 14295: printf(" V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
14296: fprintf(ficreseij,"V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 14297: }
14298: fprintf(ficreseij,"******\n");
1.235 brouard 14299: printf("******\n");
1.219 brouard 14300:
14301: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14302: oldm=oldms;savm=savms;
1.330 brouard 14303: /* printf("HELLO Entering evsij bage=%d fage=%d k=%d estepm=%d nres=%d\n",(int) bage, (int)fage, k, estepm, nres); */
1.235 brouard 14304: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 14305:
1.219 brouard 14306: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 14307: }
14308: fclose(ficreseij);
1.208 brouard 14309: printf("done evsij\n");fflush(stdout);
14310: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 14311:
1.218 brouard 14312:
1.227 brouard 14313: /*---------- State-specific expectancies and variances ------------*/
1.336 brouard 14314: /* Should be moved in a function */
1.201 brouard 14315: strcpy(filerest,"T_");
14316: strcat(filerest,fileresu);
1.127 brouard 14317: if((ficrest=fopen(filerest,"w"))==NULL) {
14318: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
14319: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
14320: }
1.208 brouard 14321: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
14322: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 14323: strcpy(fileresstde,"STDE_");
14324: strcat(fileresstde,fileresu);
1.126 brouard 14325: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 14326: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
14327: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 14328: }
1.227 brouard 14329: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
14330: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 14331:
1.201 brouard 14332: strcpy(filerescve,"CVE_");
14333: strcat(filerescve,fileresu);
1.126 brouard 14334: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 14335: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
14336: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 14337: }
1.227 brouard 14338: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
14339: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 14340:
1.201 brouard 14341: strcpy(fileresv,"V_");
14342: strcat(fileresv,fileresu);
1.126 brouard 14343: if((ficresvij=fopen(fileresv,"w"))==NULL) {
14344: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
14345: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
14346: }
1.227 brouard 14347: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
14348: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 14349:
1.235 brouard 14350: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14351: if (cptcovn < 1){i1=1;}
14352:
1.334 brouard 14353: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
14354: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
14355: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
14356: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
14357: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
14358: /* */
14359: if(i1 != 1 && TKresult[nres]!= k) /* TKresult[nres] is the combination of this nres resultline. All the i1 combinations are not output */
1.235 brouard 14360: continue;
1.321 brouard 14361: printf("\n# model %s \n#****** Result for:", model);
14362: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
14363: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.334 brouard 14364: /* It might not be a good idea to mix dummies and quantitative */
14365: /* for(j=1;j<=cptcoveff;j++){ /\* j=resultpos. Could be a loop on cptcovs: number of single dummy covariate in the result line as well as in the model *\/ */
14366: for(j=1;j<=cptcovs;j++){ /* j=resultpos. Could be a loop on cptcovs: number of single covariate (dummy or quantitative) in the result line as well as in the model */
14367: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
14368: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
14369: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
14370: * (V5 is quanti) V4 and V3 are dummies
14371: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
14372: * l=1 l=2
14373: * k=1 1 1 0 0
14374: * k=2 2 1 1 0
14375: * k=3 [1] [2] 0 1
14376: * k=4 2 2 1 1
14377: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
14378: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
14379: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
14380: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
14381: */
14382: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
14383: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
14384: /* We give up with the combinations!! */
1.342 brouard 14385: /* if(debugILK) */
14386: /* printf("\n j=%d In computing T_ Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=%d cptcovs=%d, cptcoveff=%d Fixed[modelresult[nres][j]]=%d\n", j, nres, j, Dummy[modelresult[nres][j]],nres,j,modelresult[nres][j],cptcovs, cptcoveff,Fixed[modelresult[nres][j]]); /\* end if dummy or quanti *\/ */
1.334 brouard 14387:
14388: if(Dummy[modelresult[nres][j]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to j in resultline */
1.344 brouard 14389: /* printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][j]); /\* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline *\/ */ /* TinvDoQresult[nres][Name of the variable] */
14390: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); /* Output of each value for the combination TKresult[nres], ordered by the covariate values in the resultline */
14391: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
14392: fprintf(ficrest,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
1.334 brouard 14393: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14394: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14395: }else{
14396: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14397: }
14398: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14399: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14400: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
14401: /* For each selected (single) quantitative value */
1.337 brouard 14402: printf(" V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14403: fprintf(ficlog," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14404: fprintf(ficrest," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
1.334 brouard 14405: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14406: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14407: }else{
14408: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14409: }
14410: }else{
14411: printf("Error in computing T_ Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=%d cptcovs=%d, cptcoveff=%d \n", nres, j, Dummy[modelresult[nres][j]],nres,j,modelresult[nres][j],cptcovs, cptcoveff); /* end if dummy or quanti */
14412: fprintf(ficlog,"Error in computing T_ Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=%d cptcovs=%d, cptcoveff=%d \n", nres, j, Dummy[modelresult[nres][j]],nres,j,modelresult[nres][j],cptcovs, cptcoveff); /* end if dummy or quanti */
14413: exit(1);
14414: }
1.335 brouard 14415: } /* End loop for each variable in the resultline */
1.334 brouard 14416: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
14417: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
14418: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14419: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14420: /* } */
1.208 brouard 14421: fprintf(ficrest,"******\n");
1.227 brouard 14422: fprintf(ficlog,"******\n");
14423: printf("******\n");
1.208 brouard 14424:
14425: fprintf(ficresstdeij,"\n#****** ");
14426: fprintf(ficrescveij,"\n#****** ");
1.337 brouard 14427: /* It could have been: for(j=1;j<=cptcoveff;j++) {printf("V=%d=%lg",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);} */
14428: /* But it won't be sorted and depends on how the resultline is ordered */
1.225 brouard 14429: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 14430: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14431: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14432: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14433: }
14434: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value, TvarsQind gives the position of a quantitative in model equation */
1.337 brouard 14435: fprintf(ficresstdeij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
14436: fprintf(ficrescveij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 14437: }
1.208 brouard 14438: fprintf(ficresstdeij,"******\n");
14439: fprintf(ficrescveij,"******\n");
14440:
14441: fprintf(ficresvij,"\n#****** ");
1.238 brouard 14442: /* pstamp(ficresvij); */
1.225 brouard 14443: for(j=1;j<=cptcoveff;j++)
1.335 brouard 14444: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14445: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 14446: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 14447: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
1.337 brouard 14448: fprintf(ficresvij," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 14449: }
1.208 brouard 14450: fprintf(ficresvij,"******\n");
14451:
14452: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14453: oldm=oldms;savm=savms;
1.235 brouard 14454: printf(" cvevsij ");
14455: fprintf(ficlog, " cvevsij ");
14456: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 14457: printf(" end cvevsij \n ");
14458: fprintf(ficlog, " end cvevsij \n ");
14459:
14460: /*
14461: */
14462: /* goto endfree; */
14463:
14464: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14465: pstamp(ficrest);
14466:
1.269 brouard 14467: epj=vector(1,nlstate+1);
1.208 brouard 14468: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 14469: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
14470: cptcod= 0; /* To be deleted */
14471: printf("varevsij vpopbased=%d \n",vpopbased);
14472: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 14473: varevsij(optionfilefiname, vareij, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, &ncvyear, k, estepm, cptcov,cptcod,vpopbased,mobilav, strstart, nres); /* cptcod not initialized Intel */
1.227 brouard 14474: fprintf(ficrest,"# Total life expectancy with std error and decomposition into time to be expected in each health state\n# (weighted average of eij where weights are ");
14475: if(vpopbased==1)
14476: fprintf(ficrest,"the age specific prevalence observed (cross-sectionally) in the population i.e cross-sectionally\n in each health state (popbased=1) (mobilav=%d)\n",mobilav);
14477: else
1.288 brouard 14478: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.335 brouard 14479: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 14480: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
14481: fprintf(ficrest,"\n");
14482: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 14483: printf("Computing age specific forward period (stable) prevalences in each health state \n");
14484: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 14485: for(age=bage; age <=fage ;age++){
1.235 brouard 14486: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 14487: if (vpopbased==1) {
14488: if(mobilav ==0){
14489: for(i=1; i<=nlstate;i++)
14490: prlim[i][i]=probs[(int)age][i][k];
14491: }else{ /* mobilav */
14492: for(i=1; i<=nlstate;i++)
14493: prlim[i][i]=mobaverage[(int)age][i][k];
14494: }
14495: }
1.219 brouard 14496:
1.227 brouard 14497: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
14498: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
14499: /* printf(" age %4.0f ",age); */
14500: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
14501: for(i=1, epj[j]=0.;i <=nlstate;i++) {
14502: epj[j] += prlim[i][i]*eij[i][j][(int)age];
14503: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
14504: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
14505: }
14506: epj[nlstate+1] +=epj[j];
14507: }
14508: /* printf(" age %4.0f \n",age); */
1.219 brouard 14509:
1.227 brouard 14510: for(i=1, vepp=0.;i <=nlstate;i++)
14511: for(j=1;j <=nlstate;j++)
14512: vepp += vareij[i][j][(int)age];
14513: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
14514: for(j=1;j <=nlstate;j++){
14515: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
14516: }
14517: fprintf(ficrest,"\n");
14518: }
1.208 brouard 14519: } /* End vpopbased */
1.269 brouard 14520: free_vector(epj,1,nlstate+1);
1.208 brouard 14521: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
14522: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 14523: printf("done selection\n");fflush(stdout);
14524: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 14525:
1.335 brouard 14526: } /* End k selection or end covariate selection for nres */
1.227 brouard 14527:
14528: printf("done State-specific expectancies\n");fflush(stdout);
14529: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
14530:
1.335 brouard 14531: /* variance-covariance of forward period prevalence */
1.269 brouard 14532: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14533:
1.227 brouard 14534:
1.290 brouard 14535: free_vector(weight,firstobs,lastobs);
1.330 brouard 14536: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 14537: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 14538: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
14539: free_matrix(anint,1,maxwav,firstobs,lastobs);
14540: free_matrix(mint,1,maxwav,firstobs,lastobs);
14541: free_ivector(cod,firstobs,lastobs);
1.227 brouard 14542: free_ivector(tab,1,NCOVMAX);
14543: fclose(ficresstdeij);
14544: fclose(ficrescveij);
14545: fclose(ficresvij);
14546: fclose(ficrest);
14547: fclose(ficpar);
14548:
14549:
1.126 brouard 14550: /*---------- End : free ----------------*/
1.219 brouard 14551: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 14552: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
14553: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 14554: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
14555: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 14556: } /* mle==-3 arrives here for freeing */
1.227 brouard 14557: /* endfree:*/
14558: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
14559: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
14560: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.341 brouard 14561: /* if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs); */
14562: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs);
1.290 brouard 14563: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
14564: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
14565: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 14566: free_matrix(matcov,1,npar,1,npar);
14567: free_matrix(hess,1,npar,1,npar);
14568: /*free_vector(delti,1,npar);*/
14569: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14570: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 14571: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 14572: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14573:
14574: free_ivector(ncodemax,1,NCOVMAX);
14575: free_ivector(ncodemaxwundef,1,NCOVMAX);
14576: free_ivector(Dummy,-1,NCOVMAX);
14577: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 14578: free_ivector(DummyV,1,NCOVMAX);
14579: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 14580: free_ivector(Typevar,-1,NCOVMAX);
14581: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 14582: free_ivector(TvarsQ,1,NCOVMAX);
14583: free_ivector(TvarsQind,1,NCOVMAX);
14584: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 14585: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 14586: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 14587: free_ivector(TvarFD,1,NCOVMAX);
14588: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 14589: free_ivector(TvarF,1,NCOVMAX);
14590: free_ivector(TvarFind,1,NCOVMAX);
14591: free_ivector(TvarV,1,NCOVMAX);
14592: free_ivector(TvarVind,1,NCOVMAX);
14593: free_ivector(TvarA,1,NCOVMAX);
14594: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 14595: free_ivector(TvarFQ,1,NCOVMAX);
14596: free_ivector(TvarFQind,1,NCOVMAX);
14597: free_ivector(TvarVD,1,NCOVMAX);
14598: free_ivector(TvarVDind,1,NCOVMAX);
14599: free_ivector(TvarVQ,1,NCOVMAX);
14600: free_ivector(TvarVQind,1,NCOVMAX);
1.339 brouard 14601: free_ivector(TvarVV,1,NCOVMAX);
14602: free_ivector(TvarVVind,1,NCOVMAX);
14603:
1.230 brouard 14604: free_ivector(Tvarsel,1,NCOVMAX);
14605: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 14606: free_ivector(Tposprod,1,NCOVMAX);
14607: free_ivector(Tprod,1,NCOVMAX);
14608: free_ivector(Tvaraff,1,NCOVMAX);
1.338 brouard 14609: free_ivector(invalidvarcomb,0,ncovcombmax);
1.227 brouard 14610: free_ivector(Tage,1,NCOVMAX);
14611: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 14612: free_ivector(TmodelInvind,1,NCOVMAX);
14613: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 14614:
14615: free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
14616:
1.227 brouard 14617: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
14618: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 14619: fflush(fichtm);
14620: fflush(ficgp);
14621:
1.227 brouard 14622:
1.126 brouard 14623: if((nberr >0) || (nbwarn>0)){
1.216 brouard 14624: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
14625: fprintf(ficlog,"End of Imach with %d errors and/or warnings %d. Please look at the log file for details.\n",nberr,nbwarn);
1.126 brouard 14626: }else{
14627: printf("End of Imach\n");
14628: fprintf(ficlog,"End of Imach\n");
14629: }
14630: printf("See log file on %s\n",filelog);
14631: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 14632: /*(void) gettimeofday(&end_time,&tzp);*/
14633: rend_time = time(NULL);
14634: end_time = *localtime(&rend_time);
14635: /* tml = *localtime(&end_time.tm_sec); */
14636: strcpy(strtend,asctime(&end_time));
1.126 brouard 14637: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
14638: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 14639: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 14640:
1.157 brouard 14641: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
14642: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
14643: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 14644: /* printf("Total time was %d uSec.\n", total_usecs);*/
14645: /* if(fileappend(fichtm,optionfilehtm)){ */
14646: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14647: fclose(fichtm);
14648: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14649: fclose(fichtmcov);
14650: fclose(ficgp);
14651: fclose(ficlog);
14652: /*------ End -----------*/
1.227 brouard 14653:
1.281 brouard 14654:
14655: /* Executes gnuplot */
1.227 brouard 14656:
14657: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 14658: #ifdef WIN32
1.227 brouard 14659: if (_chdir(pathcd) != 0)
14660: printf("Can't move to directory %s!\n",path);
14661: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 14662: #else
1.227 brouard 14663: if(chdir(pathcd) != 0)
14664: printf("Can't move to directory %s!\n", path);
14665: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 14666: #endif
1.126 brouard 14667: printf("Current directory %s!\n",pathcd);
14668: /*strcat(plotcmd,CHARSEPARATOR);*/
14669: sprintf(plotcmd,"gnuplot");
1.157 brouard 14670: #ifdef _WIN32
1.126 brouard 14671: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
14672: #endif
14673: if(!stat(plotcmd,&info)){
1.158 brouard 14674: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14675: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 14676: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 14677: }else
14678: strcpy(pplotcmd,plotcmd);
1.157 brouard 14679: #ifdef __unix
1.126 brouard 14680: strcpy(plotcmd,GNUPLOTPROGRAM);
14681: if(!stat(plotcmd,&info)){
1.158 brouard 14682: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14683: }else
14684: strcpy(pplotcmd,plotcmd);
14685: #endif
14686: }else
14687: strcpy(pplotcmd,plotcmd);
14688:
14689: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 14690: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 14691: strcpy(pplotcmd,plotcmd);
1.227 brouard 14692:
1.126 brouard 14693: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 14694: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 14695: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 14696: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 14697: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 14698: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 14699: strcpy(plotcmd,pplotcmd);
14700: }
1.126 brouard 14701: }
1.158 brouard 14702: printf(" Successful, please wait...");
1.126 brouard 14703: while (z[0] != 'q') {
14704: /* chdir(path); */
1.154 brouard 14705: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 14706: scanf("%s",z);
14707: /* if (z[0] == 'c') system("./imach"); */
14708: if (z[0] == 'e') {
1.158 brouard 14709: #ifdef __APPLE__
1.152 brouard 14710: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 14711: #elif __linux
14712: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 14713: #else
1.152 brouard 14714: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 14715: #endif
14716: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
14717: system(pplotcmd);
1.126 brouard 14718: }
14719: else if (z[0] == 'g') system(plotcmd);
14720: else if (z[0] == 'q') exit(0);
14721: }
1.227 brouard 14722: end:
1.126 brouard 14723: while (z[0] != 'q') {
1.195 brouard 14724: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 14725: scanf("%s",z);
14726: }
1.283 brouard 14727: printf("End\n");
1.282 brouard 14728: exit(0);
1.126 brouard 14729: }
FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>
![[BACK]](/icons/small/back.gif){kind=icon}
Return to imach.c CVS log ![[TXT]](/icons/small/text.gif){kind=icon}
![[DIR]](/icons/small/dir.gif){kind=icon}
Up to [Local Repository] / imach / src