Annotation of imach/src/imach.c, revision 1.335
1.335 ! brouard 1: /* $Id: imach.c,v 1.334 2022/08/25 09:08:41 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.335 ! brouard 4: Revision 1.334 2022/08/25 09:08:41 brouard
! 5: Summary: In progress for quantitative
! 6:
1.334 brouard 7: Revision 1.333 2022/08/21 09:10:30 brouard
8: * src/imach.c (Module): Version 0.99r33 A lot of changes in
9: reassigning covariates: my first idea was that people will always
10: use the first covariate V1 into the model but in fact they are
11: producing data with many covariates and can use an equation model
12: with some of the covariate; it means that in a model V2+V3 instead
13: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
14: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
15: the equation model is restricted to two variables only (V2, V3)
16: and the combination for V2 should be codtabm(k,1) instead of
17: (codtabm(k,2), and the code should be
18: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
19: made. All of these should be simplified once a day like we did in
20: hpxij() for example by using precov[nres] which is computed in
21: decoderesult for each nres of each resultline. Loop should be done
22: on the equation model globally by distinguishing only product with
23: age (which are changing with age) and no more on type of
24: covariates, single dummies, single covariates.
25:
1.333 brouard 26: Revision 1.332 2022/08/21 09:06:25 brouard
27: Summary: Version 0.99r33
28:
29: * src/imach.c (Module): Version 0.99r33 A lot of changes in
30: reassigning covariates: my first idea was that people will always
31: use the first covariate V1 into the model but in fact they are
32: producing data with many covariates and can use an equation model
33: with some of the covariate; it means that in a model V2+V3 instead
34: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
35: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
36: the equation model is restricted to two variables only (V2, V3)
37: and the combination for V2 should be codtabm(k,1) instead of
38: (codtabm(k,2), and the code should be
39: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
40: made. All of these should be simplified once a day like we did in
41: hpxij() for example by using precov[nres] which is computed in
42: decoderesult for each nres of each resultline. Loop should be done
43: on the equation model globally by distinguishing only product with
44: age (which are changing with age) and no more on type of
45: covariates, single dummies, single covariates.
46:
1.332 brouard 47: Revision 1.331 2022/08/07 05:40:09 brouard
48: *** empty log message ***
49:
1.331 brouard 50: Revision 1.330 2022/08/06 07:18:25 brouard
51: Summary: last 0.99r31
52:
53: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
54:
1.330 brouard 55: Revision 1.329 2022/08/03 17:29:54 brouard
56: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
57:
1.329 brouard 58: Revision 1.328 2022/07/27 17:40:48 brouard
59: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
60:
1.328 brouard 61: Revision 1.327 2022/07/27 14:47:35 brouard
62: Summary: Still a problem for one-step probabilities in case of quantitative variables
63:
1.327 brouard 64: Revision 1.326 2022/07/26 17:33:55 brouard
65: Summary: some test with nres=1
66:
1.326 brouard 67: Revision 1.325 2022/07/25 14:27:23 brouard
68: Summary: r30
69:
70: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
71: coredumped, revealed by Feiuno, thank you.
72:
1.325 brouard 73: Revision 1.324 2022/07/23 17:44:26 brouard
74: *** empty log message ***
75:
1.324 brouard 76: Revision 1.323 2022/07/22 12:30:08 brouard
77: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
78:
1.323 brouard 79: Revision 1.322 2022/07/22 12:27:48 brouard
80: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
81:
1.322 brouard 82: Revision 1.321 2022/07/22 12:04:24 brouard
83: Summary: r28
84:
85: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
86:
1.321 brouard 87: Revision 1.320 2022/06/02 05:10:11 brouard
88: *** empty log message ***
89:
1.320 brouard 90: Revision 1.319 2022/06/02 04:45:11 brouard
91: * imach.c (Module): Adding the Wald tests from the log to the main
92: htm for better display of the maximum likelihood estimators.
93:
1.319 brouard 94: Revision 1.318 2022/05/24 08:10:59 brouard
95: * imach.c (Module): Some attempts to find a bug of wrong estimates
96: of confidencce intervals with product in the equation modelC
97:
1.318 brouard 98: Revision 1.317 2022/05/15 15:06:23 brouard
99: * imach.c (Module): Some minor improvements
100:
1.317 brouard 101: Revision 1.316 2022/05/11 15:11:31 brouard
102: Summary: r27
103:
1.316 brouard 104: Revision 1.315 2022/05/11 15:06:32 brouard
105: *** empty log message ***
106:
1.315 brouard 107: Revision 1.314 2022/04/13 17:43:09 brouard
108: * imach.c (Module): Adding link to text data files
109:
1.314 brouard 110: Revision 1.313 2022/04/11 15:57:42 brouard
111: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
112:
1.313 brouard 113: Revision 1.312 2022/04/05 21:24:39 brouard
114: *** empty log message ***
115:
1.312 brouard 116: Revision 1.311 2022/04/05 21:03:51 brouard
117: Summary: Fixed quantitative covariates
118:
119: Fixed covariates (dummy or quantitative)
120: with missing values have never been allowed but are ERRORS and
121: program quits. Standard deviations of fixed covariates were
122: wrongly computed. Mean and standard deviations of time varying
123: covariates are still not computed.
124:
1.311 brouard 125: Revision 1.310 2022/03/17 08:45:53 brouard
126: Summary: 99r25
127:
128: Improving detection of errors: result lines should be compatible with
129: the model.
130:
1.310 brouard 131: Revision 1.309 2021/05/20 12:39:14 brouard
132: Summary: Version 0.99r24
133:
1.309 brouard 134: Revision 1.308 2021/03/31 13:11:57 brouard
135: Summary: Version 0.99r23
136:
137:
138: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
139:
1.308 brouard 140: Revision 1.307 2021/03/08 18:11:32 brouard
141: Summary: 0.99r22 fixed bug on result:
142:
1.307 brouard 143: Revision 1.306 2021/02/20 15:44:02 brouard
144: Summary: Version 0.99r21
145:
146: * imach.c (Module): Fix bug on quitting after result lines!
147: (Module): Version 0.99r21
148:
1.306 brouard 149: Revision 1.305 2021/02/20 15:28:30 brouard
150: * imach.c (Module): Fix bug on quitting after result lines!
151:
1.305 brouard 152: Revision 1.304 2021/02/12 11:34:20 brouard
153: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
154:
1.304 brouard 155: Revision 1.303 2021/02/11 19:50:15 brouard
156: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
157:
1.303 brouard 158: Revision 1.302 2020/02/22 21:00:05 brouard
159: * (Module): imach.c Update mle=-3 (for computing Life expectancy
160: and life table from the data without any state)
161:
1.302 brouard 162: Revision 1.301 2019/06/04 13:51:20 brouard
163: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
164:
1.301 brouard 165: Revision 1.300 2019/05/22 19:09:45 brouard
166: Summary: version 0.99r19 of May 2019
167:
1.300 brouard 168: Revision 1.299 2019/05/22 18:37:08 brouard
169: Summary: Cleaned 0.99r19
170:
1.299 brouard 171: Revision 1.298 2019/05/22 18:19:56 brouard
172: *** empty log message ***
173:
1.298 brouard 174: Revision 1.297 2019/05/22 17:56:10 brouard
175: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
176:
1.297 brouard 177: Revision 1.296 2019/05/20 13:03:18 brouard
178: Summary: Projection syntax simplified
179:
180:
181: We can now start projections, forward or backward, from the mean date
182: of inteviews up to or down to a number of years of projection:
183: prevforecast=1 yearsfproj=15.3 mobil_average=0
184: or
185: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
186: or
187: prevbackcast=1 yearsbproj=12.3 mobil_average=1
188: or
189: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
190:
1.296 brouard 191: Revision 1.295 2019/05/18 09:52:50 brouard
192: Summary: doxygen tex bug
193:
1.295 brouard 194: Revision 1.294 2019/05/16 14:54:33 brouard
195: Summary: There was some wrong lines added
196:
1.294 brouard 197: Revision 1.293 2019/05/09 15:17:34 brouard
198: *** empty log message ***
199:
1.293 brouard 200: Revision 1.292 2019/05/09 14:17:20 brouard
201: Summary: Some updates
202:
1.292 brouard 203: Revision 1.291 2019/05/09 13:44:18 brouard
204: Summary: Before ncovmax
205:
1.291 brouard 206: Revision 1.290 2019/05/09 13:39:37 brouard
207: Summary: 0.99r18 unlimited number of individuals
208:
209: 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.
210:
1.290 brouard 211: Revision 1.289 2018/12/13 09:16:26 brouard
212: Summary: Bug for young ages (<-30) will be in r17
213:
1.289 brouard 214: Revision 1.288 2018/05/02 20:58:27 brouard
215: Summary: Some bugs fixed
216:
1.288 brouard 217: Revision 1.287 2018/05/01 17:57:25 brouard
218: Summary: Bug fixed by providing frequencies only for non missing covariates
219:
1.287 brouard 220: Revision 1.286 2018/04/27 14:27:04 brouard
221: Summary: some minor bugs
222:
1.286 brouard 223: Revision 1.285 2018/04/21 21:02:16 brouard
224: Summary: Some bugs fixed, valgrind tested
225:
1.285 brouard 226: Revision 1.284 2018/04/20 05:22:13 brouard
227: Summary: Computing mean and stdeviation of fixed quantitative variables
228:
1.284 brouard 229: Revision 1.283 2018/04/19 14:49:16 brouard
230: Summary: Some minor bugs fixed
231:
1.283 brouard 232: Revision 1.282 2018/02/27 22:50:02 brouard
233: *** empty log message ***
234:
1.282 brouard 235: Revision 1.281 2018/02/27 19:25:23 brouard
236: Summary: Adding second argument for quitting
237:
1.281 brouard 238: Revision 1.280 2018/02/21 07:58:13 brouard
239: Summary: 0.99r15
240:
241: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
242:
1.280 brouard 243: Revision 1.279 2017/07/20 13:35:01 brouard
244: Summary: temporary working
245:
1.279 brouard 246: Revision 1.278 2017/07/19 14:09:02 brouard
247: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
248:
1.278 brouard 249: Revision 1.277 2017/07/17 08:53:49 brouard
250: Summary: BOM files can be read now
251:
1.277 brouard 252: Revision 1.276 2017/06/30 15:48:31 brouard
253: Summary: Graphs improvements
254:
1.276 brouard 255: Revision 1.275 2017/06/30 13:39:33 brouard
256: Summary: Saito's color
257:
1.275 brouard 258: Revision 1.274 2017/06/29 09:47:08 brouard
259: Summary: Version 0.99r14
260:
1.274 brouard 261: Revision 1.273 2017/06/27 11:06:02 brouard
262: Summary: More documentation on projections
263:
1.273 brouard 264: Revision 1.272 2017/06/27 10:22:40 brouard
265: Summary: Color of backprojection changed from 6 to 5(yellow)
266:
1.272 brouard 267: Revision 1.271 2017/06/27 10:17:50 brouard
268: Summary: Some bug with rint
269:
1.271 brouard 270: Revision 1.270 2017/05/24 05:45:29 brouard
271: *** empty log message ***
272:
1.270 brouard 273: Revision 1.269 2017/05/23 08:39:25 brouard
274: Summary: Code into subroutine, cleanings
275:
1.269 brouard 276: Revision 1.268 2017/05/18 20:09:32 brouard
277: Summary: backprojection and confidence intervals of backprevalence
278:
1.268 brouard 279: Revision 1.267 2017/05/13 10:25:05 brouard
280: Summary: temporary save for backprojection
281:
1.267 brouard 282: Revision 1.266 2017/05/13 07:26:12 brouard
283: Summary: Version 0.99r13 (improvements and bugs fixed)
284:
1.266 brouard 285: Revision 1.265 2017/04/26 16:22:11 brouard
286: Summary: imach 0.99r13 Some bugs fixed
287:
1.265 brouard 288: Revision 1.264 2017/04/26 06:01:29 brouard
289: Summary: Labels in graphs
290:
1.264 brouard 291: Revision 1.263 2017/04/24 15:23:15 brouard
292: Summary: to save
293:
1.263 brouard 294: Revision 1.262 2017/04/18 16:48:12 brouard
295: *** empty log message ***
296:
1.262 brouard 297: Revision 1.261 2017/04/05 10:14:09 brouard
298: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
299:
1.261 brouard 300: Revision 1.260 2017/04/04 17:46:59 brouard
301: Summary: Gnuplot indexations fixed (humm)
302:
1.260 brouard 303: Revision 1.259 2017/04/04 13:01:16 brouard
304: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
305:
1.259 brouard 306: Revision 1.258 2017/04/03 10:17:47 brouard
307: Summary: Version 0.99r12
308:
309: Some cleanings, conformed with updated documentation.
310:
1.258 brouard 311: Revision 1.257 2017/03/29 16:53:30 brouard
312: Summary: Temp
313:
1.257 brouard 314: Revision 1.256 2017/03/27 05:50:23 brouard
315: Summary: Temporary
316:
1.256 brouard 317: Revision 1.255 2017/03/08 16:02:28 brouard
318: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
319:
1.255 brouard 320: Revision 1.254 2017/03/08 07:13:00 brouard
321: Summary: Fixing data parameter line
322:
1.254 brouard 323: Revision 1.253 2016/12/15 11:59:41 brouard
324: Summary: 0.99 in progress
325:
1.253 brouard 326: Revision 1.252 2016/09/15 21:15:37 brouard
327: *** empty log message ***
328:
1.252 brouard 329: Revision 1.251 2016/09/15 15:01:13 brouard
330: Summary: not working
331:
1.251 brouard 332: Revision 1.250 2016/09/08 16:07:27 brouard
333: Summary: continue
334:
1.250 brouard 335: Revision 1.249 2016/09/07 17:14:18 brouard
336: Summary: Starting values from frequencies
337:
1.249 brouard 338: Revision 1.248 2016/09/07 14:10:18 brouard
339: *** empty log message ***
340:
1.248 brouard 341: Revision 1.247 2016/09/02 11:11:21 brouard
342: *** empty log message ***
343:
1.247 brouard 344: Revision 1.246 2016/09/02 08:49:22 brouard
345: *** empty log message ***
346:
1.246 brouard 347: Revision 1.245 2016/09/02 07:25:01 brouard
348: *** empty log message ***
349:
1.245 brouard 350: Revision 1.244 2016/09/02 07:17:34 brouard
351: *** empty log message ***
352:
1.244 brouard 353: Revision 1.243 2016/09/02 06:45:35 brouard
354: *** empty log message ***
355:
1.243 brouard 356: Revision 1.242 2016/08/30 15:01:20 brouard
357: Summary: Fixing a lots
358:
1.242 brouard 359: Revision 1.241 2016/08/29 17:17:25 brouard
360: Summary: gnuplot problem in Back projection to fix
361:
1.241 brouard 362: Revision 1.240 2016/08/29 07:53:18 brouard
363: Summary: Better
364:
1.240 brouard 365: Revision 1.239 2016/08/26 15:51:03 brouard
366: Summary: Improvement in Powell output in order to copy and paste
367:
368: Author:
369:
1.239 brouard 370: Revision 1.238 2016/08/26 14:23:35 brouard
371: Summary: Starting tests of 0.99
372:
1.238 brouard 373: Revision 1.237 2016/08/26 09:20:19 brouard
374: Summary: to valgrind
375:
1.237 brouard 376: Revision 1.236 2016/08/25 10:50:18 brouard
377: *** empty log message ***
378:
1.236 brouard 379: Revision 1.235 2016/08/25 06:59:23 brouard
380: *** empty log message ***
381:
1.235 brouard 382: Revision 1.234 2016/08/23 16:51:20 brouard
383: *** empty log message ***
384:
1.234 brouard 385: Revision 1.233 2016/08/23 07:40:50 brouard
386: Summary: not working
387:
1.233 brouard 388: Revision 1.232 2016/08/22 14:20:21 brouard
389: Summary: not working
390:
1.232 brouard 391: Revision 1.231 2016/08/22 07:17:15 brouard
392: Summary: not working
393:
1.231 brouard 394: Revision 1.230 2016/08/22 06:55:53 brouard
395: Summary: Not working
396:
1.230 brouard 397: Revision 1.229 2016/07/23 09:45:53 brouard
398: Summary: Completing for func too
399:
1.229 brouard 400: Revision 1.228 2016/07/22 17:45:30 brouard
401: Summary: Fixing some arrays, still debugging
402:
1.227 brouard 403: Revision 1.226 2016/07/12 18:42:34 brouard
404: Summary: temp
405:
1.226 brouard 406: Revision 1.225 2016/07/12 08:40:03 brouard
407: Summary: saving but not running
408:
1.225 brouard 409: Revision 1.224 2016/07/01 13:16:01 brouard
410: Summary: Fixes
411:
1.224 brouard 412: Revision 1.223 2016/02/19 09:23:35 brouard
413: Summary: temporary
414:
1.223 brouard 415: Revision 1.222 2016/02/17 08:14:50 brouard
416: Summary: Probably last 0.98 stable version 0.98r6
417:
1.222 brouard 418: Revision 1.221 2016/02/15 23:35:36 brouard
419: Summary: minor bug
420:
1.220 brouard 421: Revision 1.219 2016/02/15 00:48:12 brouard
422: *** empty log message ***
423:
1.219 brouard 424: Revision 1.218 2016/02/12 11:29:23 brouard
425: Summary: 0.99 Back projections
426:
1.218 brouard 427: Revision 1.217 2015/12/23 17:18:31 brouard
428: Summary: Experimental backcast
429:
1.217 brouard 430: Revision 1.216 2015/12/18 17:32:11 brouard
431: Summary: 0.98r4 Warning and status=-2
432:
433: Version 0.98r4 is now:
434: - displaying an error when status is -1, date of interview unknown and date of death known;
435: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
436: Older changes concerning s=-2, dating from 2005 have been supersed.
437:
1.216 brouard 438: Revision 1.215 2015/12/16 08:52:24 brouard
439: Summary: 0.98r4 working
440:
1.215 brouard 441: Revision 1.214 2015/12/16 06:57:54 brouard
442: Summary: temporary not working
443:
1.214 brouard 444: Revision 1.213 2015/12/11 18:22:17 brouard
445: Summary: 0.98r4
446:
1.213 brouard 447: Revision 1.212 2015/11/21 12:47:24 brouard
448: Summary: minor typo
449:
1.212 brouard 450: Revision 1.211 2015/11/21 12:41:11 brouard
451: Summary: 0.98r3 with some graph of projected cross-sectional
452:
453: Author: Nicolas Brouard
454:
1.211 brouard 455: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 456: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 457: Summary: Adding ftolpl parameter
458: Author: N Brouard
459:
460: We had difficulties to get smoothed confidence intervals. It was due
461: to the period prevalence which wasn't computed accurately. The inner
462: parameter ftolpl is now an outer parameter of the .imach parameter
463: file after estepm. If ftolpl is small 1.e-4 and estepm too,
464: computation are long.
465:
1.209 brouard 466: Revision 1.208 2015/11/17 14:31:57 brouard
467: Summary: temporary
468:
1.208 brouard 469: Revision 1.207 2015/10/27 17:36:57 brouard
470: *** empty log message ***
471:
1.207 brouard 472: Revision 1.206 2015/10/24 07:14:11 brouard
473: *** empty log message ***
474:
1.206 brouard 475: Revision 1.205 2015/10/23 15:50:53 brouard
476: Summary: 0.98r3 some clarification for graphs on likelihood contributions
477:
1.205 brouard 478: Revision 1.204 2015/10/01 16:20:26 brouard
479: Summary: Some new graphs of contribution to likelihood
480:
1.204 brouard 481: Revision 1.203 2015/09/30 17:45:14 brouard
482: Summary: looking at better estimation of the hessian
483:
484: Also a better criteria for convergence to the period prevalence And
485: therefore adding the number of years needed to converge. (The
486: prevalence in any alive state shold sum to one
487:
1.203 brouard 488: Revision 1.202 2015/09/22 19:45:16 brouard
489: Summary: Adding some overall graph on contribution to likelihood. Might change
490:
1.202 brouard 491: Revision 1.201 2015/09/15 17:34:58 brouard
492: Summary: 0.98r0
493:
494: - Some new graphs like suvival functions
495: - Some bugs fixed like model=1+age+V2.
496:
1.201 brouard 497: Revision 1.200 2015/09/09 16:53:55 brouard
498: Summary: Big bug thanks to Flavia
499:
500: Even model=1+age+V2. did not work anymore
501:
1.200 brouard 502: Revision 1.199 2015/09/07 14:09:23 brouard
503: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
504:
1.199 brouard 505: Revision 1.198 2015/09/03 07:14:39 brouard
506: Summary: 0.98q5 Flavia
507:
1.198 brouard 508: Revision 1.197 2015/09/01 18:24:39 brouard
509: *** empty log message ***
510:
1.197 brouard 511: Revision 1.196 2015/08/18 23:17:52 brouard
512: Summary: 0.98q5
513:
1.196 brouard 514: Revision 1.195 2015/08/18 16:28:39 brouard
515: Summary: Adding a hack for testing purpose
516:
517: After reading the title, ftol and model lines, if the comment line has
518: a q, starting with #q, the answer at the end of the run is quit. It
519: permits to run test files in batch with ctest. The former workaround was
520: $ echo q | imach foo.imach
521:
1.195 brouard 522: Revision 1.194 2015/08/18 13:32:00 brouard
523: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
524:
1.194 brouard 525: Revision 1.193 2015/08/04 07:17:42 brouard
526: Summary: 0.98q4
527:
1.193 brouard 528: Revision 1.192 2015/07/16 16:49:02 brouard
529: Summary: Fixing some outputs
530:
1.192 brouard 531: Revision 1.191 2015/07/14 10:00:33 brouard
532: Summary: Some fixes
533:
1.191 brouard 534: Revision 1.190 2015/05/05 08:51:13 brouard
535: Summary: Adding digits in output parameters (7 digits instead of 6)
536:
537: Fix 1+age+.
538:
1.190 brouard 539: Revision 1.189 2015/04/30 14:45:16 brouard
540: Summary: 0.98q2
541:
1.189 brouard 542: Revision 1.188 2015/04/30 08:27:53 brouard
543: *** empty log message ***
544:
1.188 brouard 545: Revision 1.187 2015/04/29 09:11:15 brouard
546: *** empty log message ***
547:
1.187 brouard 548: Revision 1.186 2015/04/23 12:01:52 brouard
549: Summary: V1*age is working now, version 0.98q1
550:
551: Some codes had been disabled in order to simplify and Vn*age was
552: working in the optimization phase, ie, giving correct MLE parameters,
553: but, as usual, outputs were not correct and program core dumped.
554:
1.186 brouard 555: Revision 1.185 2015/03/11 13:26:42 brouard
556: Summary: Inclusion of compile and links command line for Intel Compiler
557:
1.185 brouard 558: Revision 1.184 2015/03/11 11:52:39 brouard
559: Summary: Back from Windows 8. Intel Compiler
560:
1.184 brouard 561: Revision 1.183 2015/03/10 20:34:32 brouard
562: Summary: 0.98q0, trying with directest, mnbrak fixed
563:
564: We use directest instead of original Powell test; probably no
565: incidence on the results, but better justifications;
566: We fixed Numerical Recipes mnbrak routine which was wrong and gave
567: wrong results.
568:
1.183 brouard 569: Revision 1.182 2015/02/12 08:19:57 brouard
570: Summary: Trying to keep directest which seems simpler and more general
571: Author: Nicolas Brouard
572:
1.182 brouard 573: Revision 1.181 2015/02/11 23:22:24 brouard
574: Summary: Comments on Powell added
575:
576: Author:
577:
1.181 brouard 578: Revision 1.180 2015/02/11 17:33:45 brouard
579: Summary: Finishing move from main to function (hpijx and prevalence_limit)
580:
1.180 brouard 581: Revision 1.179 2015/01/04 09:57:06 brouard
582: Summary: back to OS/X
583:
1.179 brouard 584: Revision 1.178 2015/01/04 09:35:48 brouard
585: *** empty log message ***
586:
1.178 brouard 587: Revision 1.177 2015/01/03 18:40:56 brouard
588: Summary: Still testing ilc32 on OSX
589:
1.177 brouard 590: Revision 1.176 2015/01/03 16:45:04 brouard
591: *** empty log message ***
592:
1.176 brouard 593: Revision 1.175 2015/01/03 16:33:42 brouard
594: *** empty log message ***
595:
1.175 brouard 596: Revision 1.174 2015/01/03 16:15:49 brouard
597: Summary: Still in cross-compilation
598:
1.174 brouard 599: Revision 1.173 2015/01/03 12:06:26 brouard
600: Summary: trying to detect cross-compilation
601:
1.173 brouard 602: Revision 1.172 2014/12/27 12:07:47 brouard
603: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
604:
1.172 brouard 605: Revision 1.171 2014/12/23 13:26:59 brouard
606: Summary: Back from Visual C
607:
608: Still problem with utsname.h on Windows
609:
1.171 brouard 610: Revision 1.170 2014/12/23 11:17:12 brouard
611: Summary: Cleaning some \%% back to %%
612:
613: The escape was mandatory for a specific compiler (which one?), but too many warnings.
614:
1.170 brouard 615: Revision 1.169 2014/12/22 23:08:31 brouard
616: Summary: 0.98p
617:
618: Outputs some informations on compiler used, OS etc. Testing on different platforms.
619:
1.169 brouard 620: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 621: Summary: update
1.169 brouard 622:
1.168 brouard 623: Revision 1.167 2014/12/22 13:50:56 brouard
624: Summary: Testing uname and compiler version and if compiled 32 or 64
625:
626: Testing on Linux 64
627:
1.167 brouard 628: Revision 1.166 2014/12/22 11:40:47 brouard
629: *** empty log message ***
630:
1.166 brouard 631: Revision 1.165 2014/12/16 11:20:36 brouard
632: Summary: After compiling on Visual C
633:
634: * imach.c (Module): Merging 1.61 to 1.162
635:
1.165 brouard 636: Revision 1.164 2014/12/16 10:52:11 brouard
637: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
638:
639: * imach.c (Module): Merging 1.61 to 1.162
640:
1.164 brouard 641: Revision 1.163 2014/12/16 10:30:11 brouard
642: * imach.c (Module): Merging 1.61 to 1.162
643:
1.163 brouard 644: Revision 1.162 2014/09/25 11:43:39 brouard
645: Summary: temporary backup 0.99!
646:
1.162 brouard 647: Revision 1.1 2014/09/16 11:06:58 brouard
648: Summary: With some code (wrong) for nlopt
649:
650: Author:
651:
652: Revision 1.161 2014/09/15 20:41:41 brouard
653: Summary: Problem with macro SQR on Intel compiler
654:
1.161 brouard 655: Revision 1.160 2014/09/02 09:24:05 brouard
656: *** empty log message ***
657:
1.160 brouard 658: Revision 1.159 2014/09/01 10:34:10 brouard
659: Summary: WIN32
660: Author: Brouard
661:
1.159 brouard 662: Revision 1.158 2014/08/27 17:11:51 brouard
663: *** empty log message ***
664:
1.158 brouard 665: Revision 1.157 2014/08/27 16:26:55 brouard
666: Summary: Preparing windows Visual studio version
667: Author: Brouard
668:
669: In order to compile on Visual studio, time.h is now correct and time_t
670: and tm struct should be used. difftime should be used but sometimes I
671: just make the differences in raw time format (time(&now).
672: Trying to suppress #ifdef LINUX
673: Add xdg-open for __linux in order to open default browser.
674:
1.157 brouard 675: Revision 1.156 2014/08/25 20:10:10 brouard
676: *** empty log message ***
677:
1.156 brouard 678: Revision 1.155 2014/08/25 18:32:34 brouard
679: Summary: New compile, minor changes
680: Author: Brouard
681:
1.155 brouard 682: Revision 1.154 2014/06/20 17:32:08 brouard
683: Summary: Outputs now all graphs of convergence to period prevalence
684:
1.154 brouard 685: Revision 1.153 2014/06/20 16:45:46 brouard
686: Summary: If 3 live state, convergence to period prevalence on same graph
687: Author: Brouard
688:
1.153 brouard 689: Revision 1.152 2014/06/18 17:54:09 brouard
690: Summary: open browser, use gnuplot on same dir than imach if not found in the path
691:
1.152 brouard 692: Revision 1.151 2014/06/18 16:43:30 brouard
693: *** empty log message ***
694:
1.151 brouard 695: Revision 1.150 2014/06/18 16:42:35 brouard
696: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
697: Author: brouard
698:
1.150 brouard 699: Revision 1.149 2014/06/18 15:51:14 brouard
700: Summary: Some fixes in parameter files errors
701: Author: Nicolas Brouard
702:
1.149 brouard 703: Revision 1.148 2014/06/17 17:38:48 brouard
704: Summary: Nothing new
705: Author: Brouard
706:
707: Just a new packaging for OS/X version 0.98nS
708:
1.148 brouard 709: Revision 1.147 2014/06/16 10:33:11 brouard
710: *** empty log message ***
711:
1.147 brouard 712: Revision 1.146 2014/06/16 10:20:28 brouard
713: Summary: Merge
714: Author: Brouard
715:
716: Merge, before building revised version.
717:
1.146 brouard 718: Revision 1.145 2014/06/10 21:23:15 brouard
719: Summary: Debugging with valgrind
720: Author: Nicolas Brouard
721:
722: Lot of changes in order to output the results with some covariates
723: After the Edimburgh REVES conference 2014, it seems mandatory to
724: improve the code.
725: No more memory valgrind error but a lot has to be done in order to
726: continue the work of splitting the code into subroutines.
727: Also, decodemodel has been improved. Tricode is still not
728: optimal. nbcode should be improved. Documentation has been added in
729: the source code.
730:
1.144 brouard 731: Revision 1.143 2014/01/26 09:45:38 brouard
732: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
733:
734: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
735: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
736:
1.143 brouard 737: Revision 1.142 2014/01/26 03:57:36 brouard
738: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
739:
740: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
741:
1.142 brouard 742: Revision 1.141 2014/01/26 02:42:01 brouard
743: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
744:
1.141 brouard 745: Revision 1.140 2011/09/02 10:37:54 brouard
746: Summary: times.h is ok with mingw32 now.
747:
1.140 brouard 748: Revision 1.139 2010/06/14 07:50:17 brouard
749: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
750: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
751:
1.139 brouard 752: Revision 1.138 2010/04/30 18:19:40 brouard
753: *** empty log message ***
754:
1.138 brouard 755: Revision 1.137 2010/04/29 18:11:38 brouard
756: (Module): Checking covariates for more complex models
757: than V1+V2. A lot of change to be done. Unstable.
758:
1.137 brouard 759: Revision 1.136 2010/04/26 20:30:53 brouard
760: (Module): merging some libgsl code. Fixing computation
761: of likelione (using inter/intrapolation if mle = 0) in order to
762: get same likelihood as if mle=1.
763: Some cleaning of code and comments added.
764:
1.136 brouard 765: Revision 1.135 2009/10/29 15:33:14 brouard
766: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
767:
1.135 brouard 768: Revision 1.134 2009/10/29 13:18:53 brouard
769: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
770:
1.134 brouard 771: Revision 1.133 2009/07/06 10:21:25 brouard
772: just nforces
773:
1.133 brouard 774: Revision 1.132 2009/07/06 08:22:05 brouard
775: Many tings
776:
1.132 brouard 777: Revision 1.131 2009/06/20 16:22:47 brouard
778: Some dimensions resccaled
779:
1.131 brouard 780: Revision 1.130 2009/05/26 06:44:34 brouard
781: (Module): Max Covariate is now set to 20 instead of 8. A
782: lot of cleaning with variables initialized to 0. Trying to make
783: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
784:
1.130 brouard 785: Revision 1.129 2007/08/31 13:49:27 lievre
786: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
787:
1.129 lievre 788: Revision 1.128 2006/06/30 13:02:05 brouard
789: (Module): Clarifications on computing e.j
790:
1.128 brouard 791: Revision 1.127 2006/04/28 18:11:50 brouard
792: (Module): Yes the sum of survivors was wrong since
793: imach-114 because nhstepm was no more computed in the age
794: loop. Now we define nhstepma in the age loop.
795: (Module): In order to speed up (in case of numerous covariates) we
796: compute health expectancies (without variances) in a first step
797: and then all the health expectancies with variances or standard
798: deviation (needs data from the Hessian matrices) which slows the
799: computation.
800: In the future we should be able to stop the program is only health
801: expectancies and graph are needed without standard deviations.
802:
1.127 brouard 803: Revision 1.126 2006/04/28 17:23:28 brouard
804: (Module): Yes the sum of survivors was wrong since
805: imach-114 because nhstepm was no more computed in the age
806: loop. Now we define nhstepma in the age loop.
807: Version 0.98h
808:
1.126 brouard 809: Revision 1.125 2006/04/04 15:20:31 lievre
810: Errors in calculation of health expectancies. Age was not initialized.
811: Forecasting file added.
812:
813: Revision 1.124 2006/03/22 17:13:53 lievre
814: Parameters are printed with %lf instead of %f (more numbers after the comma).
815: The log-likelihood is printed in the log file
816:
817: Revision 1.123 2006/03/20 10:52:43 brouard
818: * imach.c (Module): <title> changed, corresponds to .htm file
819: name. <head> headers where missing.
820:
821: * imach.c (Module): Weights can have a decimal point as for
822: English (a comma might work with a correct LC_NUMERIC environment,
823: otherwise the weight is truncated).
824: Modification of warning when the covariates values are not 0 or
825: 1.
826: Version 0.98g
827:
828: Revision 1.122 2006/03/20 09:45:41 brouard
829: (Module): Weights can have a decimal point as for
830: English (a comma might work with a correct LC_NUMERIC environment,
831: otherwise the weight is truncated).
832: Modification of warning when the covariates values are not 0 or
833: 1.
834: Version 0.98g
835:
836: Revision 1.121 2006/03/16 17:45:01 lievre
837: * imach.c (Module): Comments concerning covariates added
838:
839: * imach.c (Module): refinements in the computation of lli if
840: status=-2 in order to have more reliable computation if stepm is
841: not 1 month. Version 0.98f
842:
843: Revision 1.120 2006/03/16 15:10:38 lievre
844: (Module): refinements in the computation of lli if
845: status=-2 in order to have more reliable computation if stepm is
846: not 1 month. Version 0.98f
847:
848: Revision 1.119 2006/03/15 17:42:26 brouard
849: (Module): Bug if status = -2, the loglikelihood was
850: computed as likelihood omitting the logarithm. Version O.98e
851:
852: Revision 1.118 2006/03/14 18:20:07 brouard
853: (Module): varevsij Comments added explaining the second
854: table of variances if popbased=1 .
855: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
856: (Module): Function pstamp added
857: (Module): Version 0.98d
858:
859: Revision 1.117 2006/03/14 17:16:22 brouard
860: (Module): varevsij Comments added explaining the second
861: table of variances if popbased=1 .
862: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
863: (Module): Function pstamp added
864: (Module): Version 0.98d
865:
866: Revision 1.116 2006/03/06 10:29:27 brouard
867: (Module): Variance-covariance wrong links and
868: varian-covariance of ej. is needed (Saito).
869:
870: Revision 1.115 2006/02/27 12:17:45 brouard
871: (Module): One freematrix added in mlikeli! 0.98c
872:
873: Revision 1.114 2006/02/26 12:57:58 brouard
874: (Module): Some improvements in processing parameter
875: filename with strsep.
876:
877: Revision 1.113 2006/02/24 14:20:24 brouard
878: (Module): Memory leaks checks with valgrind and:
879: datafile was not closed, some imatrix were not freed and on matrix
880: allocation too.
881:
882: Revision 1.112 2006/01/30 09:55:26 brouard
883: (Module): Back to gnuplot.exe instead of wgnuplot.exe
884:
885: Revision 1.111 2006/01/25 20:38:18 brouard
886: (Module): Lots of cleaning and bugs added (Gompertz)
887: (Module): Comments can be added in data file. Missing date values
888: can be a simple dot '.'.
889:
890: Revision 1.110 2006/01/25 00:51:50 brouard
891: (Module): Lots of cleaning and bugs added (Gompertz)
892:
893: Revision 1.109 2006/01/24 19:37:15 brouard
894: (Module): Comments (lines starting with a #) are allowed in data.
895:
896: Revision 1.108 2006/01/19 18:05:42 lievre
897: Gnuplot problem appeared...
898: To be fixed
899:
900: Revision 1.107 2006/01/19 16:20:37 brouard
901: Test existence of gnuplot in imach path
902:
903: Revision 1.106 2006/01/19 13:24:36 brouard
904: Some cleaning and links added in html output
905:
906: Revision 1.105 2006/01/05 20:23:19 lievre
907: *** empty log message ***
908:
909: Revision 1.104 2005/09/30 16:11:43 lievre
910: (Module): sump fixed, loop imx fixed, and simplifications.
911: (Module): If the status is missing at the last wave but we know
912: that the person is alive, then we can code his/her status as -2
913: (instead of missing=-1 in earlier versions) and his/her
914: contributions to the likelihood is 1 - Prob of dying from last
915: health status (= 1-p13= p11+p12 in the easiest case of somebody in
916: the healthy state at last known wave). Version is 0.98
917:
918: Revision 1.103 2005/09/30 15:54:49 lievre
919: (Module): sump fixed, loop imx fixed, and simplifications.
920:
921: Revision 1.102 2004/09/15 17:31:30 brouard
922: Add the possibility to read data file including tab characters.
923:
924: Revision 1.101 2004/09/15 10:38:38 brouard
925: Fix on curr_time
926:
927: Revision 1.100 2004/07/12 18:29:06 brouard
928: Add version for Mac OS X. Just define UNIX in Makefile
929:
930: Revision 1.99 2004/06/05 08:57:40 brouard
931: *** empty log message ***
932:
933: Revision 1.98 2004/05/16 15:05:56 brouard
934: New version 0.97 . First attempt to estimate force of mortality
935: directly from the data i.e. without the need of knowing the health
936: state at each age, but using a Gompertz model: log u =a + b*age .
937: This is the basic analysis of mortality and should be done before any
938: other analysis, in order to test if the mortality estimated from the
939: cross-longitudinal survey is different from the mortality estimated
940: from other sources like vital statistic data.
941:
942: The same imach parameter file can be used but the option for mle should be -3.
943:
1.324 brouard 944: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 945: former routines in order to include the new code within the former code.
946:
947: The output is very simple: only an estimate of the intercept and of
948: the slope with 95% confident intervals.
949:
950: Current limitations:
951: A) Even if you enter covariates, i.e. with the
952: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
953: B) There is no computation of Life Expectancy nor Life Table.
954:
955: Revision 1.97 2004/02/20 13:25:42 lievre
956: Version 0.96d. Population forecasting command line is (temporarily)
957: suppressed.
958:
959: Revision 1.96 2003/07/15 15:38:55 brouard
960: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
961: rewritten within the same printf. Workaround: many printfs.
962:
963: Revision 1.95 2003/07/08 07:54:34 brouard
964: * imach.c (Repository):
965: (Repository): Using imachwizard code to output a more meaningful covariance
966: matrix (cov(a12,c31) instead of numbers.
967:
968: Revision 1.94 2003/06/27 13:00:02 brouard
969: Just cleaning
970:
971: Revision 1.93 2003/06/25 16:33:55 brouard
972: (Module): On windows (cygwin) function asctime_r doesn't
973: exist so I changed back to asctime which exists.
974: (Module): Version 0.96b
975:
976: Revision 1.92 2003/06/25 16:30:45 brouard
977: (Module): On windows (cygwin) function asctime_r doesn't
978: exist so I changed back to asctime which exists.
979:
980: Revision 1.91 2003/06/25 15:30:29 brouard
981: * imach.c (Repository): Duplicated warning errors corrected.
982: (Repository): Elapsed time after each iteration is now output. It
983: helps to forecast when convergence will be reached. Elapsed time
984: is stamped in powell. We created a new html file for the graphs
985: concerning matrix of covariance. It has extension -cov.htm.
986:
987: Revision 1.90 2003/06/24 12:34:15 brouard
988: (Module): Some bugs corrected for windows. Also, when
989: mle=-1 a template is output in file "or"mypar.txt with the design
990: of the covariance matrix to be input.
991:
992: Revision 1.89 2003/06/24 12:30:52 brouard
993: (Module): Some bugs corrected for windows. Also, when
994: mle=-1 a template is output in file "or"mypar.txt with the design
995: of the covariance matrix to be input.
996:
997: Revision 1.88 2003/06/23 17:54:56 brouard
998: * 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.
999:
1000: Revision 1.87 2003/06/18 12:26:01 brouard
1001: Version 0.96
1002:
1003: Revision 1.86 2003/06/17 20:04:08 brouard
1004: (Module): Change position of html and gnuplot routines and added
1005: routine fileappend.
1006:
1007: Revision 1.85 2003/06/17 13:12:43 brouard
1008: * imach.c (Repository): Check when date of death was earlier that
1009: current date of interview. It may happen when the death was just
1010: prior to the death. In this case, dh was negative and likelihood
1011: was wrong (infinity). We still send an "Error" but patch by
1012: assuming that the date of death was just one stepm after the
1013: interview.
1014: (Repository): Because some people have very long ID (first column)
1015: we changed int to long in num[] and we added a new lvector for
1016: memory allocation. But we also truncated to 8 characters (left
1017: truncation)
1018: (Repository): No more line truncation errors.
1019:
1020: Revision 1.84 2003/06/13 21:44:43 brouard
1021: * imach.c (Repository): Replace "freqsummary" at a correct
1022: place. It differs from routine "prevalence" which may be called
1023: many times. Probs is memory consuming and must be used with
1024: parcimony.
1025: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1026:
1027: Revision 1.83 2003/06/10 13:39:11 lievre
1028: *** empty log message ***
1029:
1030: Revision 1.82 2003/06/05 15:57:20 brouard
1031: Add log in imach.c and fullversion number is now printed.
1032:
1033: */
1034: /*
1035: Interpolated Markov Chain
1036:
1037: Short summary of the programme:
1038:
1.227 brouard 1039: This program computes Healthy Life Expectancies or State-specific
1040: (if states aren't health statuses) Expectancies from
1041: cross-longitudinal data. Cross-longitudinal data consist in:
1042:
1043: -1- a first survey ("cross") where individuals from different ages
1044: are interviewed on their health status or degree of disability (in
1045: the case of a health survey which is our main interest)
1046:
1047: -2- at least a second wave of interviews ("longitudinal") which
1048: measure each change (if any) in individual health status. Health
1049: expectancies are computed from the time spent in each health state
1050: according to a model. More health states you consider, more time is
1051: necessary to reach the Maximum Likelihood of the parameters involved
1052: in the model. The simplest model is the multinomial logistic model
1053: where pij is the probability to be observed in state j at the second
1054: wave conditional to be observed in state i at the first
1055: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1056: etc , where 'age' is age and 'sex' is a covariate. If you want to
1057: have a more complex model than "constant and age", you should modify
1058: the program where the markup *Covariates have to be included here
1059: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1060: convergence.
1061:
1062: The advantage of this computer programme, compared to a simple
1063: multinomial logistic model, is clear when the delay between waves is not
1064: identical for each individual. Also, if a individual missed an
1065: intermediate interview, the information is lost, but taken into
1066: account using an interpolation or extrapolation.
1067:
1068: hPijx is the probability to be observed in state i at age x+h
1069: conditional to the observed state i at age x. The delay 'h' can be
1070: split into an exact number (nh*stepm) of unobserved intermediate
1071: states. This elementary transition (by month, quarter,
1072: semester or year) is modelled as a multinomial logistic. The hPx
1073: matrix is simply the matrix product of nh*stepm elementary matrices
1074: and the contribution of each individual to the likelihood is simply
1075: hPijx.
1076:
1077: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1078: of the life expectancies. It also computes the period (stable) prevalence.
1079:
1080: Back prevalence and projections:
1.227 brouard 1081:
1082: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1083: double agemaxpar, double ftolpl, int *ncvyearp, double
1084: dateprev1,double dateprev2, int firstpass, int lastpass, int
1085: mobilavproj)
1086:
1087: Computes the back prevalence limit for any combination of
1088: covariate values k at any age between ageminpar and agemaxpar and
1089: returns it in **bprlim. In the loops,
1090:
1091: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1092: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1093:
1094: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1095: Computes for any combination of covariates k and any age between bage and fage
1096: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1097: oldm=oldms;savm=savms;
1.227 brouard 1098:
1.267 brouard 1099: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1100: Computes the transition matrix starting at age 'age' over
1101: 'nhstepm*hstepm*stepm' months (i.e. until
1102: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1103: nhstepm*hstepm matrices.
1104:
1105: Returns p3mat[i][j][h] after calling
1106: p3mat[i][j][h]=matprod2(newm,
1107: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1108: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1109: oldm);
1.226 brouard 1110:
1111: Important routines
1112:
1113: - func (or funcone), computes logit (pij) distinguishing
1114: o fixed variables (single or product dummies or quantitative);
1115: o varying variables by:
1116: (1) wave (single, product dummies, quantitative),
1117: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1118: % fixed dummy (treated) or quantitative (not done because time-consuming);
1119: % varying dummy (not done) or quantitative (not done);
1120: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1121: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1122: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1123: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1124: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1125:
1.226 brouard 1126:
1127:
1.324 brouard 1128: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1129: Institut national d'études démographiques, Paris.
1.126 brouard 1130: This software have been partly granted by Euro-REVES, a concerted action
1131: from the European Union.
1132: It is copyrighted identically to a GNU software product, ie programme and
1133: software can be distributed freely for non commercial use. Latest version
1134: can be accessed at http://euroreves.ined.fr/imach .
1135:
1136: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1137: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1138:
1139: **********************************************************************/
1140: /*
1141: main
1142: read parameterfile
1143: read datafile
1144: concatwav
1145: freqsummary
1146: if (mle >= 1)
1147: mlikeli
1148: print results files
1149: if mle==1
1150: computes hessian
1151: read end of parameter file: agemin, agemax, bage, fage, estepm
1152: begin-prev-date,...
1153: open gnuplot file
1154: open html file
1.145 brouard 1155: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1156: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1157: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1158: freexexit2 possible for memory heap.
1159:
1160: h Pij x | pij_nom ficrestpij
1161: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1162: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1163: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1164:
1165: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1166: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1167: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1168: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1169: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1170:
1.126 brouard 1171: forecasting if prevfcast==1 prevforecast call prevalence()
1172: health expectancies
1173: Variance-covariance of DFLE
1174: prevalence()
1175: movingaverage()
1176: varevsij()
1177: if popbased==1 varevsij(,popbased)
1178: total life expectancies
1179: Variance of period (stable) prevalence
1180: end
1181: */
1182:
1.187 brouard 1183: /* #define DEBUG */
1184: /* #define DEBUGBRENT */
1.203 brouard 1185: /* #define DEBUGLINMIN */
1186: /* #define DEBUGHESS */
1187: #define DEBUGHESSIJ
1.224 brouard 1188: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1189: #define POWELL /* Instead of NLOPT */
1.224 brouard 1190: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1191: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1192: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1193: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1194:
1195: #include <math.h>
1196: #include <stdio.h>
1197: #include <stdlib.h>
1198: #include <string.h>
1.226 brouard 1199: #include <ctype.h>
1.159 brouard 1200:
1201: #ifdef _WIN32
1202: #include <io.h>
1.172 brouard 1203: #include <windows.h>
1204: #include <tchar.h>
1.159 brouard 1205: #else
1.126 brouard 1206: #include <unistd.h>
1.159 brouard 1207: #endif
1.126 brouard 1208:
1209: #include <limits.h>
1210: #include <sys/types.h>
1.171 brouard 1211:
1212: #if defined(__GNUC__)
1213: #include <sys/utsname.h> /* Doesn't work on Windows */
1214: #endif
1215:
1.126 brouard 1216: #include <sys/stat.h>
1217: #include <errno.h>
1.159 brouard 1218: /* extern int errno; */
1.126 brouard 1219:
1.157 brouard 1220: /* #ifdef LINUX */
1221: /* #include <time.h> */
1222: /* #include "timeval.h" */
1223: /* #else */
1224: /* #include <sys/time.h> */
1225: /* #endif */
1226:
1.126 brouard 1227: #include <time.h>
1228:
1.136 brouard 1229: #ifdef GSL
1230: #include <gsl/gsl_errno.h>
1231: #include <gsl/gsl_multimin.h>
1232: #endif
1233:
1.167 brouard 1234:
1.162 brouard 1235: #ifdef NLOPT
1236: #include <nlopt.h>
1237: typedef struct {
1238: double (* function)(double [] );
1239: } myfunc_data ;
1240: #endif
1241:
1.126 brouard 1242: /* #include <libintl.h> */
1243: /* #define _(String) gettext (String) */
1244:
1.251 brouard 1245: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1246:
1247: #define GNUPLOTPROGRAM "gnuplot"
1248: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1249: #define FILENAMELENGTH 256
1.126 brouard 1250:
1251: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1252: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1253:
1.144 brouard 1254: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1255: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1256:
1257: #define NINTERVMAX 8
1.144 brouard 1258: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1259: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1260: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1261: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1262: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1263: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1264: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1265: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1266: /* #define AGESUP 130 */
1.288 brouard 1267: /* #define AGESUP 150 */
1268: #define AGESUP 200
1.268 brouard 1269: #define AGEINF 0
1.218 brouard 1270: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1271: #define AGEBASE 40
1.194 brouard 1272: #define AGEOVERFLOW 1.e20
1.164 brouard 1273: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1274: #ifdef _WIN32
1275: #define DIRSEPARATOR '\\'
1276: #define CHARSEPARATOR "\\"
1277: #define ODIRSEPARATOR '/'
1278: #else
1.126 brouard 1279: #define DIRSEPARATOR '/'
1280: #define CHARSEPARATOR "/"
1281: #define ODIRSEPARATOR '\\'
1282: #endif
1283:
1.335 ! brouard 1284: /* $Id: imach.c,v 1.334 2022/08/25 09:08:41 brouard Exp $ */
1.126 brouard 1285: /* $State: Exp $ */
1.196 brouard 1286: #include "version.h"
1287: char version[]=__IMACH_VERSION__;
1.332 brouard 1288: char copyright[]="August 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";
1.335 ! brouard 1289: char fullversion[]="$Revision: 1.334 $ $Date: 2022/08/25 09:08:41 $";
1.126 brouard 1290: char strstart[80];
1291: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1292: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1293: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1294: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1295: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 ! brouard 1296: 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 1297: 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 1298: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
! 1299: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1300: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1301: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 ! brouard 1302: 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 1303: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1304: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1305: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1306: int nsd=0; /**< Total number of single dummy variables (output) */
1307: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1308: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1309: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1310: int ntveff=0; /**< ntveff number of effective time varying variables */
1311: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1312: int cptcov=0; /* Working variable */
1.334 brouard 1313: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1314: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1315: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1316: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1317: int nlstate=2; /* Number of live states */
1318: int ndeath=1; /* Number of dead states */
1.130 brouard 1319: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1320: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1321: int popbased=0;
1322:
1323: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1324: int maxwav=0; /* Maxim number of waves */
1325: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1326: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1327: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1328: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1329: int mle=1, weightopt=0;
1.126 brouard 1330: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1331: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1332: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1333: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1334: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1335: int selected(int kvar); /* Is covariate kvar selected for printing results */
1336:
1.130 brouard 1337: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1338: double **matprod2(); /* test */
1.126 brouard 1339: double **oldm, **newm, **savm; /* Working pointers to matrices */
1340: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1341: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1342:
1.136 brouard 1343: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1344: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1345: FILE *ficlog, *ficrespow;
1.130 brouard 1346: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1347: double fretone; /* Only one call to likelihood */
1.130 brouard 1348: long ipmx=0; /* Number of contributions */
1.126 brouard 1349: double sw; /* Sum of weights */
1350: char filerespow[FILENAMELENGTH];
1351: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1352: FILE *ficresilk;
1353: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1354: FILE *ficresprobmorprev;
1355: FILE *fichtm, *fichtmcov; /* Html File */
1356: FILE *ficreseij;
1357: char filerese[FILENAMELENGTH];
1358: FILE *ficresstdeij;
1359: char fileresstde[FILENAMELENGTH];
1360: FILE *ficrescveij;
1361: char filerescve[FILENAMELENGTH];
1362: FILE *ficresvij;
1363: char fileresv[FILENAMELENGTH];
1.269 brouard 1364:
1.126 brouard 1365: char title[MAXLINE];
1.234 brouard 1366: char model[MAXLINE]; /**< The model line */
1.217 brouard 1367: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1368: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1369: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1370: char command[FILENAMELENGTH];
1371: int outcmd=0;
1372:
1.217 brouard 1373: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1374: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1375: char filelog[FILENAMELENGTH]; /* Log file */
1376: char filerest[FILENAMELENGTH];
1377: char fileregp[FILENAMELENGTH];
1378: char popfile[FILENAMELENGTH];
1379:
1380: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1381:
1.157 brouard 1382: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1383: /* struct timezone tzp; */
1384: /* extern int gettimeofday(); */
1385: struct tm tml, *gmtime(), *localtime();
1386:
1387: extern time_t time();
1388:
1389: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1390: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1391: struct tm tm;
1392:
1.126 brouard 1393: char strcurr[80], strfor[80];
1394:
1395: char *endptr;
1396: long lval;
1397: double dval;
1398:
1399: #define NR_END 1
1400: #define FREE_ARG char*
1401: #define FTOL 1.0e-10
1402:
1403: #define NRANSI
1.240 brouard 1404: #define ITMAX 200
1405: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1406:
1407: #define TOL 2.0e-4
1408:
1409: #define CGOLD 0.3819660
1410: #define ZEPS 1.0e-10
1411: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1412:
1413: #define GOLD 1.618034
1414: #define GLIMIT 100.0
1415: #define TINY 1.0e-20
1416:
1417: static double maxarg1,maxarg2;
1418: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1419: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1420:
1421: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1422: #define rint(a) floor(a+0.5)
1.166 brouard 1423: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1424: #define mytinydouble 1.0e-16
1.166 brouard 1425: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1426: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1427: /* static double dsqrarg; */
1428: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1429: static double sqrarg;
1430: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1431: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1432: int agegomp= AGEGOMP;
1433:
1434: int imx;
1435: int stepm=1;
1436: /* Stepm, step in month: minimum step interpolation*/
1437:
1438: int estepm;
1439: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1440:
1441: int m,nb;
1442: long *num;
1.197 brouard 1443: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1444: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1445: covariate for which somebody answered excluding
1446: undefined. Usually 2: 0 and 1. */
1447: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1448: covariate for which somebody answered including
1449: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1450: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1451: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1452: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1453: 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 1454: double *ageexmed,*agecens;
1455: double dateintmean=0;
1.296 brouard 1456: double anprojd, mprojd, jprojd; /* For eventual projections */
1457: double anprojf, mprojf, jprojf;
1.126 brouard 1458:
1.296 brouard 1459: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1460: double anbackf, mbackf, jbackf;
1461: double jintmean,mintmean,aintmean;
1.126 brouard 1462: double *weight;
1463: int **s; /* Status */
1.141 brouard 1464: double *agedc;
1.145 brouard 1465: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1466: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1467: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1468: double **coqvar; /* Fixed quantitative covariate nqv */
1469: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1470: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1471: double idx;
1472: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1473: /* Some documentation */
1474: /* Design original data
1475: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1476: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1477: * ntv=3 nqtv=1
1.330 brouard 1478: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1479: * For time varying covariate, quanti or dummies
1480: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1481: * cotvar[wav][ntv+iv][i]= [3+(1 to nqtv)][i]=(V12) quanti
1482: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1483: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1484: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1485: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1486: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1487: * k= 1 2 3 4 5 6 7 8 9 10 11
1488: */
1489: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1490: /* 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
1491: # States 1=Coresidence, 2 Living alone, 3 Institution
1492: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1493: */
1.319 brouard 1494: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1495: /* k 1 2 3 4 5 6 7 8 9 */
1496: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1497: /* fixed or varying), 1 for age product, 2 for*/
1498: /* product */
1499: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1500: /*(single or product without age), 2 dummy*/
1501: /* with age product, 3 quant with age product*/
1502: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1503: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 brouard 1504: /*TnsdVar[Tvar] 1 2 3 */
1.319 brouard 1505: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1506: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1507: /* nsq 1 2 */ /* Counting single quantit tv */
1508: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1509: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1510: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1511: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1512: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1513: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1514: /* 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 1515: /* 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 1516: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1517: /* Type */
1518: /* V 1 2 3 4 5 */
1519: /* F F V V V */
1520: /* D Q D D Q */
1521: /* */
1522: int *TvarsD;
1.330 brouard 1523: int *TnsdVar;
1.234 brouard 1524: int *TvarsDind;
1525: int *TvarsQ;
1526: int *TvarsQind;
1527:
1.318 brouard 1528: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1529: int nresult=0;
1.258 brouard 1530: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1531: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1532: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1533: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1534: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1535: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1536: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1537: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1538: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1539: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1540: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1541:
1542: /* 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
1543: # States 1=Coresidence, 2 Living alone, 3 Institution
1544: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1545: */
1.234 brouard 1546: /* 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 1547: 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 */
1548: 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 */
1549: 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 */
1550: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1551: 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 */
1552: 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 1553: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1554: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1555: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1556: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1557: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1558: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1559: 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 */
1560: 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 */
1561:
1.230 brouard 1562: int *Tvarsel; /**< Selected covariates for output */
1563: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1564: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1565: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1566: 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 1567: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1568: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1569: int *Tage;
1.227 brouard 1570: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1571: 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 1572: 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*/
1573: 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 1574: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1575: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1576: int **Tvard;
1.330 brouard 1577: int **Tvardk;
1.227 brouard 1578: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1579: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1580: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1581: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1582: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1583: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1584: double *lsurv, *lpop, *tpop;
1585:
1.231 brouard 1586: #define FD 1; /* Fixed dummy covariate */
1587: #define FQ 2; /* Fixed quantitative covariate */
1588: #define FP 3; /* Fixed product covariate */
1589: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1590: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1591: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1592: #define VD 10; /* Varying dummy covariate */
1593: #define VQ 11; /* Varying quantitative covariate */
1594: #define VP 12; /* Varying product covariate */
1595: #define VPDD 13; /* Varying product dummy*dummy covariate */
1596: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1597: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1598: #define APFD 16; /* Age product * fixed dummy covariate */
1599: #define APFQ 17; /* Age product * fixed quantitative covariate */
1600: #define APVD 18; /* Age product * varying dummy covariate */
1601: #define APVQ 19; /* Age product * varying quantitative covariate */
1602:
1603: #define FTYPE 1; /* Fixed covariate */
1604: #define VTYPE 2; /* Varying covariate (loop in wave) */
1605: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1606:
1607: struct kmodel{
1608: int maintype; /* main type */
1609: int subtype; /* subtype */
1610: };
1611: struct kmodel modell[NCOVMAX];
1612:
1.143 brouard 1613: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1614: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1615:
1616: /**************** split *************************/
1617: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1618: {
1619: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1620: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1621: */
1622: char *ss; /* pointer */
1.186 brouard 1623: int l1=0, l2=0; /* length counters */
1.126 brouard 1624:
1625: l1 = strlen(path ); /* length of path */
1626: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1627: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1628: if ( ss == NULL ) { /* no directory, so determine current directory */
1629: strcpy( name, path ); /* we got the fullname name because no directory */
1630: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1631: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1632: /* get current working directory */
1633: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1634: #ifdef WIN32
1635: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1636: #else
1637: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1638: #endif
1.126 brouard 1639: return( GLOCK_ERROR_GETCWD );
1640: }
1641: /* got dirc from getcwd*/
1642: printf(" DIRC = %s \n",dirc);
1.205 brouard 1643: } else { /* strip directory from path */
1.126 brouard 1644: ss++; /* after this, the filename */
1645: l2 = strlen( ss ); /* length of filename */
1646: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1647: strcpy( name, ss ); /* save file name */
1648: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1649: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1650: printf(" DIRC2 = %s \n",dirc);
1651: }
1652: /* We add a separator at the end of dirc if not exists */
1653: l1 = strlen( dirc ); /* length of directory */
1654: if( dirc[l1-1] != DIRSEPARATOR ){
1655: dirc[l1] = DIRSEPARATOR;
1656: dirc[l1+1] = 0;
1657: printf(" DIRC3 = %s \n",dirc);
1658: }
1659: ss = strrchr( name, '.' ); /* find last / */
1660: if (ss >0){
1661: ss++;
1662: strcpy(ext,ss); /* save extension */
1663: l1= strlen( name);
1664: l2= strlen(ss)+1;
1665: strncpy( finame, name, l1-l2);
1666: finame[l1-l2]= 0;
1667: }
1668:
1669: return( 0 ); /* we're done */
1670: }
1671:
1672:
1673: /******************************************/
1674:
1675: void replace_back_to_slash(char *s, char*t)
1676: {
1677: int i;
1678: int lg=0;
1679: i=0;
1680: lg=strlen(t);
1681: for(i=0; i<= lg; i++) {
1682: (s[i] = t[i]);
1683: if (t[i]== '\\') s[i]='/';
1684: }
1685: }
1686:
1.132 brouard 1687: char *trimbb(char *out, char *in)
1.137 brouard 1688: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1689: char *s;
1690: s=out;
1691: while (*in != '\0'){
1.137 brouard 1692: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1693: in++;
1694: }
1695: *out++ = *in++;
1696: }
1697: *out='\0';
1698: return s;
1699: }
1700:
1.187 brouard 1701: /* char *substrchaine(char *out, char *in, char *chain) */
1702: /* { */
1703: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1704: /* char *s, *t; */
1705: /* t=in;s=out; */
1706: /* while ((*in != *chain) && (*in != '\0')){ */
1707: /* *out++ = *in++; */
1708: /* } */
1709:
1710: /* /\* *in matches *chain *\/ */
1711: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1712: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1713: /* } */
1714: /* in--; chain--; */
1715: /* while ( (*in != '\0')){ */
1716: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1717: /* *out++ = *in++; */
1718: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1719: /* } */
1720: /* *out='\0'; */
1721: /* out=s; */
1722: /* return out; */
1723: /* } */
1724: char *substrchaine(char *out, char *in, char *chain)
1725: {
1726: /* Substract chain 'chain' from 'in', return and output 'out' */
1727: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1728:
1729: char *strloc;
1730:
1731: strcpy (out, in);
1732: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1733: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1734: if(strloc != NULL){
1735: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1736: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1737: /* strcpy (strloc, strloc +strlen(chain));*/
1738: }
1739: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1740: return out;
1741: }
1742:
1743:
1.145 brouard 1744: char *cutl(char *blocc, char *alocc, char *in, char occ)
1745: {
1.187 brouard 1746: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1747: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1748: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1749: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1750: */
1.160 brouard 1751: char *s, *t;
1.145 brouard 1752: t=in;s=in;
1753: while ((*in != occ) && (*in != '\0')){
1754: *alocc++ = *in++;
1755: }
1756: if( *in == occ){
1757: *(alocc)='\0';
1758: s=++in;
1759: }
1760:
1761: if (s == t) {/* occ not found */
1762: *(alocc-(in-s))='\0';
1763: in=s;
1764: }
1765: while ( *in != '\0'){
1766: *blocc++ = *in++;
1767: }
1768:
1769: *blocc='\0';
1770: return t;
1771: }
1.137 brouard 1772: char *cutv(char *blocc, char *alocc, char *in, char occ)
1773: {
1.187 brouard 1774: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1775: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1776: gives blocc="abcdef2ghi" and alocc="j".
1777: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1778: */
1779: char *s, *t;
1780: t=in;s=in;
1781: while (*in != '\0'){
1782: while( *in == occ){
1783: *blocc++ = *in++;
1784: s=in;
1785: }
1786: *blocc++ = *in++;
1787: }
1788: if (s == t) /* occ not found */
1789: *(blocc-(in-s))='\0';
1790: else
1791: *(blocc-(in-s)-1)='\0';
1792: in=s;
1793: while ( *in != '\0'){
1794: *alocc++ = *in++;
1795: }
1796:
1797: *alocc='\0';
1798: return s;
1799: }
1800:
1.126 brouard 1801: int nbocc(char *s, char occ)
1802: {
1803: int i,j=0;
1804: int lg=20;
1805: i=0;
1806: lg=strlen(s);
1807: for(i=0; i<= lg; i++) {
1.234 brouard 1808: if (s[i] == occ ) j++;
1.126 brouard 1809: }
1810: return j;
1811: }
1812:
1.137 brouard 1813: /* void cutv(char *u,char *v, char*t, char occ) */
1814: /* { */
1815: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1816: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1817: /* gives u="abcdef2ghi" and v="j" *\/ */
1818: /* int i,lg,j,p=0; */
1819: /* i=0; */
1820: /* lg=strlen(t); */
1821: /* for(j=0; j<=lg-1; j++) { */
1822: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1823: /* } */
1.126 brouard 1824:
1.137 brouard 1825: /* for(j=0; j<p; j++) { */
1826: /* (u[j] = t[j]); */
1827: /* } */
1828: /* u[p]='\0'; */
1.126 brouard 1829:
1.137 brouard 1830: /* for(j=0; j<= lg; j++) { */
1831: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1832: /* } */
1833: /* } */
1.126 brouard 1834:
1.160 brouard 1835: #ifdef _WIN32
1836: char * strsep(char **pp, const char *delim)
1837: {
1838: char *p, *q;
1839:
1840: if ((p = *pp) == NULL)
1841: return 0;
1842: if ((q = strpbrk (p, delim)) != NULL)
1843: {
1844: *pp = q + 1;
1845: *q = '\0';
1846: }
1847: else
1848: *pp = 0;
1849: return p;
1850: }
1851: #endif
1852:
1.126 brouard 1853: /********************** nrerror ********************/
1854:
1855: void nrerror(char error_text[])
1856: {
1857: fprintf(stderr,"ERREUR ...\n");
1858: fprintf(stderr,"%s\n",error_text);
1859: exit(EXIT_FAILURE);
1860: }
1861: /*********************** vector *******************/
1862: double *vector(int nl, int nh)
1863: {
1864: double *v;
1865: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1866: if (!v) nrerror("allocation failure in vector");
1867: return v-nl+NR_END;
1868: }
1869:
1870: /************************ free vector ******************/
1871: void free_vector(double*v, int nl, int nh)
1872: {
1873: free((FREE_ARG)(v+nl-NR_END));
1874: }
1875:
1876: /************************ivector *******************************/
1877: int *ivector(long nl,long nh)
1878: {
1879: int *v;
1880: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1881: if (!v) nrerror("allocation failure in ivector");
1882: return v-nl+NR_END;
1883: }
1884:
1885: /******************free ivector **************************/
1886: void free_ivector(int *v, long nl, long nh)
1887: {
1888: free((FREE_ARG)(v+nl-NR_END));
1889: }
1890:
1891: /************************lvector *******************************/
1892: long *lvector(long nl,long nh)
1893: {
1894: long *v;
1895: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1896: if (!v) nrerror("allocation failure in ivector");
1897: return v-nl+NR_END;
1898: }
1899:
1900: /******************free lvector **************************/
1901: void free_lvector(long *v, long nl, long nh)
1902: {
1903: free((FREE_ARG)(v+nl-NR_END));
1904: }
1905:
1906: /******************* imatrix *******************************/
1907: int **imatrix(long nrl, long nrh, long ncl, long nch)
1908: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1909: {
1910: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1911: int **m;
1912:
1913: /* allocate pointers to rows */
1914: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1915: if (!m) nrerror("allocation failure 1 in matrix()");
1916: m += NR_END;
1917: m -= nrl;
1918:
1919:
1920: /* allocate rows and set pointers to them */
1921: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1922: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1923: m[nrl] += NR_END;
1924: m[nrl] -= ncl;
1925:
1926: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1927:
1928: /* return pointer to array of pointers to rows */
1929: return m;
1930: }
1931:
1932: /****************** free_imatrix *************************/
1933: void free_imatrix(m,nrl,nrh,ncl,nch)
1934: int **m;
1935: long nch,ncl,nrh,nrl;
1936: /* free an int matrix allocated by imatrix() */
1937: {
1938: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1939: free((FREE_ARG) (m+nrl-NR_END));
1940: }
1941:
1942: /******************* matrix *******************************/
1943: double **matrix(long nrl, long nrh, long ncl, long nch)
1944: {
1945: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1946: double **m;
1947:
1948: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1949: if (!m) nrerror("allocation failure 1 in matrix()");
1950: m += NR_END;
1951: m -= nrl;
1952:
1953: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1954: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1955: m[nrl] += NR_END;
1956: m[nrl] -= ncl;
1957:
1958: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1959: return m;
1.145 brouard 1960: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1961: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1962: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1963: */
1964: }
1965:
1966: /*************************free matrix ************************/
1967: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1968: {
1969: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1970: free((FREE_ARG)(m+nrl-NR_END));
1971: }
1972:
1973: /******************* ma3x *******************************/
1974: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1975: {
1976: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1977: double ***m;
1978:
1979: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1980: if (!m) nrerror("allocation failure 1 in matrix()");
1981: m += NR_END;
1982: m -= nrl;
1983:
1984: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1985: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1986: m[nrl] += NR_END;
1987: m[nrl] -= ncl;
1988:
1989: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1990:
1991: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1992: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1993: m[nrl][ncl] += NR_END;
1994: m[nrl][ncl] -= nll;
1995: for (j=ncl+1; j<=nch; j++)
1996: m[nrl][j]=m[nrl][j-1]+nlay;
1997:
1998: for (i=nrl+1; i<=nrh; i++) {
1999: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2000: for (j=ncl+1; j<=nch; j++)
2001: m[i][j]=m[i][j-1]+nlay;
2002: }
2003: return m;
2004: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2005: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2006: */
2007: }
2008:
2009: /*************************free ma3x ************************/
2010: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2011: {
2012: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2013: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2014: free((FREE_ARG)(m+nrl-NR_END));
2015: }
2016:
2017: /*************** function subdirf ***********/
2018: char *subdirf(char fileres[])
2019: {
2020: /* Caution optionfilefiname is hidden */
2021: strcpy(tmpout,optionfilefiname);
2022: strcat(tmpout,"/"); /* Add to the right */
2023: strcat(tmpout,fileres);
2024: return tmpout;
2025: }
2026:
2027: /*************** function subdirf2 ***********/
2028: char *subdirf2(char fileres[], char *preop)
2029: {
1.314 brouard 2030: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2031: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2032: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2033: /* Caution optionfilefiname is hidden */
2034: strcpy(tmpout,optionfilefiname);
2035: strcat(tmpout,"/");
2036: strcat(tmpout,preop);
2037: strcat(tmpout,fileres);
2038: return tmpout;
2039: }
2040:
2041: /*************** function subdirf3 ***********/
2042: char *subdirf3(char fileres[], char *preop, char *preop2)
2043: {
2044:
2045: /* Caution optionfilefiname is hidden */
2046: strcpy(tmpout,optionfilefiname);
2047: strcat(tmpout,"/");
2048: strcat(tmpout,preop);
2049: strcat(tmpout,preop2);
2050: strcat(tmpout,fileres);
2051: return tmpout;
2052: }
1.213 brouard 2053:
2054: /*************** function subdirfext ***********/
2055: char *subdirfext(char fileres[], char *preop, char *postop)
2056: {
2057:
2058: strcpy(tmpout,preop);
2059: strcat(tmpout,fileres);
2060: strcat(tmpout,postop);
2061: return tmpout;
2062: }
1.126 brouard 2063:
1.213 brouard 2064: /*************** function subdirfext3 ***********/
2065: char *subdirfext3(char fileres[], char *preop, char *postop)
2066: {
2067:
2068: /* Caution optionfilefiname is hidden */
2069: strcpy(tmpout,optionfilefiname);
2070: strcat(tmpout,"/");
2071: strcat(tmpout,preop);
2072: strcat(tmpout,fileres);
2073: strcat(tmpout,postop);
2074: return tmpout;
2075: }
2076:
1.162 brouard 2077: char *asc_diff_time(long time_sec, char ascdiff[])
2078: {
2079: long sec_left, days, hours, minutes;
2080: days = (time_sec) / (60*60*24);
2081: sec_left = (time_sec) % (60*60*24);
2082: hours = (sec_left) / (60*60) ;
2083: sec_left = (sec_left) %(60*60);
2084: minutes = (sec_left) /60;
2085: sec_left = (sec_left) % (60);
2086: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2087: return ascdiff;
2088: }
2089:
1.126 brouard 2090: /***************** f1dim *************************/
2091: extern int ncom;
2092: extern double *pcom,*xicom;
2093: extern double (*nrfunc)(double []);
2094:
2095: double f1dim(double x)
2096: {
2097: int j;
2098: double f;
2099: double *xt;
2100:
2101: xt=vector(1,ncom);
2102: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2103: f=(*nrfunc)(xt);
2104: free_vector(xt,1,ncom);
2105: return f;
2106: }
2107:
2108: /*****************brent *************************/
2109: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2110: {
2111: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2112: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2113: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2114: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2115: * returned function value.
2116: */
1.126 brouard 2117: int iter;
2118: double a,b,d,etemp;
1.159 brouard 2119: double fu=0,fv,fw,fx;
1.164 brouard 2120: double ftemp=0.;
1.126 brouard 2121: double p,q,r,tol1,tol2,u,v,w,x,xm;
2122: double e=0.0;
2123:
2124: a=(ax < cx ? ax : cx);
2125: b=(ax > cx ? ax : cx);
2126: x=w=v=bx;
2127: fw=fv=fx=(*f)(x);
2128: for (iter=1;iter<=ITMAX;iter++) {
2129: xm=0.5*(a+b);
2130: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2131: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2132: printf(".");fflush(stdout);
2133: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2134: #ifdef DEBUGBRENT
1.126 brouard 2135: 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);
2136: 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);
2137: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2138: #endif
2139: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2140: *xmin=x;
2141: return fx;
2142: }
2143: ftemp=fu;
2144: if (fabs(e) > tol1) {
2145: r=(x-w)*(fx-fv);
2146: q=(x-v)*(fx-fw);
2147: p=(x-v)*q-(x-w)*r;
2148: q=2.0*(q-r);
2149: if (q > 0.0) p = -p;
2150: q=fabs(q);
2151: etemp=e;
2152: e=d;
2153: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2154: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2155: else {
1.224 brouard 2156: d=p/q;
2157: u=x+d;
2158: if (u-a < tol2 || b-u < tol2)
2159: d=SIGN(tol1,xm-x);
1.126 brouard 2160: }
2161: } else {
2162: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2163: }
2164: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2165: fu=(*f)(u);
2166: if (fu <= fx) {
2167: if (u >= x) a=x; else b=x;
2168: SHFT(v,w,x,u)
1.183 brouard 2169: SHFT(fv,fw,fx,fu)
2170: } else {
2171: if (u < x) a=u; else b=u;
2172: if (fu <= fw || w == x) {
1.224 brouard 2173: v=w;
2174: w=u;
2175: fv=fw;
2176: fw=fu;
1.183 brouard 2177: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2178: v=u;
2179: fv=fu;
1.183 brouard 2180: }
2181: }
1.126 brouard 2182: }
2183: nrerror("Too many iterations in brent");
2184: *xmin=x;
2185: return fx;
2186: }
2187:
2188: /****************** mnbrak ***********************/
2189:
2190: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2191: double (*func)(double))
1.183 brouard 2192: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2193: the downhill direction (defined by the function as evaluated at the initial points) and returns
2194: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2195: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2196: */
1.126 brouard 2197: double ulim,u,r,q, dum;
2198: double fu;
1.187 brouard 2199:
2200: double scale=10.;
2201: int iterscale=0;
2202:
2203: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2204: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2205:
2206:
2207: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2208: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2209: /* *bx = *ax - (*ax - *bx)/scale; */
2210: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2211: /* } */
2212:
1.126 brouard 2213: if (*fb > *fa) {
2214: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2215: SHFT(dum,*fb,*fa,dum)
2216: }
1.126 brouard 2217: *cx=(*bx)+GOLD*(*bx-*ax);
2218: *fc=(*func)(*cx);
1.183 brouard 2219: #ifdef DEBUG
1.224 brouard 2220: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2221: 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 2222: #endif
1.224 brouard 2223: 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 2224: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2225: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2226: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2227: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2228: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2229: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2230: fu=(*func)(u);
1.163 brouard 2231: #ifdef DEBUG
2232: /* f(x)=A(x-u)**2+f(u) */
2233: double A, fparabu;
2234: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2235: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2236: 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);
2237: 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 2238: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2239: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2240: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2241: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2242: #endif
1.184 brouard 2243: #ifdef MNBRAKORIGINAL
1.183 brouard 2244: #else
1.191 brouard 2245: /* if (fu > *fc) { */
2246: /* #ifdef DEBUG */
2247: /* printf("mnbrak4 fu > fc \n"); */
2248: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2249: /* #endif */
2250: /* /\* 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 *\\/ *\/ */
2251: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2252: /* dum=u; /\* Shifting c and u *\/ */
2253: /* u = *cx; */
2254: /* *cx = dum; */
2255: /* dum = fu; */
2256: /* fu = *fc; */
2257: /* *fc =dum; */
2258: /* } else { /\* end *\/ */
2259: /* #ifdef DEBUG */
2260: /* printf("mnbrak3 fu < fc \n"); */
2261: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2262: /* #endif */
2263: /* dum=u; /\* Shifting c and u *\/ */
2264: /* u = *cx; */
2265: /* *cx = dum; */
2266: /* dum = fu; */
2267: /* fu = *fc; */
2268: /* *fc =dum; */
2269: /* } */
1.224 brouard 2270: #ifdef DEBUGMNBRAK
2271: double A, fparabu;
2272: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2273: fparabu= *fa - A*(*ax-u)*(*ax-u);
2274: 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);
2275: 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 2276: #endif
1.191 brouard 2277: dum=u; /* Shifting c and u */
2278: u = *cx;
2279: *cx = dum;
2280: dum = fu;
2281: fu = *fc;
2282: *fc =dum;
1.183 brouard 2283: #endif
1.162 brouard 2284: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2285: #ifdef DEBUG
1.224 brouard 2286: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2287: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2288: #endif
1.126 brouard 2289: fu=(*func)(u);
2290: if (fu < *fc) {
1.183 brouard 2291: #ifdef DEBUG
1.224 brouard 2292: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2293: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2294: #endif
2295: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2296: SHFT(*fb,*fc,fu,(*func)(u))
2297: #ifdef DEBUG
2298: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2299: #endif
2300: }
1.162 brouard 2301: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2302: #ifdef DEBUG
1.224 brouard 2303: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2304: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2305: #endif
1.126 brouard 2306: u=ulim;
2307: fu=(*func)(u);
1.183 brouard 2308: } else { /* u could be left to b (if r > q parabola has a maximum) */
2309: #ifdef DEBUG
1.224 brouard 2310: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2311: 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 2312: #endif
1.126 brouard 2313: u=(*cx)+GOLD*(*cx-*bx);
2314: fu=(*func)(u);
1.224 brouard 2315: #ifdef DEBUG
2316: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2317: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2318: #endif
1.183 brouard 2319: } /* end tests */
1.126 brouard 2320: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2321: SHFT(*fa,*fb,*fc,fu)
2322: #ifdef DEBUG
1.224 brouard 2323: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2324: 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 2325: #endif
2326: } /* 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 2327: }
2328:
2329: /*************** linmin ************************/
1.162 brouard 2330: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2331: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2332: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2333: the value of func at the returned location p . This is actually all accomplished by calling the
2334: routines mnbrak and brent .*/
1.126 brouard 2335: int ncom;
2336: double *pcom,*xicom;
2337: double (*nrfunc)(double []);
2338:
1.224 brouard 2339: #ifdef LINMINORIGINAL
1.126 brouard 2340: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2341: #else
2342: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2343: #endif
1.126 brouard 2344: {
2345: double brent(double ax, double bx, double cx,
2346: double (*f)(double), double tol, double *xmin);
2347: double f1dim(double x);
2348: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2349: double *fc, double (*func)(double));
2350: int j;
2351: double xx,xmin,bx,ax;
2352: double fx,fb,fa;
1.187 brouard 2353:
1.203 brouard 2354: #ifdef LINMINORIGINAL
2355: #else
2356: double scale=10., axs, xxs; /* Scale added for infinity */
2357: #endif
2358:
1.126 brouard 2359: ncom=n;
2360: pcom=vector(1,n);
2361: xicom=vector(1,n);
2362: nrfunc=func;
2363: for (j=1;j<=n;j++) {
2364: pcom[j]=p[j];
1.202 brouard 2365: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2366: }
1.187 brouard 2367:
1.203 brouard 2368: #ifdef LINMINORIGINAL
2369: xx=1.;
2370: #else
2371: axs=0.0;
2372: xxs=1.;
2373: do{
2374: xx= xxs;
2375: #endif
1.187 brouard 2376: ax=0.;
2377: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2378: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2379: /* 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)) */
2380: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2381: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2382: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2383: /* 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 2384: #ifdef LINMINORIGINAL
2385: #else
2386: if (fx != fx){
1.224 brouard 2387: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2388: printf("|");
2389: fprintf(ficlog,"|");
1.203 brouard 2390: #ifdef DEBUGLINMIN
1.224 brouard 2391: 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 2392: #endif
2393: }
1.224 brouard 2394: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2395: #endif
2396:
1.191 brouard 2397: #ifdef DEBUGLINMIN
2398: 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 2399: 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 2400: #endif
1.224 brouard 2401: #ifdef LINMINORIGINAL
2402: #else
1.317 brouard 2403: if(fb == fx){ /* Flat function in the direction */
2404: xmin=xx;
1.224 brouard 2405: *flat=1;
1.317 brouard 2406: }else{
1.224 brouard 2407: *flat=0;
2408: #endif
2409: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2410: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2411: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2412: /* fmin = f(p[j] + xmin * xi[j]) */
2413: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2414: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2415: #ifdef DEBUG
1.224 brouard 2416: 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);
2417: 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);
2418: #endif
2419: #ifdef LINMINORIGINAL
2420: #else
2421: }
1.126 brouard 2422: #endif
1.191 brouard 2423: #ifdef DEBUGLINMIN
2424: printf("linmin end ");
1.202 brouard 2425: fprintf(ficlog,"linmin end ");
1.191 brouard 2426: #endif
1.126 brouard 2427: for (j=1;j<=n;j++) {
1.203 brouard 2428: #ifdef LINMINORIGINAL
2429: xi[j] *= xmin;
2430: #else
2431: #ifdef DEBUGLINMIN
2432: if(xxs <1.0)
2433: printf(" before xi[%d]=%12.8f", j,xi[j]);
2434: #endif
2435: 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) */
2436: #ifdef DEBUGLINMIN
2437: if(xxs <1.0)
2438: 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 );
2439: #endif
2440: #endif
1.187 brouard 2441: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2442: }
1.191 brouard 2443: #ifdef DEBUGLINMIN
1.203 brouard 2444: printf("\n");
1.191 brouard 2445: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2446: 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 2447: for (j=1;j<=n;j++) {
1.202 brouard 2448: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2449: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2450: if(j % ncovmodel == 0){
1.191 brouard 2451: printf("\n");
1.202 brouard 2452: fprintf(ficlog,"\n");
2453: }
1.191 brouard 2454: }
1.203 brouard 2455: #else
1.191 brouard 2456: #endif
1.126 brouard 2457: free_vector(xicom,1,n);
2458: free_vector(pcom,1,n);
2459: }
2460:
2461:
2462: /*************** powell ************************/
1.162 brouard 2463: /*
1.317 brouard 2464: Minimization of a function func of n variables. Input consists in an initial starting point
2465: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2466: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2467: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2468: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2469: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2470: */
1.224 brouard 2471: #ifdef LINMINORIGINAL
2472: #else
2473: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2474: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2475: #endif
1.126 brouard 2476: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2477: double (*func)(double []))
2478: {
1.224 brouard 2479: #ifdef LINMINORIGINAL
2480: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2481: double (*func)(double []));
1.224 brouard 2482: #else
1.241 brouard 2483: void linmin(double p[], double xi[], int n, double *fret,
2484: double (*func)(double []),int *flat);
1.224 brouard 2485: #endif
1.239 brouard 2486: int i,ibig,j,jk,k;
1.126 brouard 2487: double del,t,*pt,*ptt,*xit;
1.181 brouard 2488: double directest;
1.126 brouard 2489: double fp,fptt;
2490: double *xits;
2491: int niterf, itmp;
2492:
2493: pt=vector(1,n);
2494: ptt=vector(1,n);
2495: xit=vector(1,n);
2496: xits=vector(1,n);
2497: *fret=(*func)(p);
2498: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2499: rcurr_time = time(NULL);
1.126 brouard 2500: for (*iter=1;;++(*iter)) {
2501: ibig=0;
2502: del=0.0;
1.157 brouard 2503: rlast_time=rcurr_time;
2504: /* (void) gettimeofday(&curr_time,&tzp); */
2505: rcurr_time = time(NULL);
2506: curr_time = *localtime(&rcurr_time);
1.324 brouard 2507: 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);
2508: 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);
1.157 brouard 2509: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2510: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2511: for (i=1;i<=n;i++) {
1.126 brouard 2512: fprintf(ficrespow," %.12lf", p[i]);
2513: }
1.239 brouard 2514: fprintf(ficrespow,"\n");fflush(ficrespow);
2515: printf("\n#model= 1 + age ");
2516: fprintf(ficlog,"\n#model= 1 + age ");
2517: if(nagesqr==1){
1.241 brouard 2518: printf(" + age*age ");
2519: fprintf(ficlog," + age*age ");
1.239 brouard 2520: }
2521: for(j=1;j <=ncovmodel-2;j++){
2522: if(Typevar[j]==0) {
2523: printf(" + V%d ",Tvar[j]);
2524: fprintf(ficlog," + V%d ",Tvar[j]);
2525: }else if(Typevar[j]==1) {
2526: printf(" + V%d*age ",Tvar[j]);
2527: fprintf(ficlog," + V%d*age ",Tvar[j]);
2528: }else if(Typevar[j]==2) {
2529: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2530: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2531: }
2532: }
1.126 brouard 2533: printf("\n");
1.239 brouard 2534: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2535: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2536: fprintf(ficlog,"\n");
1.239 brouard 2537: for(i=1,jk=1; i <=nlstate; i++){
2538: for(k=1; k <=(nlstate+ndeath); k++){
2539: if (k != i) {
2540: printf("%d%d ",i,k);
2541: fprintf(ficlog,"%d%d ",i,k);
2542: for(j=1; j <=ncovmodel; j++){
2543: printf("%12.7f ",p[jk]);
2544: fprintf(ficlog,"%12.7f ",p[jk]);
2545: jk++;
2546: }
2547: printf("\n");
2548: fprintf(ficlog,"\n");
2549: }
2550: }
2551: }
1.241 brouard 2552: if(*iter <=3 && *iter >1){
1.157 brouard 2553: tml = *localtime(&rcurr_time);
2554: strcpy(strcurr,asctime(&tml));
2555: rforecast_time=rcurr_time;
1.126 brouard 2556: itmp = strlen(strcurr);
2557: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2558: strcurr[itmp-1]='\0';
1.162 brouard 2559: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2560: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2561: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2562: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2563: forecast_time = *localtime(&rforecast_time);
2564: strcpy(strfor,asctime(&forecast_time));
2565: itmp = strlen(strfor);
2566: if(strfor[itmp-1]=='\n')
2567: strfor[itmp-1]='\0';
2568: 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);
2569: 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 2570: }
2571: }
1.187 brouard 2572: for (i=1;i<=n;i++) { /* For each direction i */
2573: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2574: fptt=(*fret);
2575: #ifdef DEBUG
1.203 brouard 2576: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2577: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2578: #endif
1.203 brouard 2579: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2580: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2581: #ifdef LINMINORIGINAL
1.188 brouard 2582: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2583: #else
2584: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2585: flatdir[i]=flat; /* Function is vanishing in that direction i */
2586: #endif
2587: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2588: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2589: /* because that direction will be replaced unless the gain del is small */
2590: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2591: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2592: /* with the new direction. */
2593: del=fabs(fptt-(*fret));
2594: ibig=i;
1.126 brouard 2595: }
2596: #ifdef DEBUG
2597: printf("%d %.12e",i,(*fret));
2598: fprintf(ficlog,"%d %.12e",i,(*fret));
2599: for (j=1;j<=n;j++) {
1.224 brouard 2600: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2601: printf(" x(%d)=%.12e",j,xit[j]);
2602: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2603: }
2604: for(j=1;j<=n;j++) {
1.225 brouard 2605: printf(" p(%d)=%.12e",j,p[j]);
2606: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2607: }
2608: printf("\n");
2609: fprintf(ficlog,"\n");
2610: #endif
1.187 brouard 2611: } /* end loop on each direction i */
2612: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2613: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2614: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2615: for(j=1;j<=n;j++) {
2616: if(flatdir[j] >0){
2617: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2618: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2619: }
1.319 brouard 2620: /* printf("\n"); */
2621: /* fprintf(ficlog,"\n"); */
2622: }
1.243 brouard 2623: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2624: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2625: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2626: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2627: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2628: /* decreased of more than 3.84 */
2629: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2630: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2631: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2632:
1.188 brouard 2633: /* Starting the program with initial values given by a former maximization will simply change */
2634: /* the scales of the directions and the directions, because the are reset to canonical directions */
2635: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2636: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2637: #ifdef DEBUG
2638: int k[2],l;
2639: k[0]=1;
2640: k[1]=-1;
2641: printf("Max: %.12e",(*func)(p));
2642: fprintf(ficlog,"Max: %.12e",(*func)(p));
2643: for (j=1;j<=n;j++) {
2644: printf(" %.12e",p[j]);
2645: fprintf(ficlog," %.12e",p[j]);
2646: }
2647: printf("\n");
2648: fprintf(ficlog,"\n");
2649: for(l=0;l<=1;l++) {
2650: for (j=1;j<=n;j++) {
2651: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2652: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2653: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2654: }
2655: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2656: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2657: }
2658: #endif
2659:
2660: free_vector(xit,1,n);
2661: free_vector(xits,1,n);
2662: free_vector(ptt,1,n);
2663: free_vector(pt,1,n);
2664: return;
1.192 brouard 2665: } /* enough precision */
1.240 brouard 2666: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2667: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2668: ptt[j]=2.0*p[j]-pt[j];
2669: xit[j]=p[j]-pt[j];
2670: pt[j]=p[j];
2671: }
1.181 brouard 2672: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2673: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2674: if (*iter <=4) {
1.225 brouard 2675: #else
2676: #endif
1.224 brouard 2677: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2678: #else
1.161 brouard 2679: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2680: #endif
1.162 brouard 2681: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2682: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2683: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2684: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2685: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2686: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2687: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2688: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2689: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2690: /* Even if f3 <f1, directest can be negative and t >0 */
2691: /* mu² and del² are equal when f3=f1 */
2692: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2693: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2694: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2695: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2696: #ifdef NRCORIGINAL
2697: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2698: #else
2699: 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 2700: t= t- del*SQR(fp-fptt);
1.183 brouard 2701: #endif
1.202 brouard 2702: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2703: #ifdef DEBUG
1.181 brouard 2704: 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);
2705: 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 2706: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2707: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2708: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2709: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2710: 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);
2711: 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);
2712: #endif
1.183 brouard 2713: #ifdef POWELLORIGINAL
2714: if (t < 0.0) { /* Then we use it for new direction */
2715: #else
1.182 brouard 2716: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2717: 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 2718: 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 2719: 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 2720: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2721: }
1.181 brouard 2722: if (directest < 0.0) { /* Then we use it for new direction */
2723: #endif
1.191 brouard 2724: #ifdef DEBUGLINMIN
1.234 brouard 2725: printf("Before linmin in direction P%d-P0\n",n);
2726: for (j=1;j<=n;j++) {
2727: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2728: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2729: if(j % ncovmodel == 0){
2730: printf("\n");
2731: fprintf(ficlog,"\n");
2732: }
2733: }
1.224 brouard 2734: #endif
2735: #ifdef LINMINORIGINAL
1.234 brouard 2736: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2737: #else
1.234 brouard 2738: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2739: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2740: #endif
1.234 brouard 2741:
1.191 brouard 2742: #ifdef DEBUGLINMIN
1.234 brouard 2743: for (j=1;j<=n;j++) {
2744: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2745: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2746: if(j % ncovmodel == 0){
2747: printf("\n");
2748: fprintf(ficlog,"\n");
2749: }
2750: }
1.224 brouard 2751: #endif
1.234 brouard 2752: for (j=1;j<=n;j++) {
2753: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2754: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2755: }
1.224 brouard 2756: #ifdef LINMINORIGINAL
2757: #else
1.234 brouard 2758: for (j=1, flatd=0;j<=n;j++) {
2759: if(flatdir[j]>0)
2760: flatd++;
2761: }
2762: if(flatd >0){
1.255 brouard 2763: printf("%d flat directions: ",flatd);
2764: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2765: for (j=1;j<=n;j++) {
2766: if(flatdir[j]>0){
2767: printf("%d ",j);
2768: fprintf(ficlog,"%d ",j);
2769: }
2770: }
2771: printf("\n");
2772: fprintf(ficlog,"\n");
1.319 brouard 2773: #ifdef FLATSUP
2774: free_vector(xit,1,n);
2775: free_vector(xits,1,n);
2776: free_vector(ptt,1,n);
2777: free_vector(pt,1,n);
2778: return;
2779: #endif
1.234 brouard 2780: }
1.191 brouard 2781: #endif
1.234 brouard 2782: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2783: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2784:
1.126 brouard 2785: #ifdef DEBUG
1.234 brouard 2786: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2787: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2788: for(j=1;j<=n;j++){
2789: printf(" %lf",xit[j]);
2790: fprintf(ficlog," %lf",xit[j]);
2791: }
2792: printf("\n");
2793: fprintf(ficlog,"\n");
1.126 brouard 2794: #endif
1.192 brouard 2795: } /* end of t or directest negative */
1.224 brouard 2796: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2797: #else
1.234 brouard 2798: } /* end if (fptt < fp) */
1.192 brouard 2799: #endif
1.225 brouard 2800: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2801: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2802: #else
1.224 brouard 2803: #endif
1.234 brouard 2804: } /* loop iteration */
1.126 brouard 2805: }
1.234 brouard 2806:
1.126 brouard 2807: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2808:
1.235 brouard 2809: 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 2810: {
1.279 brouard 2811: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2812: * (and selected quantitative values in nres)
2813: * by left multiplying the unit
2814: * matrix by transitions matrix until convergence is reached with precision ftolpl
2815: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2816: * Wx is row vector: population in state 1, population in state 2, population dead
2817: * or prevalence in state 1, prevalence in state 2, 0
2818: * newm is the matrix after multiplications, its rows are identical at a factor.
2819: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2820: * Output is prlim.
2821: * Initial matrix pimij
2822: */
1.206 brouard 2823: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2824: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2825: /* 0, 0 , 1} */
2826: /*
2827: * and after some iteration: */
2828: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2829: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2830: /* 0, 0 , 1} */
2831: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2832: /* {0.51571254859325999, 0.4842874514067399, */
2833: /* 0.51326036147820708, 0.48673963852179264} */
2834: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2835:
1.332 brouard 2836: int i, ii,j,k, k1;
1.209 brouard 2837: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2838: /* double **matprod2(); */ /* test */
1.218 brouard 2839: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2840: double **newm;
1.209 brouard 2841: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2842: int ncvloop=0;
1.288 brouard 2843: int first=0;
1.169 brouard 2844:
1.209 brouard 2845: min=vector(1,nlstate);
2846: max=vector(1,nlstate);
2847: meandiff=vector(1,nlstate);
2848:
1.218 brouard 2849: /* Starting with matrix unity */
1.126 brouard 2850: for (ii=1;ii<=nlstate+ndeath;ii++)
2851: for (j=1;j<=nlstate+ndeath;j++){
2852: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2853: }
1.169 brouard 2854:
2855: cov[1]=1.;
2856:
2857: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2858: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2859: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2860: ncvloop++;
1.126 brouard 2861: newm=savm;
2862: /* Covariates have to be included here again */
1.138 brouard 2863: cov[2]=agefin;
1.319 brouard 2864: if(nagesqr==1){
2865: cov[3]= agefin*agefin;
2866: }
1.332 brouard 2867: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
2868: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
2869: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
2870: if(Typevar[k1]==1){ /* A product with age */
2871: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
2872: }else{
2873: cov[2+nagesqr+k1]=precov[nres][k1];
2874: }
2875: }/* End of loop on model equation */
2876:
2877: /* Start of old code (replaced by a loop on position in the model equation */
2878: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
2879: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
2880: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
2881: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
2882: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
2883: /* * k 1 2 3 4 5 6 7 8 */
2884: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
2885: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
2886: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
2887: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
2888: /* *nsd=3 (1) (2) (3) */
2889: /* *TvarsD[nsd] [1]=2 1 3 */
2890: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
2891: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
2892: /* *Tage[] [1]=1 [2]=2 [3]=3 */
2893: /* *Tvard[] [1][1]=1 [2][1]=1 */
2894: /* * [1][2]=3 [2][2]=2 */
2895: /* *Tprod[](=k) [1]=1 [2]=8 */
2896: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
2897: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
2898: /* *TvarsDpType */
2899: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
2900: /* * nsd=1 (1) (2) */
2901: /* *TvarsD[nsd] 3 2 */
2902: /* *TnsdVar (3)=1 (2)=2 */
2903: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
2904: /* *Tage[] [1]=2 [2]= 3 */
2905: /* *\/ */
2906: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
2907: /* /\* 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)); *\/ */
2908: /* } */
2909: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
2910: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
2911: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
2912: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
2913: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
2914: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2915: /* /\* 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]); *\/ */
2916: /* } */
2917: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
2918: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
2919: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
2920: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
2921: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
2922: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
2923: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2924: /* } */
2925: /* /\* 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]); *\/ */
2926: /* } */
2927: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
2928: /* /\* 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]); *\/ */
2929: /* if(Dummy[Tvard[k][1]]==0){ */
2930: /* if(Dummy[Tvard[k][2]]==0){ */
2931: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
2932: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
2933: /* }else{ */
2934: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
2935: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
2936: /* } */
2937: /* }else{ */
2938: /* if(Dummy[Tvard[k][2]]==0){ */
2939: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
2940: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
2941: /* }else{ */
2942: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
2943: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
2944: /* } */
2945: /* } */
2946: /* } /\* End product without age *\/ */
2947: /* ENd of old code */
1.138 brouard 2948: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2949: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2950: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2951: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2952: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 2953: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2954: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2955:
1.126 brouard 2956: savm=oldm;
2957: oldm=newm;
1.209 brouard 2958:
2959: for(j=1; j<=nlstate; j++){
2960: max[j]=0.;
2961: min[j]=1.;
2962: }
2963: for(i=1;i<=nlstate;i++){
2964: sumnew=0;
2965: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2966: for(j=1; j<=nlstate; j++){
2967: prlim[i][j]= newm[i][j]/(1-sumnew);
2968: max[j]=FMAX(max[j],prlim[i][j]);
2969: min[j]=FMIN(min[j],prlim[i][j]);
2970: }
2971: }
2972:
1.126 brouard 2973: maxmax=0.;
1.209 brouard 2974: for(j=1; j<=nlstate; j++){
2975: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2976: maxmax=FMAX(maxmax,meandiff[j]);
2977: /* 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 2978: } /* j loop */
1.203 brouard 2979: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2980: /* 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 2981: if(maxmax < ftolpl){
1.209 brouard 2982: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2983: free_vector(min,1,nlstate);
2984: free_vector(max,1,nlstate);
2985: free_vector(meandiff,1,nlstate);
1.126 brouard 2986: return prlim;
2987: }
1.288 brouard 2988: } /* agefin loop */
1.208 brouard 2989: /* After some age loop it doesn't converge */
1.288 brouard 2990: if(!first){
2991: first=1;
2992: 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 2993: 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);
2994: }else if (first >=1 && first <10){
2995: 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);
2996: first++;
2997: }else if (first ==10){
2998: 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);
2999: 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");
3000: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
3001: first++;
1.288 brouard 3002: }
3003:
1.209 brouard 3004: /* 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); */
3005: free_vector(min,1,nlstate);
3006: free_vector(max,1,nlstate);
3007: free_vector(meandiff,1,nlstate);
1.208 brouard 3008:
1.169 brouard 3009: return prlim; /* should not reach here */
1.126 brouard 3010: }
3011:
1.217 brouard 3012:
3013: /**** Back Prevalence limit (stable or period prevalence) ****************/
3014:
1.218 brouard 3015: /* 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) */
3016: /* 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 3017: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 3018: {
1.264 brouard 3019: /* 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 3020: matrix by transitions matrix until convergence is reached with precision ftolpl */
3021: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
3022: /* Wx is row vector: population in state 1, population in state 2, population dead */
3023: /* or prevalence in state 1, prevalence in state 2, 0 */
3024: /* newm is the matrix after multiplications, its rows are identical at a factor */
3025: /* Initial matrix pimij */
3026: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
3027: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
3028: /* 0, 0 , 1} */
3029: /*
3030: * and after some iteration: */
3031: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
3032: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
3033: /* 0, 0 , 1} */
3034: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
3035: /* {0.51571254859325999, 0.4842874514067399, */
3036: /* 0.51326036147820708, 0.48673963852179264} */
3037: /* If we start from prlim again, prlim tends to a constant matrix */
3038:
1.332 brouard 3039: int i, ii,j,k, k1;
1.247 brouard 3040: int first=0;
1.217 brouard 3041: double *min, *max, *meandiff, maxmax,sumnew=0.;
3042: /* double **matprod2(); */ /* test */
3043: double **out, cov[NCOVMAX+1], **bmij();
3044: double **newm;
1.218 brouard 3045: double **dnewm, **doldm, **dsavm; /* for use */
3046: double **oldm, **savm; /* for use */
3047:
1.217 brouard 3048: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
3049: int ncvloop=0;
3050:
3051: min=vector(1,nlstate);
3052: max=vector(1,nlstate);
3053: meandiff=vector(1,nlstate);
3054:
1.266 brouard 3055: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
3056: oldm=oldms; savm=savms;
3057:
3058: /* Starting with matrix unity */
3059: for (ii=1;ii<=nlstate+ndeath;ii++)
3060: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 3061: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3062: }
3063:
3064: cov[1]=1.;
3065:
3066: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3067: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 3068: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 3069: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
3070: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 3071: ncvloop++;
1.218 brouard 3072: newm=savm; /* oldm should be kept from previous iteration or unity at start */
3073: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 3074: /* Covariates have to be included here again */
3075: cov[2]=agefin;
1.319 brouard 3076: if(nagesqr==1){
1.217 brouard 3077: cov[3]= agefin*agefin;;
1.319 brouard 3078: }
1.332 brouard 3079: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3080: if(Typevar[k1]==1){ /* A product with age */
3081: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 3082: }else{
1.332 brouard 3083: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 3084: }
1.332 brouard 3085: }/* End of loop on model equation */
3086:
3087: /* Old code */
3088:
3089: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
3090: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
3091: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
3092: /* /\* 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)); *\/ */
3093: /* } */
3094: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
3095: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3096: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3097: /* /\* /\\* 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])]); *\\/ *\/ */
3098: /* /\* } *\/ */
3099: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3100: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3101: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3102: /* /\* 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]); *\/ */
3103: /* } */
3104: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
3105: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
3106: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
3107: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3108: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3109: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
3110: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3111: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3112: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3113: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3114: /* } */
3115: /* /\* 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]); *\/ */
3116: /* } */
3117: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3118: /* /\* 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]); *\/ */
3119: /* if(Dummy[Tvard[k][1]]==0){ */
3120: /* if(Dummy[Tvard[k][2]]==0){ */
3121: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3122: /* }else{ */
3123: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3124: /* } */
3125: /* }else{ */
3126: /* if(Dummy[Tvard[k][2]]==0){ */
3127: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3128: /* }else{ */
3129: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3130: /* } */
3131: /* } */
3132: /* } */
1.217 brouard 3133:
3134: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3135: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3136: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3137: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3138: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3139: /* ij should be linked to the correct index of cov */
3140: /* age and covariate values ij are in 'cov', but we need to pass
3141: * ij for the observed prevalence at age and status and covariate
3142: * number: prevacurrent[(int)agefin][ii][ij]
3143: */
3144: /* 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 *\/ */
3145: /* 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 *\/ */
3146: 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 3147: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3148: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3149: /* for(i=1; i<=nlstate+ndeath; i++) { */
3150: /* printf("%d newm= ",i); */
3151: /* for(j=1;j<=nlstate+ndeath;j++) { */
3152: /* printf("%f ",newm[i][j]); */
3153: /* } */
3154: /* printf("oldm * "); */
3155: /* for(j=1;j<=nlstate+ndeath;j++) { */
3156: /* printf("%f ",oldm[i][j]); */
3157: /* } */
1.268 brouard 3158: /* printf(" bmmij "); */
1.266 brouard 3159: /* for(j=1;j<=nlstate+ndeath;j++) { */
3160: /* printf("%f ",pmmij[i][j]); */
3161: /* } */
3162: /* printf("\n"); */
3163: /* } */
3164: /* } */
1.217 brouard 3165: savm=oldm;
3166: oldm=newm;
1.266 brouard 3167:
1.217 brouard 3168: for(j=1; j<=nlstate; j++){
3169: max[j]=0.;
3170: min[j]=1.;
3171: }
3172: for(j=1; j<=nlstate; j++){
3173: for(i=1;i<=nlstate;i++){
1.234 brouard 3174: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3175: bprlim[i][j]= newm[i][j];
3176: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3177: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3178: }
3179: }
1.218 brouard 3180:
1.217 brouard 3181: maxmax=0.;
3182: for(i=1; i<=nlstate; i++){
1.318 brouard 3183: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3184: maxmax=FMAX(maxmax,meandiff[i]);
3185: /* 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 3186: } /* i loop */
1.217 brouard 3187: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3188: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3189: if(maxmax < ftolpl){
1.220 brouard 3190: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3191: free_vector(min,1,nlstate);
3192: free_vector(max,1,nlstate);
3193: free_vector(meandiff,1,nlstate);
3194: return bprlim;
3195: }
1.288 brouard 3196: } /* agefin loop */
1.217 brouard 3197: /* After some age loop it doesn't converge */
1.288 brouard 3198: if(!first){
1.247 brouard 3199: first=1;
3200: 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\
3201: 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);
3202: }
3203: 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 3204: 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);
3205: /* 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); */
3206: free_vector(min,1,nlstate);
3207: free_vector(max,1,nlstate);
3208: free_vector(meandiff,1,nlstate);
3209:
3210: return bprlim; /* should not reach here */
3211: }
3212:
1.126 brouard 3213: /*************** transition probabilities ***************/
3214:
3215: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3216: {
1.138 brouard 3217: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3218: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3219: model to the ncovmodel covariates (including constant and age).
3220: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3221: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3222: ncth covariate in the global vector x is given by the formula:
3223: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3224: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3225: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3226: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3227: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3228: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3229: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3230: */
3231: double s1, lnpijopii;
1.126 brouard 3232: /*double t34;*/
1.164 brouard 3233: int i,j, nc, ii, jj;
1.126 brouard 3234:
1.223 brouard 3235: for(i=1; i<= nlstate; i++){
3236: for(j=1; j<i;j++){
3237: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3238: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3239: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3240: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3241: }
3242: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3243: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3244: }
3245: for(j=i+1; j<=nlstate+ndeath;j++){
3246: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3247: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3248: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3249: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3250: }
3251: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3252: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3253: }
3254: }
1.218 brouard 3255:
1.223 brouard 3256: for(i=1; i<= nlstate; i++){
3257: s1=0;
3258: for(j=1; j<i; j++){
3259: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 brouard 3260: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
1.223 brouard 3261: }
3262: for(j=i+1; j<=nlstate+ndeath; j++){
3263: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 brouard 3264: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
1.223 brouard 3265: }
3266: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3267: ps[i][i]=1./(s1+1.);
3268: /* Computing other pijs */
3269: for(j=1; j<i; j++)
1.325 brouard 3270: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3271: for(j=i+1; j<=nlstate+ndeath; j++)
3272: ps[i][j]= exp(ps[i][j])*ps[i][i];
3273: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3274: } /* end i */
1.218 brouard 3275:
1.223 brouard 3276: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3277: for(jj=1; jj<= nlstate+ndeath; jj++){
3278: ps[ii][jj]=0;
3279: ps[ii][ii]=1;
3280: }
3281: }
1.294 brouard 3282:
3283:
1.223 brouard 3284: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3285: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3286: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3287: /* } */
3288: /* printf("\n "); */
3289: /* } */
3290: /* printf("\n ");printf("%lf ",cov[2]);*/
3291: /*
3292: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3293: goto end;*/
1.266 brouard 3294: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3295: }
3296:
1.218 brouard 3297: /*************** backward transition probabilities ***************/
3298:
3299: /* 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 ) */
3300: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3301: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3302: {
1.302 brouard 3303: /* 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 3304: * 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 3305: */
1.218 brouard 3306: int i, ii, j,k;
1.222 brouard 3307:
3308: double **out, **pmij();
3309: double sumnew=0.;
1.218 brouard 3310: double agefin;
1.292 brouard 3311: 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 3312: double **dnewm, **dsavm, **doldm;
3313: double **bbmij;
3314:
1.218 brouard 3315: doldm=ddoldms; /* global pointers */
1.222 brouard 3316: dnewm=ddnewms;
3317: dsavm=ddsavms;
1.318 brouard 3318:
3319: /* Debug */
3320: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3321: agefin=cov[2];
1.268 brouard 3322: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3323: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3324: the observed prevalence (with this covariate ij) at beginning of transition */
3325: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3326:
3327: /* P_x */
1.325 brouard 3328: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3329: /* outputs pmmij which is a stochastic matrix in row */
3330:
3331: /* Diag(w_x) */
1.292 brouard 3332: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3333: sumnew=0.;
1.269 brouard 3334: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3335: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3336: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3337: sumnew+=prevacurrent[(int)agefin][ii][ij];
3338: }
3339: if(sumnew >0.01){ /* At least some value in the prevalence */
3340: for (ii=1;ii<=nlstate+ndeath;ii++){
3341: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3342: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3343: }
3344: }else{
3345: for (ii=1;ii<=nlstate+ndeath;ii++){
3346: for (j=1;j<=nlstate+ndeath;j++)
3347: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3348: }
3349: /* if(sumnew <0.9){ */
3350: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3351: /* } */
3352: }
3353: k3=0.0; /* We put the last diagonal to 0 */
3354: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3355: doldm[ii][ii]= k3;
3356: }
3357: /* End doldm, At the end doldm is diag[(w_i)] */
3358:
1.292 brouard 3359: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3360: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3361:
1.292 brouard 3362: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3363: /* 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 3364: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3365: sumnew=0.;
1.222 brouard 3366: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3367: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3368: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3369: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3370: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3371: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3372: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3373: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3374: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3375: /* }else */
1.268 brouard 3376: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3377: } /*End ii */
3378: } /* 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 */
3379:
1.292 brouard 3380: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3381: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3382: /* end bmij */
1.266 brouard 3383: return ps; /*pointer is unchanged */
1.218 brouard 3384: }
1.217 brouard 3385: /*************** transition probabilities ***************/
3386:
1.218 brouard 3387: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3388: {
3389: /* According to parameters values stored in x and the covariate's values stored in cov,
3390: computes the probability to be observed in state j being in state i by appying the
3391: model to the ncovmodel covariates (including constant and age).
3392: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3393: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3394: ncth covariate in the global vector x is given by the formula:
3395: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3396: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3397: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3398: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3399: Outputs ps[i][j] the probability to be observed in j being in j according to
3400: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3401: */
3402: double s1, lnpijopii;
3403: /*double t34;*/
3404: int i,j, nc, ii, jj;
3405:
1.234 brouard 3406: for(i=1; i<= nlstate; i++){
3407: for(j=1; j<i;j++){
3408: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3409: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3410: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3411: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3412: }
3413: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3414: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3415: }
3416: for(j=i+1; j<=nlstate+ndeath;j++){
3417: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3418: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3419: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3420: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3421: }
3422: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3423: }
3424: }
3425:
3426: for(i=1; i<= nlstate; i++){
3427: s1=0;
3428: for(j=1; j<i; j++){
3429: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3430: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3431: }
3432: for(j=i+1; j<=nlstate+ndeath; j++){
3433: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3434: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3435: }
3436: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3437: ps[i][i]=1./(s1+1.);
3438: /* Computing other pijs */
3439: for(j=1; j<i; j++)
3440: ps[i][j]= exp(ps[i][j])*ps[i][i];
3441: for(j=i+1; j<=nlstate+ndeath; j++)
3442: ps[i][j]= exp(ps[i][j])*ps[i][i];
3443: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3444: } /* end i */
3445:
3446: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3447: for(jj=1; jj<= nlstate+ndeath; jj++){
3448: ps[ii][jj]=0;
3449: ps[ii][ii]=1;
3450: }
3451: }
1.296 brouard 3452: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3453: for(jj=1; jj<= nlstate+ndeath; jj++){
3454: s1=0.;
3455: for(ii=1; ii<= nlstate+ndeath; ii++){
3456: s1+=ps[ii][jj];
3457: }
3458: for(ii=1; ii<= nlstate; ii++){
3459: ps[ii][jj]=ps[ii][jj]/s1;
3460: }
3461: }
3462: /* Transposition */
3463: for(jj=1; jj<= nlstate+ndeath; jj++){
3464: for(ii=jj; ii<= nlstate+ndeath; ii++){
3465: s1=ps[ii][jj];
3466: ps[ii][jj]=ps[jj][ii];
3467: ps[jj][ii]=s1;
3468: }
3469: }
3470: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3471: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3472: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3473: /* } */
3474: /* printf("\n "); */
3475: /* } */
3476: /* printf("\n ");printf("%lf ",cov[2]);*/
3477: /*
3478: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3479: goto end;*/
3480: return ps;
1.217 brouard 3481: }
3482:
3483:
1.126 brouard 3484: /**************** Product of 2 matrices ******************/
3485:
1.145 brouard 3486: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3487: {
3488: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3489: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3490: /* in, b, out are matrice of pointers which should have been initialized
3491: before: only the contents of out is modified. The function returns
3492: a pointer to pointers identical to out */
1.145 brouard 3493: int i, j, k;
1.126 brouard 3494: for(i=nrl; i<= nrh; i++)
1.145 brouard 3495: for(k=ncolol; k<=ncoloh; k++){
3496: out[i][k]=0.;
3497: for(j=ncl; j<=nch; j++)
3498: out[i][k] +=in[i][j]*b[j][k];
3499: }
1.126 brouard 3500: return out;
3501: }
3502:
3503:
3504: /************* Higher Matrix Product ***************/
3505:
1.235 brouard 3506: 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 3507: {
1.332 brouard 3508: /* 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 3509: 'nhstepm*hstepm*stepm' months (i.e. until
3510: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3511: nhstepm*hstepm matrices.
3512: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3513: (typically every 2 years instead of every month which is too big
3514: for the memory).
3515: Model is determined by parameters x and covariates have to be
3516: included manually here.
3517:
3518: */
3519:
1.330 brouard 3520: int i, j, d, h, k, k1;
1.131 brouard 3521: double **out, cov[NCOVMAX+1];
1.126 brouard 3522: double **newm;
1.187 brouard 3523: double agexact;
1.214 brouard 3524: double agebegin, ageend;
1.126 brouard 3525:
3526: /* Hstepm could be zero and should return the unit matrix */
3527: for (i=1;i<=nlstate+ndeath;i++)
3528: for (j=1;j<=nlstate+ndeath;j++){
3529: oldm[i][j]=(i==j ? 1.0 : 0.0);
3530: po[i][j][0]=(i==j ? 1.0 : 0.0);
3531: }
3532: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3533: for(h=1; h <=nhstepm; h++){
3534: for(d=1; d <=hstepm; d++){
3535: newm=savm;
3536: /* Covariates have to be included here again */
3537: cov[1]=1.;
1.214 brouard 3538: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3539: cov[2]=agexact;
1.319 brouard 3540: if(nagesqr==1){
1.227 brouard 3541: cov[3]= agexact*agexact;
1.319 brouard 3542: }
1.330 brouard 3543: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
3544: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
3545: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.332 brouard 3546: if(Typevar[k1]==1){ /* A product with age */
3547: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
3548: }else{
3549: cov[2+nagesqr+k1]=precov[nres][k1];
3550: }
3551: }/* End of loop on model equation */
3552: /* Old code */
3553: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
3554: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
3555: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
3556: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
3557: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
3558: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3559: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3560: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
3561: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
3562: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
3563: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
3564: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
3565: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3566: /* /\* 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]])); *\/ */
3567: /* 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); */
3568: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3569: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
3570: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
3571: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
3572: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
3573: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
3574: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3575: /* 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]]); */
3576: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3577: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
3578: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
3579: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
3580: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
3581: /* 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]); */
3582: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3583:
3584: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
3585: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
3586: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
3587: /* /\* *\/ */
1.330 brouard 3588: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3589: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3590: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 3591: /* /\*cptcovage=2 1 2 *\/ */
3592: /* /\*Tage[k]= 5 8 *\/ */
3593: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
3594: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3595: /* 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]]); */
3596: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3597: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
3598: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
3599: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
3600: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
3601: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
3602: /* /\* 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); *\/ */
3603: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
3604: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
3605: /* /\* } *\/ */
3606: /* /\* 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]); *\/ */
3607: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
3608: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
3609: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
3610: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
3611: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
3612: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
3613: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
3614: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
3615: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 3616:
1.332 brouard 3617: /* /\* 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])]); *\/ */
3618: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3619: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
3620: /* 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]]); */
3621: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3622:
3623: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
3624: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
3625: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3626: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
3627: /* /\* 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]])]; *\/ */
3628: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
3629: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
3630: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
3631: /* /\* } *\/ */
3632: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
3633: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
3634: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
3635: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3636: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
3637: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
3638: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3639: /* /\* } *\/ */
3640: /* /\* }/\\*end of products quantitative *\\/ *\/ */
3641: /* }/\*end of products *\/ */
3642: /* } /\* End of loop on model equation *\/ */
1.235 brouard 3643: /* for (k=1; k<=cptcovn;k++) */
3644: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3645: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3646: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3647: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3648: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3649:
3650:
1.126 brouard 3651: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3652: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3653: /* right multiplication of oldm by the current matrix */
1.126 brouard 3654: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3655: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3656: /* if((int)age == 70){ */
3657: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3658: /* for(i=1; i<=nlstate+ndeath; i++) { */
3659: /* printf("%d pmmij ",i); */
3660: /* for(j=1;j<=nlstate+ndeath;j++) { */
3661: /* printf("%f ",pmmij[i][j]); */
3662: /* } */
3663: /* printf(" oldm "); */
3664: /* for(j=1;j<=nlstate+ndeath;j++) { */
3665: /* printf("%f ",oldm[i][j]); */
3666: /* } */
3667: /* printf("\n"); */
3668: /* } */
3669: /* } */
1.126 brouard 3670: savm=oldm;
3671: oldm=newm;
3672: }
3673: for(i=1; i<=nlstate+ndeath; i++)
3674: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3675: po[i][j][h]=newm[i][j];
3676: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3677: }
1.128 brouard 3678: /*printf("h=%d ",h);*/
1.126 brouard 3679: } /* end h */
1.267 brouard 3680: /* printf("\n H=%d \n",h); */
1.126 brouard 3681: return po;
3682: }
3683:
1.217 brouard 3684: /************* Higher Back Matrix Product ***************/
1.218 brouard 3685: /* 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 3686: 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 3687: {
1.332 brouard 3688: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
3689: computes the transition matrix starting at age 'age' over
1.217 brouard 3690: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3691: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3692: nhstepm*hstepm matrices.
3693: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3694: (typically every 2 years instead of every month which is too big
1.217 brouard 3695: for the memory).
1.218 brouard 3696: Model is determined by parameters x and covariates have to be
1.266 brouard 3697: included manually here. Then we use a call to bmij(x and cov)
3698: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3699: */
1.217 brouard 3700:
1.332 brouard 3701: int i, j, d, h, k, k1;
1.266 brouard 3702: double **out, cov[NCOVMAX+1], **bmij();
3703: double **newm, ***newmm;
1.217 brouard 3704: double agexact;
3705: double agebegin, ageend;
1.222 brouard 3706: double **oldm, **savm;
1.217 brouard 3707:
1.266 brouard 3708: newmm=po; /* To be saved */
3709: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3710: /* Hstepm could be zero and should return the unit matrix */
3711: for (i=1;i<=nlstate+ndeath;i++)
3712: for (j=1;j<=nlstate+ndeath;j++){
3713: oldm[i][j]=(i==j ? 1.0 : 0.0);
3714: po[i][j][0]=(i==j ? 1.0 : 0.0);
3715: }
3716: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3717: for(h=1; h <=nhstepm; h++){
3718: for(d=1; d <=hstepm; d++){
3719: newm=savm;
3720: /* Covariates have to be included here again */
3721: cov[1]=1.;
1.271 brouard 3722: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3723: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3724: /* Debug */
3725: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3726: cov[2]=agexact;
1.332 brouard 3727: if(nagesqr==1){
1.222 brouard 3728: cov[3]= agexact*agexact;
1.332 brouard 3729: }
3730: /** New code */
3731: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3732: if(Typevar[k1]==1){ /* A product with age */
3733: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 3734: }else{
1.332 brouard 3735: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 3736: }
1.332 brouard 3737: }/* End of loop on model equation */
3738: /** End of new code */
3739: /** This was old code */
3740: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
3741: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3742: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3743: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
3744: /* /\* 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)); *\/ */
3745: /* } */
3746: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3747: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3748: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3749: /* /\* 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]); *\/ */
3750: /* } */
3751: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
3752: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
3753: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3754: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3755: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3756: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3757: /* } */
3758: /* /\* 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]); *\/ */
3759: /* } */
3760: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
3761: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3762: /* if(Dummy[Tvard[k][1]]==0){ */
3763: /* if(Dummy[Tvard[k][2]]==0){ */
3764: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
3765: /* }else{ */
3766: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3767: /* } */
3768: /* }else{ */
3769: /* if(Dummy[Tvard[k][2]]==0){ */
3770: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3771: /* }else{ */
3772: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3773: /* } */
3774: /* } */
3775: /* } */
3776: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
3777: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
3778: /** End of old code */
3779:
1.218 brouard 3780: /* Careful transposed matrix */
1.266 brouard 3781: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3782: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3783: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3784: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3785: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3786: /* if((int)age == 70){ */
3787: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3788: /* for(i=1; i<=nlstate+ndeath; i++) { */
3789: /* printf("%d pmmij ",i); */
3790: /* for(j=1;j<=nlstate+ndeath;j++) { */
3791: /* printf("%f ",pmmij[i][j]); */
3792: /* } */
3793: /* printf(" oldm "); */
3794: /* for(j=1;j<=nlstate+ndeath;j++) { */
3795: /* printf("%f ",oldm[i][j]); */
3796: /* } */
3797: /* printf("\n"); */
3798: /* } */
3799: /* } */
3800: savm=oldm;
3801: oldm=newm;
3802: }
3803: for(i=1; i<=nlstate+ndeath; i++)
3804: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3805: po[i][j][h]=newm[i][j];
1.268 brouard 3806: /* if(h==nhstepm) */
3807: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3808: }
1.268 brouard 3809: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3810: } /* end h */
1.268 brouard 3811: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3812: return po;
3813: }
3814:
3815:
1.162 brouard 3816: #ifdef NLOPT
3817: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3818: double fret;
3819: double *xt;
3820: int j;
3821: myfunc_data *d2 = (myfunc_data *) pd;
3822: /* xt = (p1-1); */
3823: xt=vector(1,n);
3824: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3825:
3826: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3827: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3828: printf("Function = %.12lf ",fret);
3829: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3830: printf("\n");
3831: free_vector(xt,1,n);
3832: return fret;
3833: }
3834: #endif
1.126 brouard 3835:
3836: /*************** log-likelihood *************/
3837: double func( double *x)
3838: {
1.226 brouard 3839: int i, ii, j, k, mi, d, kk;
3840: int ioffset=0;
3841: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3842: double **out;
3843: double lli; /* Individual log likelihood */
3844: int s1, s2;
1.228 brouard 3845: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
1.226 brouard 3846: double bbh, survp;
3847: long ipmx;
3848: double agexact;
3849: /*extern weight */
3850: /* We are differentiating ll according to initial status */
3851: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3852: /*for(i=1;i<imx;i++)
3853: printf(" %d\n",s[4][i]);
3854: */
1.162 brouard 3855:
1.226 brouard 3856: ++countcallfunc;
1.162 brouard 3857:
1.226 brouard 3858: cov[1]=1.;
1.126 brouard 3859:
1.226 brouard 3860: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3861: ioffset=0;
1.226 brouard 3862: if(mle==1){
3863: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3864: /* Computes the values of the ncovmodel covariates of the model
3865: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3866: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3867: to be observed in j being in i according to the model.
3868: */
1.243 brouard 3869: ioffset=2+nagesqr ;
1.233 brouard 3870: /* Fixed */
1.319 brouard 3871: for (k=1; k<=ncovf;k++){ /* For each fixed covariate dummu or quant or prod */
3872: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3873: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3874: /* 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 3875: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.319 brouard 3876: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (TvarFind[1]=6)*/
3877: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3878: }
1.226 brouard 3879: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3880: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3881: has been calculated etc */
3882: /* For an individual i, wav[i] gives the number of effective waves */
3883: /* We compute the contribution to Likelihood of each effective transition
3884: mw[mi][i] is real wave of the mi th effectve wave */
3885: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3886: s2=s[mw[mi+1][i]][i];
3887: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3888: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3889: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3890: */
3891: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3892: 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*/
3893: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? */
1.242 brouard 3894: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3895: }
3896: for (ii=1;ii<=nlstate+ndeath;ii++)
3897: for (j=1;j<=nlstate+ndeath;j++){
3898: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3899: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3900: }
3901: for(d=0; d<dh[mi][i]; d++){
3902: newm=savm;
3903: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3904: cov[2]=agexact;
3905: if(nagesqr==1)
3906: cov[3]= agexact*agexact; /* Should be changed here */
3907: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3908: if(!FixedV[Tvar[Tage[kk]]])
3909: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3910: else
3911: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3912: }
3913: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3914: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3915: savm=oldm;
3916: oldm=newm;
3917: } /* end mult */
3918:
3919: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3920: /* But now since version 0.9 we anticipate for bias at large stepm.
3921: * If stepm is larger than one month (smallest stepm) and if the exact delay
3922: * (in months) between two waves is not a multiple of stepm, we rounded to
3923: * the nearest (and in case of equal distance, to the lowest) interval but now
3924: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3925: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3926: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3927: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3928: * -stepm/2 to stepm/2 .
3929: * For stepm=1 the results are the same as for previous versions of Imach.
3930: * For stepm > 1 the results are less biased than in previous versions.
3931: */
1.234 brouard 3932: s1=s[mw[mi][i]][i];
3933: s2=s[mw[mi+1][i]][i];
3934: bbh=(double)bh[mi][i]/(double)stepm;
3935: /* bias bh is positive if real duration
3936: * is higher than the multiple of stepm and negative otherwise.
3937: */
3938: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3939: if( s2 > nlstate){
3940: /* i.e. if s2 is a death state and if the date of death is known
3941: then the contribution to the likelihood is the probability to
3942: die between last step unit time and current step unit time,
3943: which is also equal to probability to die before dh
3944: minus probability to die before dh-stepm .
3945: In version up to 0.92 likelihood was computed
3946: as if date of death was unknown. Death was treated as any other
3947: health state: the date of the interview describes the actual state
3948: and not the date of a change in health state. The former idea was
3949: to consider that at each interview the state was recorded
3950: (healthy, disable or death) and IMaCh was corrected; but when we
3951: introduced the exact date of death then we should have modified
3952: the contribution of an exact death to the likelihood. This new
3953: contribution is smaller and very dependent of the step unit
3954: stepm. It is no more the probability to die between last interview
3955: and month of death but the probability to survive from last
3956: interview up to one month before death multiplied by the
3957: probability to die within a month. Thanks to Chris
3958: Jackson for correcting this bug. Former versions increased
3959: mortality artificially. The bad side is that we add another loop
3960: which slows down the processing. The difference can be up to 10%
3961: lower mortality.
3962: */
3963: /* If, at the beginning of the maximization mostly, the
3964: cumulative probability or probability to be dead is
3965: constant (ie = 1) over time d, the difference is equal to
3966: 0. out[s1][3] = savm[s1][3]: probability, being at state
3967: s1 at precedent wave, to be dead a month before current
3968: wave is equal to probability, being at state s1 at
3969: precedent wave, to be dead at mont of the current
3970: wave. Then the observed probability (that this person died)
3971: is null according to current estimated parameter. In fact,
3972: it should be very low but not zero otherwise the log go to
3973: infinity.
3974: */
1.183 brouard 3975: /* #ifdef INFINITYORIGINAL */
3976: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3977: /* #else */
3978: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3979: /* lli=log(mytinydouble); */
3980: /* else */
3981: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3982: /* #endif */
1.226 brouard 3983: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3984:
1.226 brouard 3985: } else if ( s2==-1 ) { /* alive */
3986: for (j=1,survp=0. ; j<=nlstate; j++)
3987: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3988: /*survp += out[s1][j]; */
3989: lli= log(survp);
3990: }
3991: else if (s2==-4) {
3992: for (j=3,survp=0. ; j<=nlstate; j++)
3993: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3994: lli= log(survp);
3995: }
3996: else if (s2==-5) {
3997: for (j=1,survp=0. ; j<=2; j++)
3998: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3999: lli= log(survp);
4000: }
4001: else{
4002: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4003: /* 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 */
4004: }
4005: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
4006: /*if(lli ==000.0)*/
4007: /*printf("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); */
4008: ipmx +=1;
4009: sw += weight[i];
4010: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4011: /* if (lli < log(mytinydouble)){ */
4012: /* 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); */
4013: /* 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]); */
4014: /* } */
4015: } /* end of wave */
4016: } /* end of individual */
4017: } else if(mle==2){
4018: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 4019: ioffset=2+nagesqr ;
4020: for (k=1; k<=ncovf;k++)
4021: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 4022: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 4023: for(k=1; k <= ncovv ; k++){
4024: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4025: }
1.226 brouard 4026: for (ii=1;ii<=nlstate+ndeath;ii++)
4027: for (j=1;j<=nlstate+ndeath;j++){
4028: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4029: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4030: }
4031: for(d=0; d<=dh[mi][i]; d++){
4032: newm=savm;
4033: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4034: cov[2]=agexact;
4035: if(nagesqr==1)
4036: cov[3]= agexact*agexact;
4037: for (kk=1; kk<=cptcovage;kk++) {
4038: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4039: }
4040: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4041: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4042: savm=oldm;
4043: oldm=newm;
4044: } /* end mult */
4045:
4046: s1=s[mw[mi][i]][i];
4047: s2=s[mw[mi+1][i]][i];
4048: bbh=(double)bh[mi][i]/(double)stepm;
4049: 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 */
4050: ipmx +=1;
4051: sw += weight[i];
4052: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4053: } /* end of wave */
4054: } /* end of individual */
4055: } else if(mle==3){ /* exponential inter-extrapolation */
4056: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4057: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4058: for(mi=1; mi<= wav[i]-1; mi++){
4059: for (ii=1;ii<=nlstate+ndeath;ii++)
4060: for (j=1;j<=nlstate+ndeath;j++){
4061: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4062: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4063: }
4064: for(d=0; d<dh[mi][i]; d++){
4065: newm=savm;
4066: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4067: cov[2]=agexact;
4068: if(nagesqr==1)
4069: cov[3]= agexact*agexact;
4070: for (kk=1; kk<=cptcovage;kk++) {
4071: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4072: }
4073: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4074: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4075: savm=oldm;
4076: oldm=newm;
4077: } /* end mult */
4078:
4079: s1=s[mw[mi][i]][i];
4080: s2=s[mw[mi+1][i]][i];
4081: bbh=(double)bh[mi][i]/(double)stepm;
4082: 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 */
4083: ipmx +=1;
4084: sw += weight[i];
4085: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4086: } /* end of wave */
4087: } /* end of individual */
4088: }else if (mle==4){ /* ml=4 no inter-extrapolation */
4089: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4090: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4091: for(mi=1; mi<= wav[i]-1; mi++){
4092: for (ii=1;ii<=nlstate+ndeath;ii++)
4093: for (j=1;j<=nlstate+ndeath;j++){
4094: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4095: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4096: }
4097: for(d=0; d<dh[mi][i]; d++){
4098: newm=savm;
4099: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4100: cov[2]=agexact;
4101: if(nagesqr==1)
4102: cov[3]= agexact*agexact;
4103: for (kk=1; kk<=cptcovage;kk++) {
4104: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4105: }
1.126 brouard 4106:
1.226 brouard 4107: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4108: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4109: savm=oldm;
4110: oldm=newm;
4111: } /* end mult */
4112:
4113: s1=s[mw[mi][i]][i];
4114: s2=s[mw[mi+1][i]][i];
4115: if( s2 > nlstate){
4116: lli=log(out[s1][s2] - savm[s1][s2]);
4117: } else if ( s2==-1 ) { /* alive */
4118: for (j=1,survp=0. ; j<=nlstate; j++)
4119: survp += out[s1][j];
4120: lli= log(survp);
4121: }else{
4122: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4123: }
4124: ipmx +=1;
4125: sw += weight[i];
4126: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 4127: /* 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]); */
1.226 brouard 4128: } /* end of wave */
4129: } /* end of individual */
4130: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4131: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4132: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4133: for(mi=1; mi<= wav[i]-1; mi++){
4134: for (ii=1;ii<=nlstate+ndeath;ii++)
4135: for (j=1;j<=nlstate+ndeath;j++){
4136: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4137: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4138: }
4139: for(d=0; d<dh[mi][i]; d++){
4140: newm=savm;
4141: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4142: cov[2]=agexact;
4143: if(nagesqr==1)
4144: cov[3]= agexact*agexact;
4145: for (kk=1; kk<=cptcovage;kk++) {
4146: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4147: }
1.126 brouard 4148:
1.226 brouard 4149: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4150: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4151: savm=oldm;
4152: oldm=newm;
4153: } /* end mult */
4154:
4155: s1=s[mw[mi][i]][i];
4156: s2=s[mw[mi+1][i]][i];
4157: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4158: ipmx +=1;
4159: sw += weight[i];
4160: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4161: /*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]);*/
4162: } /* end of wave */
4163: } /* end of individual */
4164: } /* End of if */
4165: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4166: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4167: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4168: return -l;
1.126 brouard 4169: }
4170:
4171: /*************** log-likelihood *************/
4172: double funcone( double *x)
4173: {
1.228 brouard 4174: /* Same as func but slower because of a lot of printf and if */
1.335 ! brouard 4175: int i, ii, j, k, mi, d, kk, kf=0;
1.228 brouard 4176: int ioffset=0;
1.131 brouard 4177: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4178: double **out;
4179: double lli; /* Individual log likelihood */
4180: double llt;
4181: int s1, s2;
1.228 brouard 4182: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4183:
1.126 brouard 4184: double bbh, survp;
1.187 brouard 4185: double agexact;
1.214 brouard 4186: double agebegin, ageend;
1.126 brouard 4187: /*extern weight */
4188: /* We are differentiating ll according to initial status */
4189: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4190: /*for(i=1;i<imx;i++)
4191: printf(" %d\n",s[4][i]);
4192: */
4193: cov[1]=1.;
4194:
4195: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4196: ioffset=0;
4197: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 4198: /* ioffset=2+nagesqr+cptcovage; */
4199: ioffset=2+nagesqr;
1.232 brouard 4200: /* Fixed */
1.224 brouard 4201: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4202: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.335 ! brouard 4203: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
! 4204: 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 4205: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4206: /* cov[2+6]=covar[Tvar[6]][i]; */
4207: /* cov[2+6]=covar[2][i]; V2 */
4208: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4209: /* cov[2+7]=covar[Tvar[7]][i]; */
4210: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4211: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4212: /* cov[2+9]=covar[Tvar[9]][i]; */
4213: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4214: }
1.232 brouard 4215: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4216: /* 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?)*\/ */
4217: /* } */
1.231 brouard 4218: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4219: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4220: /* } */
1.225 brouard 4221:
1.233 brouard 4222:
4223: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 4224: /* Wave varying (but not age varying) */
4225: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 4226: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
4227: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4228: }
1.232 brouard 4229: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 4230: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4231: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4232: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4233: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4234: /* 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]); */
1.232 brouard 4235: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4236: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4237: /* /\* 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]); *\/ */
4238: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4239: /* } */
1.126 brouard 4240: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4241: for (j=1;j<=nlstate+ndeath;j++){
4242: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4243: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4244: }
1.214 brouard 4245:
4246: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4247: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4248: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4249: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4250: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4251: and mw[mi+1][i]. dh depends on stepm.*/
4252: newm=savm;
1.247 brouard 4253: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4254: cov[2]=agexact;
4255: if(nagesqr==1)
4256: cov[3]= agexact*agexact;
4257: for (kk=1; kk<=cptcovage;kk++) {
4258: if(!FixedV[Tvar[Tage[kk]]])
4259: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4260: else
4261: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
4262: }
4263: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4264: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4265: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4266: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4267: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4268: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4269: savm=oldm;
4270: oldm=newm;
1.126 brouard 4271: } /* end mult */
4272:
4273: s1=s[mw[mi][i]][i];
4274: s2=s[mw[mi+1][i]][i];
1.217 brouard 4275: /* if(s2==-1){ */
1.268 brouard 4276: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4277: /* /\* exit(1); *\/ */
4278: /* } */
1.126 brouard 4279: bbh=(double)bh[mi][i]/(double)stepm;
4280: /* bias is positive if real duration
4281: * is higher than the multiple of stepm and negative otherwise.
4282: */
4283: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4284: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4285: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4286: for (j=1,survp=0. ; j<=nlstate; j++)
4287: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4288: lli= log(survp);
1.126 brouard 4289: }else if (mle==1){
1.242 brouard 4290: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4291: } else if(mle==2){
1.242 brouard 4292: 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 4293: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4294: 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 4295: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4296: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4297: } else{ /* mle=0 back to 1 */
1.242 brouard 4298: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4299: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4300: } /* End of if */
4301: ipmx +=1;
4302: sw += weight[i];
4303: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.335 ! brouard 4304: /* printf("Funcone 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],(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.126 brouard 4305: if(globpr){
1.246 brouard 4306: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4307: %11.6f %11.6f %11.6f ", \
1.242 brouard 4308: 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 4309: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.335 ! brouard 4310: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
! 4311: /* %11.6f %11.6f %11.6f ", \ */
! 4312: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
! 4313: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 4314: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4315: llt +=ll[k]*gipmx/gsw;
4316: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 ! brouard 4317: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 4318: }
4319: fprintf(ficresilk," %10.6f\n", -llt);
1.335 ! brouard 4320: /* printf(" %10.6f\n", -llt); */
1.126 brouard 4321: }
1.335 ! brouard 4322: } /* end of wave */
! 4323: } /* end of individual */
! 4324: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 4325: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 ! brouard 4326: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
! 4327: if(globpr==0){ /* First time we count the contributions and weights */
! 4328: gipmx=ipmx;
! 4329: gsw=sw;
! 4330: }
1.232 brouard 4331: return -l;
1.126 brouard 4332: }
4333:
4334:
4335: /*************** function likelione ***********/
1.292 brouard 4336: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4337: {
4338: /* This routine should help understanding what is done with
4339: the selection of individuals/waves and
4340: to check the exact contribution to the likelihood.
4341: Plotting could be done.
4342: */
4343: int k;
4344:
4345: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4346: strcpy(fileresilk,"ILK_");
1.202 brouard 4347: strcat(fileresilk,fileresu);
1.126 brouard 4348: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4349: printf("Problem with resultfile: %s\n", fileresilk);
4350: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4351: }
1.214 brouard 4352: 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");
4353: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4354: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4355: for(k=1; k<=nlstate; k++)
4356: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4357: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4358: }
4359:
1.292 brouard 4360: *fretone=(*func)(p);
1.126 brouard 4361: if(*globpri !=0){
4362: fclose(ficresilk);
1.205 brouard 4363: if (mle ==0)
4364: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4365: else if(mle >=1)
4366: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4367: 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 4368: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4369:
4370: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4371: 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> \
1.208 brouard 4372: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4373: }
1.207 brouard 4374: 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> \
1.204 brouard 4375: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4376: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4377: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4378: fflush(fichtm);
1.205 brouard 4379: }
1.126 brouard 4380: return;
4381: }
4382:
4383:
4384: /*********** Maximum Likelihood Estimation ***************/
4385:
4386: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4387: {
1.319 brouard 4388: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4389: double **xi;
4390: double fret;
4391: double fretone; /* Only one call to likelihood */
4392: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4393:
4394: #ifdef NLOPT
4395: int creturn;
4396: nlopt_opt opt;
4397: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4398: double *lb;
4399: double minf; /* the minimum objective value, upon return */
4400: double * p1; /* Shifted parameters from 0 instead of 1 */
4401: myfunc_data dinst, *d = &dinst;
4402: #endif
4403:
4404:
1.126 brouard 4405: xi=matrix(1,npar,1,npar);
4406: for (i=1;i<=npar;i++)
4407: for (j=1;j<=npar;j++)
4408: xi[i][j]=(i==j ? 1.0 : 0.0);
4409: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4410: strcpy(filerespow,"POW_");
1.126 brouard 4411: strcat(filerespow,fileres);
4412: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4413: printf("Problem with resultfile: %s\n", filerespow);
4414: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4415: }
4416: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4417: for (i=1;i<=nlstate;i++)
4418: for(j=1;j<=nlstate+ndeath;j++)
4419: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4420: fprintf(ficrespow,"\n");
1.162 brouard 4421: #ifdef POWELL
1.319 brouard 4422: #ifdef LINMINORIGINAL
4423: #else /* LINMINORIGINAL */
4424:
4425: flatdir=ivector(1,npar);
4426: for (j=1;j<=npar;j++) flatdir[j]=0;
4427: #endif /*LINMINORIGINAL */
4428:
4429: #ifdef FLATSUP
4430: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4431: /* reorganizing p by suppressing flat directions */
4432: for(i=1, jk=1; i <=nlstate; i++){
4433: for(k=1; k <=(nlstate+ndeath); k++){
4434: if (k != i) {
4435: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4436: if(flatdir[jk]==1){
4437: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4438: }
4439: for(j=1; j <=ncovmodel; j++){
4440: printf("%12.7f ",p[jk]);
4441: jk++;
4442: }
4443: printf("\n");
4444: }
4445: }
4446: }
4447: /* skipping */
4448: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4449: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4450: for(k=1; k <=(nlstate+ndeath); k++){
4451: if (k != i) {
4452: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4453: if(flatdir[jk]==1){
4454: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4455: for(j=1; j <=ncovmodel; jk++,j++){
4456: printf(" p[%d]=%12.7f",jk, p[jk]);
4457: /*q[jjk]=p[jk];*/
4458: }
4459: }else{
4460: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4461: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4462: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4463: /*q[jjk]=p[jk];*/
4464: }
4465: }
4466: printf("\n");
4467: }
4468: fflush(stdout);
4469: }
4470: }
4471: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4472: #else /* FLATSUP */
1.126 brouard 4473: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4474: #endif /* FLATSUP */
4475:
4476: #ifdef LINMINORIGINAL
4477: #else
4478: free_ivector(flatdir,1,npar);
4479: #endif /* LINMINORIGINAL*/
4480: #endif /* POWELL */
1.126 brouard 4481:
1.162 brouard 4482: #ifdef NLOPT
4483: #ifdef NEWUOA
4484: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4485: #else
4486: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4487: #endif
4488: lb=vector(0,npar-1);
4489: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4490: nlopt_set_lower_bounds(opt, lb);
4491: nlopt_set_initial_step1(opt, 0.1);
4492:
4493: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4494: d->function = func;
4495: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4496: nlopt_set_min_objective(opt, myfunc, d);
4497: nlopt_set_xtol_rel(opt, ftol);
4498: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4499: printf("nlopt failed! %d\n",creturn);
4500: }
4501: else {
4502: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4503: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4504: iter=1; /* not equal */
4505: }
4506: nlopt_destroy(opt);
4507: #endif
1.319 brouard 4508: #ifdef FLATSUP
4509: /* npared = npar -flatd/ncovmodel; */
4510: /* xired= matrix(1,npared,1,npared); */
4511: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4512: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4513: /* free_matrix(xire,1,npared,1,npared); */
4514: #else /* FLATSUP */
4515: #endif /* FLATSUP */
1.126 brouard 4516: free_matrix(xi,1,npar,1,npar);
4517: fclose(ficrespow);
1.203 brouard 4518: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4519: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4520: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4521:
4522: }
4523:
4524: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4525: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4526: {
4527: double **a,**y,*x,pd;
1.203 brouard 4528: /* double **hess; */
1.164 brouard 4529: int i, j;
1.126 brouard 4530: int *indx;
4531:
4532: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4533: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4534: void lubksb(double **a, int npar, int *indx, double b[]) ;
4535: void ludcmp(double **a, int npar, int *indx, double *d) ;
4536: double gompertz(double p[]);
1.203 brouard 4537: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4538:
4539: printf("\nCalculation of the hessian matrix. Wait...\n");
4540: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4541: for (i=1;i<=npar;i++){
1.203 brouard 4542: printf("%d-",i);fflush(stdout);
4543: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4544:
4545: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4546:
4547: /* printf(" %f ",p[i]);
4548: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4549: }
4550:
4551: for (i=1;i<=npar;i++) {
4552: for (j=1;j<=npar;j++) {
4553: if (j>i) {
1.203 brouard 4554: printf(".%d-%d",i,j);fflush(stdout);
4555: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4556: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4557:
4558: hess[j][i]=hess[i][j];
4559: /*printf(" %lf ",hess[i][j]);*/
4560: }
4561: }
4562: }
4563: printf("\n");
4564: fprintf(ficlog,"\n");
4565:
4566: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4567: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4568:
4569: a=matrix(1,npar,1,npar);
4570: y=matrix(1,npar,1,npar);
4571: x=vector(1,npar);
4572: indx=ivector(1,npar);
4573: for (i=1;i<=npar;i++)
4574: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4575: ludcmp(a,npar,indx,&pd);
4576:
4577: for (j=1;j<=npar;j++) {
4578: for (i=1;i<=npar;i++) x[i]=0;
4579: x[j]=1;
4580: lubksb(a,npar,indx,x);
4581: for (i=1;i<=npar;i++){
4582: matcov[i][j]=x[i];
4583: }
4584: }
4585:
4586: printf("\n#Hessian matrix#\n");
4587: fprintf(ficlog,"\n#Hessian matrix#\n");
4588: for (i=1;i<=npar;i++) {
4589: for (j=1;j<=npar;j++) {
1.203 brouard 4590: printf("%.6e ",hess[i][j]);
4591: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4592: }
4593: printf("\n");
4594: fprintf(ficlog,"\n");
4595: }
4596:
1.203 brouard 4597: /* printf("\n#Covariance matrix#\n"); */
4598: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4599: /* for (i=1;i<=npar;i++) { */
4600: /* for (j=1;j<=npar;j++) { */
4601: /* printf("%.6e ",matcov[i][j]); */
4602: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4603: /* } */
4604: /* printf("\n"); */
4605: /* fprintf(ficlog,"\n"); */
4606: /* } */
4607:
1.126 brouard 4608: /* Recompute Inverse */
1.203 brouard 4609: /* for (i=1;i<=npar;i++) */
4610: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4611: /* ludcmp(a,npar,indx,&pd); */
4612:
4613: /* printf("\n#Hessian matrix recomputed#\n"); */
4614:
4615: /* for (j=1;j<=npar;j++) { */
4616: /* for (i=1;i<=npar;i++) x[i]=0; */
4617: /* x[j]=1; */
4618: /* lubksb(a,npar,indx,x); */
4619: /* for (i=1;i<=npar;i++){ */
4620: /* y[i][j]=x[i]; */
4621: /* printf("%.3e ",y[i][j]); */
4622: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4623: /* } */
4624: /* printf("\n"); */
4625: /* fprintf(ficlog,"\n"); */
4626: /* } */
4627:
4628: /* Verifying the inverse matrix */
4629: #ifdef DEBUGHESS
4630: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4631:
1.203 brouard 4632: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4633: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4634:
4635: for (j=1;j<=npar;j++) {
4636: for (i=1;i<=npar;i++){
1.203 brouard 4637: printf("%.2f ",y[i][j]);
4638: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4639: }
4640: printf("\n");
4641: fprintf(ficlog,"\n");
4642: }
1.203 brouard 4643: #endif
1.126 brouard 4644:
4645: free_matrix(a,1,npar,1,npar);
4646: free_matrix(y,1,npar,1,npar);
4647: free_vector(x,1,npar);
4648: free_ivector(indx,1,npar);
1.203 brouard 4649: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4650:
4651:
4652: }
4653:
4654: /*************** hessian matrix ****************/
4655: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4656: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4657: int i;
4658: int l=1, lmax=20;
1.203 brouard 4659: double k1,k2, res, fx;
1.132 brouard 4660: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4661: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4662: int k=0,kmax=10;
4663: double l1;
4664:
4665: fx=func(x);
4666: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4667: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4668: l1=pow(10,l);
4669: delts=delt;
4670: for(k=1 ; k <kmax; k=k+1){
4671: delt = delta*(l1*k);
4672: p2[theta]=x[theta] +delt;
1.145 brouard 4673: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4674: p2[theta]=x[theta]-delt;
4675: k2=func(p2)-fx;
4676: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4677: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4678:
1.203 brouard 4679: #ifdef DEBUGHESSII
1.126 brouard 4680: 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);
4681: 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);
4682: #endif
4683: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4684: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4685: k=kmax;
4686: }
4687: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4688: k=kmax; l=lmax*10;
1.126 brouard 4689: }
4690: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4691: delts=delt;
4692: }
1.203 brouard 4693: } /* End loop k */
1.126 brouard 4694: }
4695: delti[theta]=delts;
4696: return res;
4697:
4698: }
4699:
1.203 brouard 4700: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4701: {
4702: int i;
1.164 brouard 4703: int l=1, lmax=20;
1.126 brouard 4704: double k1,k2,k3,k4,res,fx;
1.132 brouard 4705: double p2[MAXPARM+1];
1.203 brouard 4706: int k, kmax=1;
4707: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4708:
4709: int firstime=0;
1.203 brouard 4710:
1.126 brouard 4711: fx=func(x);
1.203 brouard 4712: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4713: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4714: p2[thetai]=x[thetai]+delti[thetai]*k;
4715: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4716: k1=func(p2)-fx;
4717:
1.203 brouard 4718: p2[thetai]=x[thetai]+delti[thetai]*k;
4719: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4720: k2=func(p2)-fx;
4721:
1.203 brouard 4722: p2[thetai]=x[thetai]-delti[thetai]*k;
4723: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4724: k3=func(p2)-fx;
4725:
1.203 brouard 4726: p2[thetai]=x[thetai]-delti[thetai]*k;
4727: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4728: k4=func(p2)-fx;
1.203 brouard 4729: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4730: if(k1*k2*k3*k4 <0.){
1.208 brouard 4731: firstime=1;
1.203 brouard 4732: kmax=kmax+10;
1.208 brouard 4733: }
4734: if(kmax >=10 || firstime ==1){
1.246 brouard 4735: 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);
4736: 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 4737: 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);
4738: 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);
4739: }
4740: #ifdef DEBUGHESSIJ
4741: v1=hess[thetai][thetai];
4742: v2=hess[thetaj][thetaj];
4743: cv12=res;
4744: /* Computing eigen value of Hessian matrix */
4745: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4746: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4747: if ((lc2 <0) || (lc1 <0) ){
4748: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4749: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4750: 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);
4751: 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);
4752: }
1.126 brouard 4753: #endif
4754: }
4755: return res;
4756: }
4757:
1.203 brouard 4758: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4759: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4760: /* { */
4761: /* int i; */
4762: /* int l=1, lmax=20; */
4763: /* double k1,k2,k3,k4,res,fx; */
4764: /* double p2[MAXPARM+1]; */
4765: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4766: /* int k=0,kmax=10; */
4767: /* double l1; */
4768:
4769: /* fx=func(x); */
4770: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4771: /* l1=pow(10,l); */
4772: /* delts=delt; */
4773: /* for(k=1 ; k <kmax; k=k+1){ */
4774: /* delt = delti*(l1*k); */
4775: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4776: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4777: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4778: /* k1=func(p2)-fx; */
4779:
4780: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4781: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4782: /* k2=func(p2)-fx; */
4783:
4784: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4785: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4786: /* k3=func(p2)-fx; */
4787:
4788: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4789: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4790: /* k4=func(p2)-fx; */
4791: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4792: /* #ifdef DEBUGHESSIJ */
4793: /* 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); */
4794: /* 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); */
4795: /* #endif */
4796: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4797: /* k=kmax; */
4798: /* } */
4799: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4800: /* k=kmax; l=lmax*10; */
4801: /* } */
4802: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4803: /* delts=delt; */
4804: /* } */
4805: /* } /\* End loop k *\/ */
4806: /* } */
4807: /* delti[theta]=delts; */
4808: /* return res; */
4809: /* } */
4810:
4811:
1.126 brouard 4812: /************** Inverse of matrix **************/
4813: void ludcmp(double **a, int n, int *indx, double *d)
4814: {
4815: int i,imax,j,k;
4816: double big,dum,sum,temp;
4817: double *vv;
4818:
4819: vv=vector(1,n);
4820: *d=1.0;
4821: for (i=1;i<=n;i++) {
4822: big=0.0;
4823: for (j=1;j<=n;j++)
4824: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4825: if (big == 0.0){
4826: printf(" Singular Hessian matrix at row %d:\n",i);
4827: for (j=1;j<=n;j++) {
4828: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4829: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4830: }
4831: fflush(ficlog);
4832: fclose(ficlog);
4833: nrerror("Singular matrix in routine ludcmp");
4834: }
1.126 brouard 4835: vv[i]=1.0/big;
4836: }
4837: for (j=1;j<=n;j++) {
4838: for (i=1;i<j;i++) {
4839: sum=a[i][j];
4840: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4841: a[i][j]=sum;
4842: }
4843: big=0.0;
4844: for (i=j;i<=n;i++) {
4845: sum=a[i][j];
4846: for (k=1;k<j;k++)
4847: sum -= a[i][k]*a[k][j];
4848: a[i][j]=sum;
4849: if ( (dum=vv[i]*fabs(sum)) >= big) {
4850: big=dum;
4851: imax=i;
4852: }
4853: }
4854: if (j != imax) {
4855: for (k=1;k<=n;k++) {
4856: dum=a[imax][k];
4857: a[imax][k]=a[j][k];
4858: a[j][k]=dum;
4859: }
4860: *d = -(*d);
4861: vv[imax]=vv[j];
4862: }
4863: indx[j]=imax;
4864: if (a[j][j] == 0.0) a[j][j]=TINY;
4865: if (j != n) {
4866: dum=1.0/(a[j][j]);
4867: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4868: }
4869: }
4870: free_vector(vv,1,n); /* Doesn't work */
4871: ;
4872: }
4873:
4874: void lubksb(double **a, int n, int *indx, double b[])
4875: {
4876: int i,ii=0,ip,j;
4877: double sum;
4878:
4879: for (i=1;i<=n;i++) {
4880: ip=indx[i];
4881: sum=b[ip];
4882: b[ip]=b[i];
4883: if (ii)
4884: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4885: else if (sum) ii=i;
4886: b[i]=sum;
4887: }
4888: for (i=n;i>=1;i--) {
4889: sum=b[i];
4890: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4891: b[i]=sum/a[i][i];
4892: }
4893: }
4894:
4895: void pstamp(FILE *fichier)
4896: {
1.196 brouard 4897: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4898: }
4899:
1.297 brouard 4900: void date2dmy(double date,double *day, double *month, double *year){
4901: double yp=0., yp1=0., yp2=0.;
4902:
4903: yp1=modf(date,&yp);/* extracts integral of date in yp and
4904: fractional in yp1 */
4905: *year=yp;
4906: yp2=modf((yp1*12),&yp);
4907: *month=yp;
4908: yp1=modf((yp2*30.5),&yp);
4909: *day=yp;
4910: if(*day==0) *day=1;
4911: if(*month==0) *month=1;
4912: }
4913:
1.253 brouard 4914:
4915:
1.126 brouard 4916: /************ Frequencies ********************/
1.251 brouard 4917: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4918: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4919: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4920: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 4921: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 4922: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4923: int iind=0, iage=0;
4924: int mi; /* Effective wave */
4925: int first;
4926: double ***freq; /* Frequencies */
1.268 brouard 4927: 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 */
4928: 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 4929: double *meanq, *stdq, *idq;
1.226 brouard 4930: double **meanqt;
4931: double *pp, **prop, *posprop, *pospropt;
4932: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4933: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4934: double agebegin, ageend;
4935:
4936: pp=vector(1,nlstate);
1.251 brouard 4937: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4938: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4939: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4940: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4941: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4942: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4943: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4944: meanqt=matrix(1,lastpass,1,nqtveff);
4945: strcpy(fileresp,"P_");
4946: strcat(fileresp,fileresu);
4947: /*strcat(fileresphtm,fileresu);*/
4948: if((ficresp=fopen(fileresp,"w"))==NULL) {
4949: printf("Problem with prevalence resultfile: %s\n", fileresp);
4950: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4951: exit(0);
4952: }
1.240 brouard 4953:
1.226 brouard 4954: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4955: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4956: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4957: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4958: fflush(ficlog);
4959: exit(70);
4960: }
4961: else{
4962: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4963: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4964: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4965: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4966: }
1.319 brouard 4967: 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 4968:
1.226 brouard 4969: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4970: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4971: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4972: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4973: fflush(ficlog);
4974: exit(70);
1.240 brouard 4975: } else{
1.226 brouard 4976: 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 4977: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4978: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4979: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4980: }
1.319 brouard 4981: 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 4982:
1.253 brouard 4983: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4984: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4985: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4986: j1=0;
1.126 brouard 4987:
1.227 brouard 4988: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 ! brouard 4989: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 4990: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 4991: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4992:
4993:
1.226 brouard 4994: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4995: reference=low_education V1=0,V2=0
4996: med_educ V1=1 V2=0,
4997: high_educ V1=0 V2=1
1.330 brouard 4998: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 4999: */
1.249 brouard 5000: dateintsum=0;
5001: k2cpt=0;
5002:
1.253 brouard 5003: if(cptcoveff == 0 )
1.265 brouard 5004: nl=1; /* Constant and age model only */
1.253 brouard 5005: else
5006: nl=2;
1.265 brouard 5007:
5008: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
5009: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 ! brouard 5010: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 5011: * freq[s1][s2][iage] =0.
5012: * Loop on iind
5013: * ++freq[s1][s2][iage] weighted
5014: * end iind
5015: * if covariate and j!0
5016: * headers Variable on one line
5017: * endif cov j!=0
5018: * header of frequency table by age
5019: * Loop on age
5020: * pp[s1]+=freq[s1][s2][iage] weighted
5021: * pos+=freq[s1][s2][iage] weighted
5022: * Loop on s1 initial state
5023: * fprintf(ficresp
5024: * end s1
5025: * end age
5026: * if j!=0 computes starting values
5027: * end compute starting values
5028: * end j1
5029: * end nl
5030: */
1.253 brouard 5031: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
5032: if(nj==1)
5033: j=0; /* First pass for the constant */
1.265 brouard 5034: else{
1.335 ! brouard 5035: 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 5036: }
1.251 brouard 5037: first=1;
1.332 brouard 5038: 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 5039: posproptt=0.;
1.330 brouard 5040: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 5041: scanf("%d", i);*/
5042: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 5043: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 5044: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 5045: freq[i][s2][m]=0;
1.251 brouard 5046:
5047: for (i=1; i<=nlstate; i++) {
1.240 brouard 5048: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 5049: prop[i][m]=0;
5050: posprop[i]=0;
5051: pospropt[i]=0;
5052: }
1.283 brouard 5053: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 5054: idq[z1]=0.;
5055: meanq[z1]=0.;
5056: stdq[z1]=0.;
1.283 brouard 5057: }
5058: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 5059: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 5060: /* meanqt[m][z1]=0.; */
5061: /* } */
5062: /* } */
1.251 brouard 5063: /* dateintsum=0; */
5064: /* k2cpt=0; */
5065:
1.265 brouard 5066: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 5067: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
5068: bool=1;
5069: if(j !=0){
5070: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 ! brouard 5071: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
! 5072: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 5073: /* if(Tvaraff[z1] ==-20){ */
5074: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
5075: /* }else if(Tvaraff[z1] ==-10){ */
5076: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 5077: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 ! brouard 5078: /* 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); */
! 5079: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
! 5080: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 5081: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 5082: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 5083: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 5084: /* 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", */
5085: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
5086: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 5087: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
5088: } /* Onlyf fixed */
5089: } /* end z1 */
1.335 ! brouard 5090: } /* cptcoveff > 0 */
1.251 brouard 5091: } /* end any */
5092: }/* end j==0 */
1.265 brouard 5093: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 5094: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 5095: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 5096: m=mw[mi][iind];
5097: if(j!=0){
5098: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 ! brouard 5099: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 5100: if( Fixed[Tmodelind[z1]]==1){
5101: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.332 brouard 5102: 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 5103: value is -1, we don't select. It differs from the
5104: constant and age model which counts them. */
5105: bool=0; /* not selected */
5106: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 5107: /* i1=Tvaraff[z1]; */
5108: /* i2=TnsdVar[i1]; */
5109: /* i3=nbcode[i1][i2]; */
5110: /* i4=covar[i1][iind]; */
5111: /* if(i4 != i3){ */
5112: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 5113: bool=0;
5114: }
5115: }
5116: }
5117: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
5118: } /* end j==0 */
5119: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 5120: if(bool==1){ /*Selected */
1.251 brouard 5121: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
5122: and mw[mi+1][iind]. dh depends on stepm. */
5123: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
5124: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
5125: if(m >=firstpass && m <=lastpass){
5126: k2=anint[m][iind]+(mint[m][iind]/12.);
5127: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
5128: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
5129: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
5130: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
5131: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
5132: if (m<lastpass) {
5133: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
5134: /* 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]); */
5135: if(s[m][iind]==-1)
5136: 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.));
5137: 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 5138: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
5139: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 5140: idq[z1]=idq[z1]+weight[iind];
5141: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
5142: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
5143: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 5144: }
1.284 brouard 5145: }
1.251 brouard 5146: /* if((int)agev[m][iind] == 55) */
5147: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5148: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5149: 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 5150: }
1.251 brouard 5151: } /* end if between passes */
5152: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5153: dateintsum=dateintsum+k2; /* on all covariates ?*/
5154: k2cpt++;
5155: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5156: }
1.251 brouard 5157: }else{
5158: bool=1;
5159: }/* end bool 2 */
5160: } /* end m */
1.284 brouard 5161: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5162: /* idq[z1]=idq[z1]+weight[iind]; */
5163: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5164: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5165: /* } */
1.251 brouard 5166: } /* end bool */
5167: } /* end iind = 1 to imx */
1.319 brouard 5168: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5169: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5170:
5171:
5172: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 ! brouard 5173: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5174: pstamp(ficresp);
1.335 ! brouard 5175: if (cptcoveff>0 && j!=0){
1.265 brouard 5176: pstamp(ficresp);
1.251 brouard 5177: printf( "\n#********** Variable ");
5178: fprintf(ficresp, "\n#********** Variable ");
5179: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5180: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5181: fprintf(ficlog, "\n#********** Variable ");
1.330 brouard 5182: for (z1=1; z1<=cptcovs; z1++){
1.251 brouard 5183: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 5184: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5185: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5186: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5187: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5188: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5189: }else{
1.330 brouard 5190: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5191: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5192: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5193: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5194: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5195: }
5196: }
5197: printf( "**********\n#");
5198: fprintf(ficresp, "**********\n#");
5199: fprintf(ficresphtm, "**********</h3>\n");
5200: fprintf(ficresphtmfr, "**********</h3>\n");
5201: fprintf(ficlog, "**********\n");
5202: }
1.284 brouard 5203: /*
5204: Printing means of quantitative variables if any
5205: */
5206: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5207: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5208: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5209: if(weightopt==1){
5210: printf(" Weighted mean and standard deviation of");
5211: fprintf(ficlog," Weighted mean and standard deviation of");
5212: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5213: }
1.311 brouard 5214: /* mu = \frac{w x}{\sum w}
5215: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5216: */
5217: 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]));
5218: 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]));
5219: 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 5220: }
5221: /* for (z1=1; z1<= nqtveff; z1++) { */
5222: /* for(m=1;m<=lastpass;m++){ */
5223: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5224: /* } */
5225: /* } */
1.283 brouard 5226:
1.251 brouard 5227: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 ! brouard 5228: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5229: fprintf(ficresp, " Age");
1.335 ! brouard 5230: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
! 5231: 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]]);
! 5232: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
! 5233: }
1.251 brouard 5234: for(i=1; i<=nlstate;i++) {
1.335 ! brouard 5235: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5236: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5237: }
1.335 ! brouard 5238: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5239: fprintf(ficresphtm, "\n");
5240:
5241: /* Header of frequency table by age */
5242: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5243: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5244: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5245: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5246: if(s2!=0 && m!=0)
5247: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5248: }
1.226 brouard 5249: }
1.251 brouard 5250: fprintf(ficresphtmfr, "\n");
5251:
5252: /* For each age */
5253: for(iage=iagemin; iage <= iagemax+3; iage++){
5254: fprintf(ficresphtm,"<tr>");
5255: if(iage==iagemax+1){
5256: fprintf(ficlog,"1");
5257: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5258: }else if(iage==iagemax+2){
5259: fprintf(ficlog,"0");
5260: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5261: }else if(iage==iagemax+3){
5262: fprintf(ficlog,"Total");
5263: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5264: }else{
1.240 brouard 5265: if(first==1){
1.251 brouard 5266: first=0;
5267: printf("See log file for details...\n");
5268: }
5269: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5270: fprintf(ficlog,"Age %d", iage);
5271: }
1.265 brouard 5272: for(s1=1; s1 <=nlstate ; s1++){
5273: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5274: pp[s1] += freq[s1][m][iage];
1.251 brouard 5275: }
1.265 brouard 5276: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5277: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5278: pos += freq[s1][m][iage];
5279: if(pp[s1]>=1.e-10){
1.251 brouard 5280: if(first==1){
1.265 brouard 5281: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5282: }
1.265 brouard 5283: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5284: }else{
5285: if(first==1)
1.265 brouard 5286: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5287: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5288: }
5289: }
5290:
1.265 brouard 5291: for(s1=1; s1 <=nlstate ; s1++){
5292: /* posprop[s1]=0; */
5293: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5294: pp[s1] += freq[s1][m][iage];
5295: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5296:
5297: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5298: pos += pp[s1]; /* pos is the total number of transitions until this age */
5299: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5300: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5301: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5302: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5303: }
5304:
5305: /* Writing ficresp */
1.335 ! brouard 5306: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5307: if( iage <= iagemax){
5308: fprintf(ficresp," %d",iage);
5309: }
5310: }else if( nj==2){
5311: if( iage <= iagemax){
5312: fprintf(ficresp," %d",iage);
1.335 ! brouard 5313: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 5314: }
1.240 brouard 5315: }
1.265 brouard 5316: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5317: if(pos>=1.e-5){
1.251 brouard 5318: if(first==1)
1.265 brouard 5319: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5320: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5321: }else{
5322: if(first==1)
1.265 brouard 5323: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5324: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5325: }
5326: if( iage <= iagemax){
5327: if(pos>=1.e-5){
1.335 ! brouard 5328: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5329: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5330: }else if( nj==2){
5331: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5332: }
5333: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5334: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5335: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5336: } else{
1.335 ! brouard 5337: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5338: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5339: }
1.240 brouard 5340: }
1.265 brouard 5341: pospropt[s1] +=posprop[s1];
5342: } /* end loop s1 */
1.251 brouard 5343: /* pospropt=0.; */
1.265 brouard 5344: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5345: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5346: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5347: if(first==1){
1.265 brouard 5348: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5349: }
1.265 brouard 5350: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5351: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5352: }
1.265 brouard 5353: if(s1!=0 && m!=0)
5354: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5355: }
1.265 brouard 5356: } /* end loop s1 */
1.251 brouard 5357: posproptt=0.;
1.265 brouard 5358: for(s1=1; s1 <=nlstate; s1++){
5359: posproptt += pospropt[s1];
1.251 brouard 5360: }
5361: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5362: fprintf(ficresphtm,"</tr>\n");
1.335 ! brouard 5363: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 5364: if(iage <= iagemax)
5365: fprintf(ficresp,"\n");
1.240 brouard 5366: }
1.251 brouard 5367: if(first==1)
5368: printf("Others in log...\n");
5369: fprintf(ficlog,"\n");
5370: } /* end loop age iage */
1.265 brouard 5371:
1.251 brouard 5372: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5373: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5374: if(posproptt < 1.e-5){
1.265 brouard 5375: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5376: }else{
1.265 brouard 5377: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5378: }
1.226 brouard 5379: }
1.251 brouard 5380: fprintf(ficresphtm,"</tr>\n");
5381: fprintf(ficresphtm,"</table>\n");
5382: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5383: if(posproptt < 1.e-5){
1.251 brouard 5384: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5385: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5386: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5387: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5388: invalidvarcomb[j1]=1;
1.226 brouard 5389: }else{
1.251 brouard 5390: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
5391: invalidvarcomb[j1]=0;
1.226 brouard 5392: }
1.251 brouard 5393: fprintf(ficresphtmfr,"</table>\n");
5394: fprintf(ficlog,"\n");
5395: if(j!=0){
5396: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5397: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5398: for(k=1; k <=(nlstate+ndeath); k++){
5399: if (k != i) {
1.265 brouard 5400: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5401: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5402: if(j1==1){ /* All dummy covariates to zero */
5403: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5404: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5405: printf("%d%d ",i,k);
5406: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5407: 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]));
5408: 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]));
5409: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5410: }
1.253 brouard 5411: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5412: for(iage=iagemin; iage <= iagemax+3; iage++){
5413: x[iage]= (double)iage;
5414: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5415: /* 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 5416: }
1.268 brouard 5417: /* Some are not finite, but linreg will ignore these ages */
5418: no=0;
1.253 brouard 5419: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5420: pstart[s1]=b;
5421: pstart[s1-1]=a;
1.252 brouard 5422: }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 */
5423: 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]);
5424: 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 5425: 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 5426: printf("%d%d ",i,k);
5427: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5428: 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 5429: }else{ /* Other cases, like quantitative fixed or varying covariates */
5430: ;
5431: }
5432: /* printf("%12.7f )", param[i][jj][k]); */
5433: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5434: s1++;
1.251 brouard 5435: } /* end jj */
5436: } /* end k!= i */
5437: } /* end k */
1.265 brouard 5438: } /* end i, s1 */
1.251 brouard 5439: } /* end j !=0 */
5440: } /* end selected combination of covariate j1 */
5441: if(j==0){ /* We can estimate starting values from the occurences in each case */
5442: printf("#Freqsummary: Starting values for the constants:\n");
5443: fprintf(ficlog,"\n");
1.265 brouard 5444: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5445: for(k=1; k <=(nlstate+ndeath); k++){
5446: if (k != i) {
5447: printf("%d%d ",i,k);
5448: fprintf(ficlog,"%d%d ",i,k);
5449: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5450: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5451: if(jj==1){ /* Age has to be done */
1.265 brouard 5452: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5453: 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]));
5454: 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 5455: }
5456: /* printf("%12.7f )", param[i][jj][k]); */
5457: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5458: s1++;
1.250 brouard 5459: }
1.251 brouard 5460: printf("\n");
5461: fprintf(ficlog,"\n");
1.250 brouard 5462: }
5463: }
1.284 brouard 5464: } /* end of state i */
1.251 brouard 5465: printf("#Freqsummary\n");
5466: fprintf(ficlog,"\n");
1.265 brouard 5467: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5468: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5469: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5470: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5471: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5472: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5473: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5474: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5475: /* } */
5476: }
1.265 brouard 5477: } /* end loop s1 */
1.251 brouard 5478:
5479: printf("\n");
5480: fprintf(ficlog,"\n");
5481: } /* end j=0 */
1.249 brouard 5482: } /* end j */
1.252 brouard 5483:
1.253 brouard 5484: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5485: for(i=1, jk=1; i <=nlstate; i++){
5486: for(j=1; j <=nlstate+ndeath; j++){
5487: if(j!=i){
5488: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5489: printf("%1d%1d",i,j);
5490: fprintf(ficparo,"%1d%1d",i,j);
5491: for(k=1; k<=ncovmodel;k++){
5492: /* printf(" %lf",param[i][j][k]); */
5493: /* fprintf(ficparo," %lf",param[i][j][k]); */
5494: p[jk]=pstart[jk];
5495: printf(" %f ",pstart[jk]);
5496: fprintf(ficparo," %f ",pstart[jk]);
5497: jk++;
5498: }
5499: printf("\n");
5500: fprintf(ficparo,"\n");
5501: }
5502: }
5503: }
5504: } /* end mle=-2 */
1.226 brouard 5505: dateintmean=dateintsum/k2cpt;
1.296 brouard 5506: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5507:
1.226 brouard 5508: fclose(ficresp);
5509: fclose(ficresphtm);
5510: fclose(ficresphtmfr);
1.283 brouard 5511: free_vector(idq,1,nqfveff);
1.226 brouard 5512: free_vector(meanq,1,nqfveff);
1.284 brouard 5513: free_vector(stdq,1,nqfveff);
1.226 brouard 5514: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5515: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5516: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5517: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5518: free_vector(pospropt,1,nlstate);
5519: free_vector(posprop,1,nlstate);
1.251 brouard 5520: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5521: free_vector(pp,1,nlstate);
5522: /* End of freqsummary */
5523: }
1.126 brouard 5524:
1.268 brouard 5525: /* Simple linear regression */
5526: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5527:
5528: /* y=a+bx regression */
5529: double sumx = 0.0; /* sum of x */
5530: double sumx2 = 0.0; /* sum of x**2 */
5531: double sumxy = 0.0; /* sum of x * y */
5532: double sumy = 0.0; /* sum of y */
5533: double sumy2 = 0.0; /* sum of y**2 */
5534: double sume2 = 0.0; /* sum of square or residuals */
5535: double yhat;
5536:
5537: double denom=0;
5538: int i;
5539: int ne=*no;
5540:
5541: for ( i=ifi, ne=0;i<=ila;i++) {
5542: if(!isfinite(x[i]) || !isfinite(y[i])){
5543: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5544: continue;
5545: }
5546: ne=ne+1;
5547: sumx += x[i];
5548: sumx2 += x[i]*x[i];
5549: sumxy += x[i] * y[i];
5550: sumy += y[i];
5551: sumy2 += y[i]*y[i];
5552: denom = (ne * sumx2 - sumx*sumx);
5553: /* 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); */
5554: }
5555:
5556: denom = (ne * sumx2 - sumx*sumx);
5557: if (denom == 0) {
5558: // vertical, slope m is infinity
5559: *b = INFINITY;
5560: *a = 0;
5561: if (r) *r = 0;
5562: return 1;
5563: }
5564:
5565: *b = (ne * sumxy - sumx * sumy) / denom;
5566: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5567: if (r!=NULL) {
5568: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5569: sqrt((sumx2 - sumx*sumx/ne) *
5570: (sumy2 - sumy*sumy/ne));
5571: }
5572: *no=ne;
5573: for ( i=ifi, ne=0;i<=ila;i++) {
5574: if(!isfinite(x[i]) || !isfinite(y[i])){
5575: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5576: continue;
5577: }
5578: ne=ne+1;
5579: yhat = y[i] - *a -*b* x[i];
5580: sume2 += yhat * yhat ;
5581:
5582: denom = (ne * sumx2 - sumx*sumx);
5583: /* 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); */
5584: }
5585: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5586: *sa= *sb * sqrt(sumx2/ne);
5587:
5588: return 0;
5589: }
5590:
1.126 brouard 5591: /************ Prevalence ********************/
1.227 brouard 5592: 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)
5593: {
5594: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5595: in each health status at the date of interview (if between dateprev1 and dateprev2).
5596: We still use firstpass and lastpass as another selection.
5597: */
1.126 brouard 5598:
1.227 brouard 5599: int i, m, jk, j1, bool, z1,j, iv;
5600: int mi; /* Effective wave */
5601: int iage;
5602: double agebegin, ageend;
5603:
5604: double **prop;
5605: double posprop;
5606: double y2; /* in fractional years */
5607: int iagemin, iagemax;
5608: int first; /** to stop verbosity which is redirected to log file */
5609:
5610: iagemin= (int) agemin;
5611: iagemax= (int) agemax;
5612: /*pp=vector(1,nlstate);*/
1.251 brouard 5613: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5614: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5615: j1=0;
1.222 brouard 5616:
1.227 brouard 5617: /*j=cptcoveff;*/
5618: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5619:
1.288 brouard 5620: first=0;
1.335 ! brouard 5621: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 5622: for (i=1; i<=nlstate; i++)
1.251 brouard 5623: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5624: prop[i][iage]=0.0;
5625: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5626: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5627: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5628:
5629: for (i=1; i<=imx; i++) { /* Each individual */
5630: bool=1;
5631: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5632: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5633: m=mw[mi][i];
5634: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5635: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5636: for (z1=1; z1<=cptcoveff; z1++){
5637: if( Fixed[Tmodelind[z1]]==1){
5638: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.332 brouard 5639: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 5640: bool=0;
5641: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 5642: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 5643: bool=0;
5644: }
5645: }
5646: if(bool==1){ /* Otherwise we skip that wave/person */
5647: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5648: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5649: if(m >=firstpass && m <=lastpass){
5650: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5651: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5652: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5653: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5654: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5655: 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);
5656: exit(1);
5657: }
5658: if (s[m][i]>0 && s[m][i]<=nlstate) {
5659: /*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]]);*/
5660: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5661: prop[s[m][i]][iagemax+3] += weight[i];
5662: } /* end valid statuses */
5663: } /* end selection of dates */
5664: } /* end selection of waves */
5665: } /* end bool */
5666: } /* end wave */
5667: } /* end individual */
5668: for(i=iagemin; i <= iagemax+3; i++){
5669: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5670: posprop += prop[jk][i];
5671: }
5672:
5673: for(jk=1; jk <=nlstate ; jk++){
5674: if( i <= iagemax){
5675: if(posprop>=1.e-5){
5676: probs[i][jk][j1]= prop[jk][i]/posprop;
5677: } else{
1.288 brouard 5678: if(!first){
5679: first=1;
1.266 brouard 5680: 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]);
5681: }else{
1.288 brouard 5682: 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 5683: }
5684: }
5685: }
5686: }/* end jk */
5687: }/* end i */
1.222 brouard 5688: /*} *//* end i1 */
1.227 brouard 5689: } /* end j1 */
1.222 brouard 5690:
1.227 brouard 5691: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5692: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5693: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5694: } /* End of prevalence */
1.126 brouard 5695:
5696: /************* Waves Concatenation ***************/
5697:
5698: 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)
5699: {
1.298 brouard 5700: /* 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 5701: Death is a valid wave (if date is known).
5702: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5703: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5704: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5705: */
1.126 brouard 5706:
1.224 brouard 5707: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5708: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5709: double sum=0., jmean=0.;*/
1.224 brouard 5710: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5711: int j, k=0,jk, ju, jl;
5712: double sum=0.;
5713: first=0;
1.214 brouard 5714: firstwo=0;
1.217 brouard 5715: firsthree=0;
1.218 brouard 5716: firstfour=0;
1.164 brouard 5717: jmin=100000;
1.126 brouard 5718: jmax=-1;
5719: jmean=0.;
1.224 brouard 5720:
5721: /* Treating live states */
1.214 brouard 5722: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5723: mi=0; /* First valid wave */
1.227 brouard 5724: mli=0; /* Last valid wave */
1.309 brouard 5725: m=firstpass; /* Loop on waves */
5726: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5727: 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 */
5728: mli=m-1;/* mw[++mi][i]=m-1; */
5729: }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 5730: 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 5731: mli=m;
1.224 brouard 5732: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5733: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5734: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5735: }
1.309 brouard 5736: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5737: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5738: break;
1.224 brouard 5739: #else
1.317 brouard 5740: 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 5741: if(firsthree == 0){
1.302 brouard 5742: 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 5743: firsthree=1;
1.317 brouard 5744: }else if(firsthree >=1 && firsthree < 10){
5745: 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);
5746: firsthree++;
5747: }else if(firsthree == 10){
5748: printf("Information, too many Information flags: no more reported to log either\n");
5749: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5750: firsthree++;
5751: }else{
5752: firsthree++;
1.227 brouard 5753: }
1.309 brouard 5754: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5755: mli=m;
5756: }
5757: if(s[m][i]==-2){ /* Vital status is really unknown */
5758: nbwarn++;
1.309 brouard 5759: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5760: 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);
5761: 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);
5762: }
5763: break;
5764: }
5765: break;
1.224 brouard 5766: #endif
1.227 brouard 5767: }/* End m >= lastpass */
1.126 brouard 5768: }/* end while */
1.224 brouard 5769:
1.227 brouard 5770: /* 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 5771: /* After last pass */
1.224 brouard 5772: /* Treating death states */
1.214 brouard 5773: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5774: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5775: /* } */
1.126 brouard 5776: mi++; /* Death is another wave */
5777: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5778: /* Only death is a correct wave */
1.126 brouard 5779: mw[mi][i]=m;
1.257 brouard 5780: } /* else not in a death state */
1.224 brouard 5781: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5782: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5783: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5784: 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 5785: nbwarn++;
5786: if(firstfiv==0){
1.309 brouard 5787: 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 5788: firstfiv=1;
5789: }else{
1.309 brouard 5790: 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 5791: }
1.309 brouard 5792: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5793: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5794: nberr++;
5795: if(firstwo==0){
1.309 brouard 5796: 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 5797: firstwo=1;
5798: }
1.309 brouard 5799: 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 5800: }
1.257 brouard 5801: }else{ /* if date of interview is unknown */
1.227 brouard 5802: /* death is known but not confirmed by death status at any wave */
5803: if(firstfour==0){
1.309 brouard 5804: 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 5805: firstfour=1;
5806: }
1.309 brouard 5807: 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 5808: }
1.224 brouard 5809: } /* end if date of death is known */
5810: #endif
1.309 brouard 5811: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5812: /* wav[i]=mw[mi][i]; */
1.126 brouard 5813: if(mi==0){
5814: nbwarn++;
5815: if(first==0){
1.227 brouard 5816: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5817: first=1;
1.126 brouard 5818: }
5819: if(first==1){
1.227 brouard 5820: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5821: }
5822: } /* end mi==0 */
5823: } /* End individuals */
1.214 brouard 5824: /* wav and mw are no more changed */
1.223 brouard 5825:
1.317 brouard 5826: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
5827: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
5828:
5829:
1.126 brouard 5830: for(i=1; i<=imx; i++){
5831: for(mi=1; mi<wav[i];mi++){
5832: if (stepm <=0)
1.227 brouard 5833: dh[mi][i]=1;
1.126 brouard 5834: else{
1.260 brouard 5835: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5836: if (agedc[i] < 2*AGESUP) {
5837: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5838: if(j==0) j=1; /* Survives at least one month after exam */
5839: else if(j<0){
5840: nberr++;
5841: 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]);
5842: j=1; /* Temporary Dangerous patch */
5843: 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);
5844: 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]);
5845: 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);
5846: }
5847: k=k+1;
5848: if (j >= jmax){
5849: jmax=j;
5850: ijmax=i;
5851: }
5852: if (j <= jmin){
5853: jmin=j;
5854: ijmin=i;
5855: }
5856: sum=sum+j;
5857: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5858: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5859: }
5860: }
5861: else{
5862: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5863: /* 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 5864:
1.227 brouard 5865: k=k+1;
5866: if (j >= jmax) {
5867: jmax=j;
5868: ijmax=i;
5869: }
5870: else if (j <= jmin){
5871: jmin=j;
5872: ijmin=i;
5873: }
5874: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5875: /*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]);*/
5876: if(j<0){
5877: nberr++;
5878: 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]);
5879: 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]);
5880: }
5881: sum=sum+j;
5882: }
5883: jk= j/stepm;
5884: jl= j -jk*stepm;
5885: ju= j -(jk+1)*stepm;
5886: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5887: if(jl==0){
5888: dh[mi][i]=jk;
5889: bh[mi][i]=0;
5890: }else{ /* We want a negative bias in order to only have interpolation ie
5891: * to avoid the price of an extra matrix product in likelihood */
5892: dh[mi][i]=jk+1;
5893: bh[mi][i]=ju;
5894: }
5895: }else{
5896: if(jl <= -ju){
5897: dh[mi][i]=jk;
5898: bh[mi][i]=jl; /* bias is positive if real duration
5899: * is higher than the multiple of stepm and negative otherwise.
5900: */
5901: }
5902: else{
5903: dh[mi][i]=jk+1;
5904: bh[mi][i]=ju;
5905: }
5906: if(dh[mi][i]==0){
5907: dh[mi][i]=1; /* At least one step */
5908: bh[mi][i]=ju; /* At least one step */
5909: /* 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);*/
5910: }
5911: } /* end if mle */
1.126 brouard 5912: }
5913: } /* end wave */
5914: }
5915: jmean=sum/k;
5916: 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 5917: 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 5918: }
1.126 brouard 5919:
5920: /*********** Tricode ****************************/
1.220 brouard 5921: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5922: {
5923: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5924: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5925: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5926: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5927: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5928: */
1.130 brouard 5929:
1.242 brouard 5930: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5931: int modmaxcovj=0; /* Modality max of covariates j */
5932: int cptcode=0; /* Modality max of covariates j */
5933: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5934:
5935:
1.242 brouard 5936: /* cptcoveff=0; */
5937: /* *cptcov=0; */
1.126 brouard 5938:
1.242 brouard 5939: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5940: for (k=1; k <= maxncov; k++)
5941: for(j=1; j<=2; j++)
5942: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5943:
1.242 brouard 5944: /* Loop on covariates without age and products and no quantitative variable */
1.335 ! brouard 5945: 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 5946: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5947: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5948: switch(Fixed[k]) {
5949: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5950: modmaxcovj=0;
5951: modmincovj=0;
1.242 brouard 5952: 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*/
5953: ij=(int)(covar[Tvar[k]][i]);
5954: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5955: * If product of Vn*Vm, still boolean *:
5956: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5957: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5958: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5959: modality of the nth covariate of individual i. */
5960: if (ij > modmaxcovj)
5961: modmaxcovj=ij;
5962: else if (ij < modmincovj)
5963: modmincovj=ij;
1.287 brouard 5964: if (ij <0 || ij >1 ){
1.311 brouard 5965: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5966: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5967: fflush(ficlog);
5968: exit(1);
1.287 brouard 5969: }
5970: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5971: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5972: exit(1);
5973: }else
5974: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5975: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5976: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5977: /* getting the maximum value of the modality of the covariate
5978: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5979: female ies 1, then modmaxcovj=1.
5980: */
5981: } /* end for loop on individuals i */
5982: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5983: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5984: cptcode=modmaxcovj;
5985: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5986: /*for (i=0; i<=cptcode; i++) {*/
5987: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5988: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5989: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5990: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5991: if( j != -1){
5992: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5993: covariate for which somebody answered excluding
5994: undefined. Usually 2: 0 and 1. */
5995: }
5996: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5997: covariate for which somebody answered including
5998: undefined. Usually 3: -1, 0 and 1. */
5999: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
6000: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
6001: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 6002:
1.242 brouard 6003: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
6004: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
6005: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
6006: /* modmincovj=3; modmaxcovj = 7; */
6007: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
6008: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
6009: /* defining two dummy variables: variables V1_1 and V1_2.*/
6010: /* nbcode[Tvar[j]][ij]=k; */
6011: /* nbcode[Tvar[j]][1]=0; */
6012: /* nbcode[Tvar[j]][2]=1; */
6013: /* nbcode[Tvar[j]][3]=2; */
6014: /* To be continued (not working yet). */
6015: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 6016:
6017: /* 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*/
6018: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
6019: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
6020: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
6021: /*, could be restored in the future */
6022: 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 6023: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
6024: break;
6025: }
6026: ij++;
1.287 brouard 6027: 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 6028: cptcode = ij; /* New max modality for covar j */
6029: } /* end of loop on modality i=-1 to 1 or more */
6030: break;
6031: case 1: /* Testing on varying covariate, could be simple and
6032: * should look at waves or product of fixed *
6033: * varying. No time to test -1, assuming 0 and 1 only */
6034: ij=0;
6035: for(i=0; i<=1;i++){
6036: nbcode[Tvar[k]][++ij]=i;
6037: }
6038: break;
6039: default:
6040: break;
6041: } /* end switch */
6042: } /* end dummy test */
1.334 brouard 6043: if(Dummy[k]==1 && Typevar[k] !=1){ /* Quantitative covariate and not age product */
1.311 brouard 6044: 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 6045: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
! 6046: printf("Error k=%d \n",k);
! 6047: exit(1);
! 6048: }
1.311 brouard 6049: if(isnan(covar[Tvar[k]][i])){
6050: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6051: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6052: fflush(ficlog);
6053: exit(1);
6054: }
6055: }
1.335 ! brouard 6056: } /* end Quanti */
1.287 brouard 6057: } /* 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 6058:
6059: for (k=-1; k< maxncov; k++) Ndum[k]=0;
6060: /* Look at fixed dummy (single or product) covariates to check empty modalities */
6061: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
6062: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
6063: 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 */
6064: 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 */
6065: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
6066: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
6067:
6068: ij=0;
6069: /* 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 6070: 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 */
! 6071: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 6072: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
6073: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 ! brouard 6074: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
! 6075: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
! 6076: /* 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 6077: /* If product not in single variable we don't print results */
6078: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 ! brouard 6079: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
! 6080: /* k= 1 2 3 4 5 6 7 8 9 */
! 6081: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
! 6082: /* ij 1 2 3 */
! 6083: /* Tvaraff[ij]= 4 3 1 */
! 6084: /* Tmodelind[ij]=2 3 9 */
! 6085: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 6086: 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*/
6087: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
6088: 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 */
6089: if(Fixed[k]!=0)
6090: anyvaryingduminmodel=1;
6091: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
6092: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
6093: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
6094: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
6095: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
6096: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
6097: }
6098: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
6099: /* ij--; */
6100: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 ! brouard 6101: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 6102: * because they can be excluded from the model and real
6103: * if in the model but excluded because missing values, but how to get k from ij?*/
6104: for(j=ij+1; j<= cptcovt; j++){
6105: Tvaraff[j]=0;
6106: Tmodelind[j]=0;
6107: }
6108: for(j=ntveff+1; j<= cptcovt; j++){
6109: TmodelInvind[j]=0;
6110: }
6111: /* To be sorted */
6112: ;
6113: }
1.126 brouard 6114:
1.145 brouard 6115:
1.126 brouard 6116: /*********** Health Expectancies ****************/
6117:
1.235 brouard 6118: 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 6119:
6120: {
6121: /* Health expectancies, no variances */
1.329 brouard 6122: /* cij is the combination in the list of combination of dummy covariates */
6123: /* strstart is a string of time at start of computing */
1.164 brouard 6124: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 6125: int nhstepma, nstepma; /* Decreasing with age */
6126: double age, agelim, hf;
6127: double ***p3mat;
6128: double eip;
6129:
1.238 brouard 6130: /* pstamp(ficreseij); */
1.126 brouard 6131: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
6132: fprintf(ficreseij,"# Age");
6133: for(i=1; i<=nlstate;i++){
6134: for(j=1; j<=nlstate;j++){
6135: fprintf(ficreseij," e%1d%1d ",i,j);
6136: }
6137: fprintf(ficreseij," e%1d. ",i);
6138: }
6139: fprintf(ficreseij,"\n");
6140:
6141:
6142: if(estepm < stepm){
6143: printf ("Problem %d lower than %d\n",estepm, stepm);
6144: }
6145: else hstepm=estepm;
6146: /* We compute the life expectancy from trapezoids spaced every estepm months
6147: * This is mainly to measure the difference between two models: for example
6148: * if stepm=24 months pijx are given only every 2 years and by summing them
6149: * we are calculating an estimate of the Life Expectancy assuming a linear
6150: * progression in between and thus overestimating or underestimating according
6151: * to the curvature of the survival function. If, for the same date, we
6152: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6153: * to compare the new estimate of Life expectancy with the same linear
6154: * hypothesis. A more precise result, taking into account a more precise
6155: * curvature will be obtained if estepm is as small as stepm. */
6156:
6157: /* For example we decided to compute the life expectancy with the smallest unit */
6158: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6159: nhstepm is the number of hstepm from age to agelim
6160: nstepm is the number of stepm from age to agelin.
1.270 brouard 6161: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6162: and note for a fixed period like estepm months */
6163: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6164: survival function given by stepm (the optimization length). Unfortunately it
6165: means that if the survival funtion is printed only each two years of age and if
6166: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6167: results. So we changed our mind and took the option of the best precision.
6168: */
6169: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6170:
6171: agelim=AGESUP;
6172: /* If stepm=6 months */
6173: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6174: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6175:
6176: /* nhstepm age range expressed in number of stepm */
6177: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6178: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6179: /* if (stepm >= YEARM) hstepm=1;*/
6180: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6181: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6182:
6183: for (age=bage; age<=fage; age ++){
6184: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6185: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6186: /* if (stepm >= YEARM) hstepm=1;*/
6187: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6188:
6189: /* If stepm=6 months */
6190: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6191: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 6192: /* 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 6193: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6194:
6195: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6196:
6197: printf("%d|",(int)age);fflush(stdout);
6198: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6199:
6200: /* Computing expectancies */
6201: for(i=1; i<=nlstate;i++)
6202: for(j=1; j<=nlstate;j++)
6203: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6204: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6205:
6206: /* 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]);*/
6207:
6208: }
6209:
6210: fprintf(ficreseij,"%3.0f",age );
6211: for(i=1; i<=nlstate;i++){
6212: eip=0;
6213: for(j=1; j<=nlstate;j++){
6214: eip +=eij[i][j][(int)age];
6215: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6216: }
6217: fprintf(ficreseij,"%9.4f", eip );
6218: }
6219: fprintf(ficreseij,"\n");
6220:
6221: }
6222: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6223: printf("\n");
6224: fprintf(ficlog,"\n");
6225:
6226: }
6227:
1.235 brouard 6228: 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 6229:
6230: {
6231: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6232: to initial status i, ei. .
1.126 brouard 6233: */
6234: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6235: int nhstepma, nstepma; /* Decreasing with age */
6236: double age, agelim, hf;
6237: double ***p3matp, ***p3matm, ***varhe;
6238: double **dnewm,**doldm;
6239: double *xp, *xm;
6240: double **gp, **gm;
6241: double ***gradg, ***trgradg;
6242: int theta;
6243:
6244: double eip, vip;
6245:
6246: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6247: xp=vector(1,npar);
6248: xm=vector(1,npar);
6249: dnewm=matrix(1,nlstate*nlstate,1,npar);
6250: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6251:
6252: pstamp(ficresstdeij);
6253: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6254: fprintf(ficresstdeij,"# Age");
6255: for(i=1; i<=nlstate;i++){
6256: for(j=1; j<=nlstate;j++)
6257: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6258: fprintf(ficresstdeij," e%1d. ",i);
6259: }
6260: fprintf(ficresstdeij,"\n");
6261:
6262: pstamp(ficrescveij);
6263: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6264: fprintf(ficrescveij,"# Age");
6265: for(i=1; i<=nlstate;i++)
6266: for(j=1; j<=nlstate;j++){
6267: cptj= (j-1)*nlstate+i;
6268: for(i2=1; i2<=nlstate;i2++)
6269: for(j2=1; j2<=nlstate;j2++){
6270: cptj2= (j2-1)*nlstate+i2;
6271: if(cptj2 <= cptj)
6272: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6273: }
6274: }
6275: fprintf(ficrescveij,"\n");
6276:
6277: if(estepm < stepm){
6278: printf ("Problem %d lower than %d\n",estepm, stepm);
6279: }
6280: else hstepm=estepm;
6281: /* We compute the life expectancy from trapezoids spaced every estepm months
6282: * This is mainly to measure the difference between two models: for example
6283: * if stepm=24 months pijx are given only every 2 years and by summing them
6284: * we are calculating an estimate of the Life Expectancy assuming a linear
6285: * progression in between and thus overestimating or underestimating according
6286: * to the curvature of the survival function. If, for the same date, we
6287: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6288: * to compare the new estimate of Life expectancy with the same linear
6289: * hypothesis. A more precise result, taking into account a more precise
6290: * curvature will be obtained if estepm is as small as stepm. */
6291:
6292: /* For example we decided to compute the life expectancy with the smallest unit */
6293: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6294: nhstepm is the number of hstepm from age to agelim
6295: nstepm is the number of stepm from age to agelin.
6296: Look at hpijx to understand the reason of that which relies in memory size
6297: and note for a fixed period like estepm months */
6298: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6299: survival function given by stepm (the optimization length). Unfortunately it
6300: means that if the survival funtion is printed only each two years of age and if
6301: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6302: results. So we changed our mind and took the option of the best precision.
6303: */
6304: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6305:
6306: /* If stepm=6 months */
6307: /* nhstepm age range expressed in number of stepm */
6308: agelim=AGESUP;
6309: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6310: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6311: /* if (stepm >= YEARM) hstepm=1;*/
6312: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6313:
6314: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6315: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6316: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6317: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6318: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6319: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6320:
6321: for (age=bage; age<=fage; age ++){
6322: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6323: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6324: /* if (stepm >= YEARM) hstepm=1;*/
6325: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6326:
1.126 brouard 6327: /* If stepm=6 months */
6328: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6329: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6330:
6331: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6332:
1.126 brouard 6333: /* Computing Variances of health expectancies */
6334: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6335: decrease memory allocation */
6336: for(theta=1; theta <=npar; theta++){
6337: for(i=1; i<=npar; i++){
1.222 brouard 6338: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6339: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6340: }
1.235 brouard 6341: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6342: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6343:
1.126 brouard 6344: for(j=1; j<= nlstate; j++){
1.222 brouard 6345: for(i=1; i<=nlstate; i++){
6346: for(h=0; h<=nhstepm-1; h++){
6347: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6348: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6349: }
6350: }
1.126 brouard 6351: }
1.218 brouard 6352:
1.126 brouard 6353: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6354: for(h=0; h<=nhstepm-1; h++){
6355: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6356: }
1.126 brouard 6357: }/* End theta */
6358:
6359:
6360: for(h=0; h<=nhstepm-1; h++)
6361: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6362: for(theta=1; theta <=npar; theta++)
6363: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6364:
1.218 brouard 6365:
1.222 brouard 6366: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6367: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6368: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6369:
1.222 brouard 6370: printf("%d|",(int)age);fflush(stdout);
6371: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6372: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6373: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6374: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6375: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6376: for(ij=1;ij<=nlstate*nlstate;ij++)
6377: for(ji=1;ji<=nlstate*nlstate;ji++)
6378: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6379: }
6380: }
1.320 brouard 6381: /* if((int)age ==50){ */
6382: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6383: /* } */
1.126 brouard 6384: /* Computing expectancies */
1.235 brouard 6385: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6386: for(i=1; i<=nlstate;i++)
6387: for(j=1; j<=nlstate;j++)
1.222 brouard 6388: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6389: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6390:
1.222 brouard 6391: /* 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 6392:
1.222 brouard 6393: }
1.269 brouard 6394:
6395: /* Standard deviation of expectancies ij */
1.126 brouard 6396: fprintf(ficresstdeij,"%3.0f",age );
6397: for(i=1; i<=nlstate;i++){
6398: eip=0.;
6399: vip=0.;
6400: for(j=1; j<=nlstate;j++){
1.222 brouard 6401: eip += eij[i][j][(int)age];
6402: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6403: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6404: 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 6405: }
6406: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6407: }
6408: fprintf(ficresstdeij,"\n");
1.218 brouard 6409:
1.269 brouard 6410: /* Variance of expectancies ij */
1.126 brouard 6411: fprintf(ficrescveij,"%3.0f",age );
6412: for(i=1; i<=nlstate;i++)
6413: for(j=1; j<=nlstate;j++){
1.222 brouard 6414: cptj= (j-1)*nlstate+i;
6415: for(i2=1; i2<=nlstate;i2++)
6416: for(j2=1; j2<=nlstate;j2++){
6417: cptj2= (j2-1)*nlstate+i2;
6418: if(cptj2 <= cptj)
6419: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6420: }
1.126 brouard 6421: }
6422: fprintf(ficrescveij,"\n");
1.218 brouard 6423:
1.126 brouard 6424: }
6425: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6426: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6427: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6428: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6429: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6430: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6431: printf("\n");
6432: fprintf(ficlog,"\n");
1.218 brouard 6433:
1.126 brouard 6434: free_vector(xm,1,npar);
6435: free_vector(xp,1,npar);
6436: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6437: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6438: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6439: }
1.218 brouard 6440:
1.126 brouard 6441: /************ Variance ******************/
1.235 brouard 6442: 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 6443: {
1.279 brouard 6444: /** Variance of health expectancies
6445: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6446: * double **newm;
6447: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6448: */
1.218 brouard 6449:
6450: /* int movingaverage(); */
6451: double **dnewm,**doldm;
6452: double **dnewmp,**doldmp;
6453: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6454: int first=0;
1.218 brouard 6455: int k;
6456: double *xp;
1.279 brouard 6457: double **gp, **gm; /**< for var eij */
6458: double ***gradg, ***trgradg; /**< for var eij */
6459: double **gradgp, **trgradgp; /**< for var p point j */
6460: double *gpp, *gmp; /**< for var p point j */
6461: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6462: double ***p3mat;
6463: double age,agelim, hf;
6464: /* double ***mobaverage; */
6465: int theta;
6466: char digit[4];
6467: char digitp[25];
6468:
6469: char fileresprobmorprev[FILENAMELENGTH];
6470:
6471: if(popbased==1){
6472: if(mobilav!=0)
6473: strcpy(digitp,"-POPULBASED-MOBILAV_");
6474: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6475: }
6476: else
6477: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6478:
1.218 brouard 6479: /* if (mobilav!=0) { */
6480: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6481: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6482: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6483: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6484: /* } */
6485: /* } */
6486:
6487: strcpy(fileresprobmorprev,"PRMORPREV-");
6488: sprintf(digit,"%-d",ij);
6489: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6490: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6491: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6492: strcat(fileresprobmorprev,fileresu);
6493: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6494: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6495: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6496: }
6497: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6498: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6499: pstamp(ficresprobmorprev);
6500: 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 6501: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.334 brouard 6502: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */ /* To be done*/
1.332 brouard 6503: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 6504: }
6505: for(j=1;j<=cptcoveff;j++)
1.334 brouard 6506: fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]);
1.238 brouard 6507: fprintf(ficresprobmorprev,"\n");
6508:
1.218 brouard 6509: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6510: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6511: fprintf(ficresprobmorprev," p.%-d SE",j);
6512: for(i=1; i<=nlstate;i++)
6513: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6514: }
6515: fprintf(ficresprobmorprev,"\n");
6516:
6517: fprintf(ficgp,"\n# Routine varevsij");
6518: fprintf(ficgp,"\nunset title \n");
6519: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6520: 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");
6521: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6522:
1.218 brouard 6523: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6524: pstamp(ficresvij);
6525: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6526: if(popbased==1)
6527: 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);
6528: else
6529: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6530: fprintf(ficresvij,"# Age");
6531: for(i=1; i<=nlstate;i++)
6532: for(j=1; j<=nlstate;j++)
6533: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6534: fprintf(ficresvij,"\n");
6535:
6536: xp=vector(1,npar);
6537: dnewm=matrix(1,nlstate,1,npar);
6538: doldm=matrix(1,nlstate,1,nlstate);
6539: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6540: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6541:
6542: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6543: gpp=vector(nlstate+1,nlstate+ndeath);
6544: gmp=vector(nlstate+1,nlstate+ndeath);
6545: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6546:
1.218 brouard 6547: if(estepm < stepm){
6548: printf ("Problem %d lower than %d\n",estepm, stepm);
6549: }
6550: else hstepm=estepm;
6551: /* For example we decided to compute the life expectancy with the smallest unit */
6552: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6553: nhstepm is the number of hstepm from age to agelim
6554: nstepm is the number of stepm from age to agelim.
6555: Look at function hpijx to understand why because of memory size limitations,
6556: we decided (b) to get a life expectancy respecting the most precise curvature of the
6557: survival function given by stepm (the optimization length). Unfortunately it
6558: means that if the survival funtion is printed every two years of age and if
6559: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6560: results. So we changed our mind and took the option of the best precision.
6561: */
6562: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6563: agelim = AGESUP;
6564: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6565: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6566: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6567: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6568: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6569: gp=matrix(0,nhstepm,1,nlstate);
6570: gm=matrix(0,nhstepm,1,nlstate);
6571:
6572:
6573: for(theta=1; theta <=npar; theta++){
6574: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6575: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6576: }
1.279 brouard 6577: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6578: * returns into prlim .
1.288 brouard 6579: */
1.242 brouard 6580: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6581:
6582: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6583: if (popbased==1) {
6584: if(mobilav ==0){
6585: for(i=1; i<=nlstate;i++)
6586: prlim[i][i]=probs[(int)age][i][ij];
6587: }else{ /* mobilav */
6588: for(i=1; i<=nlstate;i++)
6589: prlim[i][i]=mobaverage[(int)age][i][ij];
6590: }
6591: }
1.295 brouard 6592: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6593: */
6594: 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 6595: /**< 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 6596: * at horizon h in state j including mortality.
6597: */
1.218 brouard 6598: for(j=1; j<= nlstate; j++){
6599: for(h=0; h<=nhstepm; h++){
6600: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6601: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6602: }
6603: }
1.279 brouard 6604: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6605: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6606: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6607: */
6608: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6609: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6610: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6611: }
6612:
6613: /* Again with minus shift */
1.218 brouard 6614:
6615: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6616: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6617:
1.242 brouard 6618: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6619:
6620: if (popbased==1) {
6621: if(mobilav ==0){
6622: for(i=1; i<=nlstate;i++)
6623: prlim[i][i]=probs[(int)age][i][ij];
6624: }else{ /* mobilav */
6625: for(i=1; i<=nlstate;i++)
6626: prlim[i][i]=mobaverage[(int)age][i][ij];
6627: }
6628: }
6629:
1.235 brouard 6630: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6631:
6632: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6633: for(h=0; h<=nhstepm; h++){
6634: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6635: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6636: }
6637: }
6638: /* This for computing probability of death (h=1 means
6639: computed over hstepm matrices product = hstepm*stepm months)
6640: as a weighted average of prlim.
6641: */
6642: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6643: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6644: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6645: }
1.279 brouard 6646: /* end shifting computations */
6647:
6648: /**< Computing gradient matrix at horizon h
6649: */
1.218 brouard 6650: for(j=1; j<= nlstate; j++) /* vareij */
6651: for(h=0; h<=nhstepm; h++){
6652: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6653: }
1.279 brouard 6654: /**< Gradient of overall mortality p.3 (or p.j)
6655: */
6656: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6657: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6658: }
6659:
6660: } /* End theta */
1.279 brouard 6661:
6662: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6663: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6664:
6665: for(h=0; h<=nhstepm; h++) /* veij */
6666: for(j=1; j<=nlstate;j++)
6667: for(theta=1; theta <=npar; theta++)
6668: trgradg[h][j][theta]=gradg[h][theta][j];
6669:
6670: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6671: for(theta=1; theta <=npar; theta++)
6672: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6673: /**< as well as its transposed matrix
6674: */
1.218 brouard 6675:
6676: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6677: for(i=1;i<=nlstate;i++)
6678: for(j=1;j<=nlstate;j++)
6679: vareij[i][j][(int)age] =0.;
1.279 brouard 6680:
6681: /* Computing trgradg by matcov by gradg at age and summing over h
6682: * and k (nhstepm) formula 15 of article
6683: * Lievre-Brouard-Heathcote
6684: */
6685:
1.218 brouard 6686: for(h=0;h<=nhstepm;h++){
6687: for(k=0;k<=nhstepm;k++){
6688: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6689: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6690: for(i=1;i<=nlstate;i++)
6691: for(j=1;j<=nlstate;j++)
6692: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6693: }
6694: }
6695:
1.279 brouard 6696: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6697: * p.j overall mortality formula 49 but computed directly because
6698: * we compute the grad (wix pijx) instead of grad (pijx),even if
6699: * wix is independent of theta.
6700: */
1.218 brouard 6701: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6702: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6703: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6704: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6705: varppt[j][i]=doldmp[j][i];
6706: /* end ppptj */
6707: /* x centered again */
6708:
1.242 brouard 6709: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6710:
6711: if (popbased==1) {
6712: if(mobilav ==0){
6713: for(i=1; i<=nlstate;i++)
6714: prlim[i][i]=probs[(int)age][i][ij];
6715: }else{ /* mobilav */
6716: for(i=1; i<=nlstate;i++)
6717: prlim[i][i]=mobaverage[(int)age][i][ij];
6718: }
6719: }
6720:
6721: /* This for computing probability of death (h=1 means
6722: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6723: as a weighted average of prlim.
6724: */
1.235 brouard 6725: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6726: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6727: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6728: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6729: }
6730: /* end probability of death */
6731:
6732: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6733: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6734: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6735: for(i=1; i<=nlstate;i++){
6736: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6737: }
6738: }
6739: fprintf(ficresprobmorprev,"\n");
6740:
6741: fprintf(ficresvij,"%.0f ",age );
6742: for(i=1; i<=nlstate;i++)
6743: for(j=1; j<=nlstate;j++){
6744: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6745: }
6746: fprintf(ficresvij,"\n");
6747: free_matrix(gp,0,nhstepm,1,nlstate);
6748: free_matrix(gm,0,nhstepm,1,nlstate);
6749: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6750: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6751: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6752: } /* End age */
6753: free_vector(gpp,nlstate+1,nlstate+ndeath);
6754: free_vector(gmp,nlstate+1,nlstate+ndeath);
6755: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6756: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6757: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6758: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6759: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6760: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6761: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6762: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6763: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6764: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6765: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6766: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6767: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6768: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6769: 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);
6770: /* 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 6771: */
1.218 brouard 6772: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6773: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6774:
1.218 brouard 6775: free_vector(xp,1,npar);
6776: free_matrix(doldm,1,nlstate,1,nlstate);
6777: free_matrix(dnewm,1,nlstate,1,npar);
6778: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6779: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6780: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6781: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6782: fclose(ficresprobmorprev);
6783: fflush(ficgp);
6784: fflush(fichtm);
6785: } /* end varevsij */
1.126 brouard 6786:
6787: /************ Variance of prevlim ******************/
1.269 brouard 6788: 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 6789: {
1.205 brouard 6790: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6791: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6792:
1.268 brouard 6793: double **dnewmpar,**doldm;
1.126 brouard 6794: int i, j, nhstepm, hstepm;
6795: double *xp;
6796: double *gp, *gm;
6797: double **gradg, **trgradg;
1.208 brouard 6798: double **mgm, **mgp;
1.126 brouard 6799: double age,agelim;
6800: int theta;
6801:
6802: pstamp(ficresvpl);
1.288 brouard 6803: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6804: fprintf(ficresvpl,"# Age ");
6805: if(nresult >=1)
6806: fprintf(ficresvpl," Result# ");
1.126 brouard 6807: for(i=1; i<=nlstate;i++)
6808: fprintf(ficresvpl," %1d-%1d",i,i);
6809: fprintf(ficresvpl,"\n");
6810:
6811: xp=vector(1,npar);
1.268 brouard 6812: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6813: doldm=matrix(1,nlstate,1,nlstate);
6814:
6815: hstepm=1*YEARM; /* Every year of age */
6816: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6817: agelim = AGESUP;
6818: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6819: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6820: if (stepm >= YEARM) hstepm=1;
6821: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6822: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6823: mgp=matrix(1,npar,1,nlstate);
6824: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6825: gp=vector(1,nlstate);
6826: gm=vector(1,nlstate);
6827:
6828: for(theta=1; theta <=npar; theta++){
6829: for(i=1; i<=npar; i++){ /* Computes gradient */
6830: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6831: }
1.288 brouard 6832: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6833: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6834: /* else */
6835: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6836: for(i=1;i<=nlstate;i++){
1.126 brouard 6837: gp[i] = prlim[i][i];
1.208 brouard 6838: mgp[theta][i] = prlim[i][i];
6839: }
1.126 brouard 6840: for(i=1; i<=npar; i++) /* Computes gradient */
6841: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6842: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6843: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6844: /* else */
6845: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6846: for(i=1;i<=nlstate;i++){
1.126 brouard 6847: gm[i] = prlim[i][i];
1.208 brouard 6848: mgm[theta][i] = prlim[i][i];
6849: }
1.126 brouard 6850: for(i=1;i<=nlstate;i++)
6851: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6852: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6853: } /* End theta */
6854:
6855: trgradg =matrix(1,nlstate,1,npar);
6856:
6857: for(j=1; j<=nlstate;j++)
6858: for(theta=1; theta <=npar; theta++)
6859: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6860: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6861: /* printf("\nmgm mgp %d ",(int)age); */
6862: /* for(j=1; j<=nlstate;j++){ */
6863: /* printf(" %d ",j); */
6864: /* for(theta=1; theta <=npar; theta++) */
6865: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6866: /* printf("\n "); */
6867: /* } */
6868: /* } */
6869: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6870: /* printf("\n gradg %d ",(int)age); */
6871: /* for(j=1; j<=nlstate;j++){ */
6872: /* printf("%d ",j); */
6873: /* for(theta=1; theta <=npar; theta++) */
6874: /* printf("%d %lf ",theta,gradg[theta][j]); */
6875: /* printf("\n "); */
6876: /* } */
6877: /* } */
1.126 brouard 6878:
6879: for(i=1;i<=nlstate;i++)
6880: varpl[i][(int)age] =0.;
1.209 brouard 6881: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6882: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6883: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6884: }else{
1.268 brouard 6885: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6886: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6887: }
1.126 brouard 6888: for(i=1;i<=nlstate;i++)
6889: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6890:
6891: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6892: if(nresult >=1)
6893: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6894: for(i=1; i<=nlstate;i++){
1.126 brouard 6895: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6896: /* for(j=1;j<=nlstate;j++) */
6897: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6898: }
1.126 brouard 6899: fprintf(ficresvpl,"\n");
6900: free_vector(gp,1,nlstate);
6901: free_vector(gm,1,nlstate);
1.208 brouard 6902: free_matrix(mgm,1,npar,1,nlstate);
6903: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6904: free_matrix(gradg,1,npar,1,nlstate);
6905: free_matrix(trgradg,1,nlstate,1,npar);
6906: } /* End age */
6907:
6908: free_vector(xp,1,npar);
6909: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6910: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6911:
6912: }
6913:
6914:
6915: /************ Variance of backprevalence limit ******************/
1.269 brouard 6916: 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 6917: {
6918: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6919: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6920:
6921: double **dnewmpar,**doldm;
6922: int i, j, nhstepm, hstepm;
6923: double *xp;
6924: double *gp, *gm;
6925: double **gradg, **trgradg;
6926: double **mgm, **mgp;
6927: double age,agelim;
6928: int theta;
6929:
6930: pstamp(ficresvbl);
6931: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6932: fprintf(ficresvbl,"# Age ");
6933: if(nresult >=1)
6934: fprintf(ficresvbl," Result# ");
6935: for(i=1; i<=nlstate;i++)
6936: fprintf(ficresvbl," %1d-%1d",i,i);
6937: fprintf(ficresvbl,"\n");
6938:
6939: xp=vector(1,npar);
6940: dnewmpar=matrix(1,nlstate,1,npar);
6941: doldm=matrix(1,nlstate,1,nlstate);
6942:
6943: hstepm=1*YEARM; /* Every year of age */
6944: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6945: agelim = AGEINF;
6946: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6947: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6948: if (stepm >= YEARM) hstepm=1;
6949: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6950: gradg=matrix(1,npar,1,nlstate);
6951: mgp=matrix(1,npar,1,nlstate);
6952: mgm=matrix(1,npar,1,nlstate);
6953: gp=vector(1,nlstate);
6954: gm=vector(1,nlstate);
6955:
6956: for(theta=1; theta <=npar; theta++){
6957: for(i=1; i<=npar; i++){ /* Computes gradient */
6958: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6959: }
6960: if(mobilavproj > 0 )
6961: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6962: else
6963: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6964: for(i=1;i<=nlstate;i++){
6965: gp[i] = bprlim[i][i];
6966: mgp[theta][i] = bprlim[i][i];
6967: }
6968: for(i=1; i<=npar; i++) /* Computes gradient */
6969: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6970: if(mobilavproj > 0 )
6971: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6972: else
6973: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6974: for(i=1;i<=nlstate;i++){
6975: gm[i] = bprlim[i][i];
6976: mgm[theta][i] = bprlim[i][i];
6977: }
6978: for(i=1;i<=nlstate;i++)
6979: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6980: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6981: } /* End theta */
6982:
6983: trgradg =matrix(1,nlstate,1,npar);
6984:
6985: for(j=1; j<=nlstate;j++)
6986: for(theta=1; theta <=npar; theta++)
6987: trgradg[j][theta]=gradg[theta][j];
6988: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6989: /* printf("\nmgm mgp %d ",(int)age); */
6990: /* for(j=1; j<=nlstate;j++){ */
6991: /* printf(" %d ",j); */
6992: /* for(theta=1; theta <=npar; theta++) */
6993: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6994: /* printf("\n "); */
6995: /* } */
6996: /* } */
6997: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6998: /* printf("\n gradg %d ",(int)age); */
6999: /* for(j=1; j<=nlstate;j++){ */
7000: /* printf("%d ",j); */
7001: /* for(theta=1; theta <=npar; theta++) */
7002: /* printf("%d %lf ",theta,gradg[theta][j]); */
7003: /* printf("\n "); */
7004: /* } */
7005: /* } */
7006:
7007: for(i=1;i<=nlstate;i++)
7008: varbpl[i][(int)age] =0.;
7009: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
7010: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7011: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7012: }else{
7013: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7014: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7015: }
7016: for(i=1;i<=nlstate;i++)
7017: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7018:
7019: fprintf(ficresvbl,"%.0f ",age );
7020: if(nresult >=1)
7021: fprintf(ficresvbl,"%d ",nres );
7022: for(i=1; i<=nlstate;i++)
7023: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
7024: fprintf(ficresvbl,"\n");
7025: free_vector(gp,1,nlstate);
7026: free_vector(gm,1,nlstate);
7027: free_matrix(mgm,1,npar,1,nlstate);
7028: free_matrix(mgp,1,npar,1,nlstate);
7029: free_matrix(gradg,1,npar,1,nlstate);
7030: free_matrix(trgradg,1,nlstate,1,npar);
7031: } /* End age */
7032:
7033: free_vector(xp,1,npar);
7034: free_matrix(doldm,1,nlstate,1,npar);
7035: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 7036:
7037: }
7038:
7039: /************ Variance of one-step probabilities ******************/
7040: 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 7041: {
7042: int i, j=0, k1, l1, tj;
7043: int k2, l2, j1, z1;
7044: int k=0, l;
7045: int first=1, first1, first2;
1.326 brouard 7046: int nres=0; /* New */
1.222 brouard 7047: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
7048: double **dnewm,**doldm;
7049: double *xp;
7050: double *gp, *gm;
7051: double **gradg, **trgradg;
7052: double **mu;
7053: double age, cov[NCOVMAX+1];
7054: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
7055: int theta;
7056: char fileresprob[FILENAMELENGTH];
7057: char fileresprobcov[FILENAMELENGTH];
7058: char fileresprobcor[FILENAMELENGTH];
7059: double ***varpij;
7060:
7061: strcpy(fileresprob,"PROB_");
7062: strcat(fileresprob,fileres);
7063: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
7064: printf("Problem with resultfile: %s\n", fileresprob);
7065: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
7066: }
7067: strcpy(fileresprobcov,"PROBCOV_");
7068: strcat(fileresprobcov,fileresu);
7069: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
7070: printf("Problem with resultfile: %s\n", fileresprobcov);
7071: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
7072: }
7073: strcpy(fileresprobcor,"PROBCOR_");
7074: strcat(fileresprobcor,fileresu);
7075: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
7076: printf("Problem with resultfile: %s\n", fileresprobcor);
7077: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
7078: }
7079: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7080: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7081: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7082: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7083: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7084: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7085: pstamp(ficresprob);
7086: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
7087: fprintf(ficresprob,"# Age");
7088: pstamp(ficresprobcov);
7089: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
7090: fprintf(ficresprobcov,"# Age");
7091: pstamp(ficresprobcor);
7092: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
7093: fprintf(ficresprobcor,"# Age");
1.126 brouard 7094:
7095:
1.222 brouard 7096: for(i=1; i<=nlstate;i++)
7097: for(j=1; j<=(nlstate+ndeath);j++){
7098: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
7099: fprintf(ficresprobcov," p%1d-%1d ",i,j);
7100: fprintf(ficresprobcor," p%1d-%1d ",i,j);
7101: }
7102: /* fprintf(ficresprob,"\n");
7103: fprintf(ficresprobcov,"\n");
7104: fprintf(ficresprobcor,"\n");
7105: */
7106: xp=vector(1,npar);
7107: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7108: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7109: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
7110: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
7111: first=1;
7112: fprintf(ficgp,"\n# Routine varprob");
7113: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
7114: fprintf(fichtm,"\n");
7115:
1.288 brouard 7116: 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 7117: 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);
7118: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 7119: and drawn. It helps understanding how is the covariance between two incidences.\
7120: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 7121: 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 7122: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
7123: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
7124: standard deviations wide on each axis. <br>\
7125: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
7126: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
7127: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
7128:
1.222 brouard 7129: cov[1]=1;
7130: /* tj=cptcoveff; */
1.225 brouard 7131: tj = (int) pow(2,cptcoveff);
1.222 brouard 7132: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
7133: j1=0;
1.332 brouard 7134:
7135: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
7136: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.334 brouard 7137: 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 7138: if(tj != 1 && TKresult[nres]!= j1)
7139: continue;
7140:
7141: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
7142: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
7143: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 7144: if (cptcovn>0) {
1.334 brouard 7145: fprintf(ficresprob, "\n#********** Variable ");
7146: fprintf(ficresprobcov, "\n#********** Variable ");
7147: fprintf(ficgp, "\n#********** Variable ");
7148: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
7149: fprintf(ficresprobcor, "\n#********** Variable ");
7150:
7151: /* Including quantitative variables of the resultline to be done */
7152: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
7153: printf("Varprob modelresult[%d][%d]=%d model=%s \n",nres, z1, modelresult[nres][z1], model);
7154: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=%s \n",nres, z1, modelresult[nres][z1], model);
7155: /* fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=%s resultline[%d]=%s \n",nres, z1, modelresult[nres][z1], model, nres, resultline[nres]); */
7156: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
7157: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
7158: 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 */
7159: 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 */
7160: 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 */
7161: 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 */
7162: 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 */
7163: fprintf(ficresprob,"fixed ");
7164: fprintf(ficresprobcov,"fixed ");
7165: fprintf(ficgp,"fixed ");
7166: fprintf(fichtmcov,"fixed ");
7167: fprintf(ficresprobcor,"fixed ");
7168: }else{
7169: fprintf(ficresprob,"varyi ");
7170: fprintf(ficresprobcov,"varyi ");
7171: fprintf(ficgp,"varyi ");
7172: fprintf(fichtmcov,"varyi ");
7173: fprintf(ficresprobcor,"varyi ");
7174: }
7175: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
7176: /* For each selected (single) quantitative value */
7177: fprintf(ficresprob," V%d=%f ",Tvqresult[nres][z1],Tqresult[nres][z1]);
7178: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
7179: fprintf(ficresprob,"fixed ");
7180: fprintf(ficresprobcov,"fixed ");
7181: fprintf(ficgp,"fixed ");
7182: fprintf(fichtmcov,"fixed ");
7183: fprintf(ficresprobcor,"fixed ");
7184: }else{
7185: fprintf(ficresprob,"varyi ");
7186: fprintf(ficresprobcov,"varyi ");
7187: fprintf(ficgp,"varyi ");
7188: fprintf(fichtmcov,"varyi ");
7189: fprintf(ficresprobcor,"varyi ");
7190: }
7191: }else{
7192: 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 */
7193: 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 */
7194: exit(1);
7195: }
7196: } /* End loop on variable of this resultline */
7197: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 7198: fprintf(ficresprob, "**********\n#\n");
7199: fprintf(ficresprobcov, "**********\n#\n");
7200: fprintf(ficgp, "**********\n#\n");
7201: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
7202: fprintf(ficresprobcor, "**********\n#");
7203: if(invalidvarcomb[j1]){
7204: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7205: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7206: continue;
7207: }
7208: }
7209: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7210: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7211: gp=vector(1,(nlstate)*(nlstate+ndeath));
7212: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 7213: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 7214: cov[2]=age;
7215: if(nagesqr==1)
7216: cov[3]= age*age;
1.334 brouard 7217: /* New code end of combination but for each resultline */
7218: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
7219: if(Typevar[k1]==1){ /* A product with age */
7220: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 7221: }else{
1.334 brouard 7222: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 7223: }
1.334 brouard 7224: }/* End of loop on model equation */
7225: /* Old code */
7226: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
7227: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
7228: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
7229: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
7230: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
7231: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
7232: /* * 1 1 1 1 1 */
7233: /* * 2 2 1 1 1 */
7234: /* * 3 1 2 1 1 */
7235: /* *\/ */
7236: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
7237: /* } */
7238: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
7239: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
7240: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
7241: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
7242: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
7243: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
7244: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7245: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
7246: /* 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]); */
7247: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
7248: /* /\* exit(1); *\/ */
7249: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
7250: /* } */
7251: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7252: /* } */
7253: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
7254: /* if(Dummy[Tvard[k][1]]==0){ */
7255: /* if(Dummy[Tvard[k][2]]==0){ */
7256: /* 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]])]; */
7257: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7258: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
7259: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
7260: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
7261: /* } */
7262: /* }else{ */
7263: /* if(Dummy[Tvard[k][2]]==0){ */
7264: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
7265: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
7266: /* }else{ */
7267: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
7268: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
7269: /* } */
7270: /* } */
7271: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7272: /* } */
1.326 brouard 7273: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7274: for(theta=1; theta <=npar; theta++){
7275: for(i=1; i<=npar; i++)
7276: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7277:
1.222 brouard 7278: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7279:
1.222 brouard 7280: k=0;
7281: for(i=1; i<= (nlstate); i++){
7282: for(j=1; j<=(nlstate+ndeath);j++){
7283: k=k+1;
7284: gp[k]=pmmij[i][j];
7285: }
7286: }
1.220 brouard 7287:
1.222 brouard 7288: for(i=1; i<=npar; i++)
7289: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7290:
1.222 brouard 7291: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7292: k=0;
7293: for(i=1; i<=(nlstate); i++){
7294: for(j=1; j<=(nlstate+ndeath);j++){
7295: k=k+1;
7296: gm[k]=pmmij[i][j];
7297: }
7298: }
1.220 brouard 7299:
1.222 brouard 7300: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7301: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7302: }
1.126 brouard 7303:
1.222 brouard 7304: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7305: for(theta=1; theta <=npar; theta++)
7306: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7307:
1.222 brouard 7308: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7309: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7310:
1.222 brouard 7311: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7312:
1.222 brouard 7313: k=0;
7314: for(i=1; i<=(nlstate); i++){
7315: for(j=1; j<=(nlstate+ndeath);j++){
7316: k=k+1;
7317: mu[k][(int) age]=pmmij[i][j];
7318: }
7319: }
7320: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7321: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7322: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7323:
1.222 brouard 7324: /*printf("\n%d ",(int)age);
7325: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7326: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7327: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7328: }*/
1.220 brouard 7329:
1.222 brouard 7330: fprintf(ficresprob,"\n%d ",(int)age);
7331: fprintf(ficresprobcov,"\n%d ",(int)age);
7332: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7333:
1.222 brouard 7334: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7335: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7336: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7337: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7338: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7339: }
7340: i=0;
7341: for (k=1; k<=(nlstate);k++){
7342: for (l=1; l<=(nlstate+ndeath);l++){
7343: i++;
7344: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7345: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7346: for (j=1; j<=i;j++){
7347: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7348: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7349: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7350: }
7351: }
7352: }/* end of loop for state */
7353: } /* end of loop for age */
7354: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7355: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7356: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7357: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7358:
7359: /* Confidence intervalle of pij */
7360: /*
7361: fprintf(ficgp,"\nunset parametric;unset label");
7362: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7363: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7364: 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);
7365: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7366: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7367: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7368: */
7369:
7370: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7371: first1=1;first2=2;
7372: for (k2=1; k2<=(nlstate);k2++){
7373: for (l2=1; l2<=(nlstate+ndeath);l2++){
7374: if(l2==k2) continue;
7375: j=(k2-1)*(nlstate+ndeath)+l2;
7376: for (k1=1; k1<=(nlstate);k1++){
7377: for (l1=1; l1<=(nlstate+ndeath);l1++){
7378: if(l1==k1) continue;
7379: i=(k1-1)*(nlstate+ndeath)+l1;
7380: if(i<=j) continue;
7381: for (age=bage; age<=fage; age ++){
7382: if ((int)age %5==0){
7383: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7384: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7385: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7386: mu1=mu[i][(int) age]/stepm*YEARM ;
7387: mu2=mu[j][(int) age]/stepm*YEARM;
7388: c12=cv12/sqrt(v1*v2);
7389: /* Computing eigen value of matrix of covariance */
7390: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7391: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7392: if ((lc2 <0) || (lc1 <0) ){
7393: if(first2==1){
7394: first1=0;
7395: 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);
7396: }
7397: 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);
7398: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7399: /* lc2=fabs(lc2); */
7400: }
1.220 brouard 7401:
1.222 brouard 7402: /* Eigen vectors */
1.280 brouard 7403: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7404: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7405: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7406: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7407: }else
7408: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7409: /*v21=sqrt(1.-v11*v11); *//* error */
7410: v21=(lc1-v1)/cv12*v11;
7411: v12=-v21;
7412: v22=v11;
7413: tnalp=v21/v11;
7414: if(first1==1){
7415: first1=0;
7416: 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);
7417: }
7418: 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);
7419: /*printf(fignu*/
7420: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7421: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7422: if(first==1){
7423: first=0;
7424: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7425: fprintf(ficgp,"\nset parametric;unset label");
7426: 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);
7427: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7428: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7429: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7430: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7431: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7432: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7433: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7434: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7435: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7436: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7437: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7438: 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 7439: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7440: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7441: }else{
7442: first=0;
7443: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7444: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7445: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7446: 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 7447: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7448: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7449: }/* if first */
7450: } /* age mod 5 */
7451: } /* end loop age */
7452: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7453: first=1;
7454: } /*l12 */
7455: } /* k12 */
7456: } /*l1 */
7457: }/* k1 */
1.332 brouard 7458: } /* loop on combination of covariates j1 */
1.326 brouard 7459: } /* loop on nres */
1.222 brouard 7460: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7461: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7462: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7463: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7464: free_vector(xp,1,npar);
7465: fclose(ficresprob);
7466: fclose(ficresprobcov);
7467: fclose(ficresprobcor);
7468: fflush(ficgp);
7469: fflush(fichtmcov);
7470: }
1.126 brouard 7471:
7472:
7473: /******************* Printing html file ***********/
1.201 brouard 7474: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7475: int lastpass, int stepm, int weightopt, char model[],\
7476: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7477: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7478: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7479: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7480: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7481: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7482: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7483: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7484: </ul>");
1.319 brouard 7485: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7486: /* </ul>", model); */
1.214 brouard 7487: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7488: 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",
7489: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 7490: 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 7491: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7492: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7493: fprintf(fichtm,"\
7494: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7495: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7496: fprintf(fichtm,"\
1.217 brouard 7497: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7498: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7499: fprintf(fichtm,"\
1.288 brouard 7500: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7501: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7502: fprintf(fichtm,"\
1.288 brouard 7503: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7504: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7505: fprintf(fichtm,"\
1.211 brouard 7506: - (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 7507: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7508: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7509: if(prevfcast==1){
7510: fprintf(fichtm,"\
7511: - Prevalence projections by age and states: \
1.201 brouard 7512: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7513: }
1.126 brouard 7514:
7515:
1.225 brouard 7516: m=pow(2,cptcoveff);
1.222 brouard 7517: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7518:
1.317 brouard 7519: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7520:
7521: jj1=0;
7522:
7523: fprintf(fichtm," \n<ul>");
7524: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7525: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7526: if(m != 1 && TKresult[nres]!= k1)
7527: continue;
7528: jj1++;
7529: if (cptcovn > 0) {
7530: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
7531: for (cpt=1; cpt<=cptcoveff;cpt++){
7532: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7533: }
7534: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7535: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7536: }
7537: fprintf(fichtm,"\">");
7538:
7539: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7540: fprintf(fichtm,"************ Results for covariates");
7541: for (cpt=1; cpt<=cptcoveff;cpt++){
7542: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7543: }
7544: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7545: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7546: }
7547: if(invalidvarcomb[k1]){
7548: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7549: continue;
7550: }
7551: fprintf(fichtm,"</a></li>");
7552: } /* cptcovn >0 */
7553: }
1.317 brouard 7554: fprintf(fichtm," \n</ul>");
1.264 brouard 7555:
1.222 brouard 7556: jj1=0;
1.237 brouard 7557:
7558: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7559: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7560: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7561: continue;
1.220 brouard 7562:
1.222 brouard 7563: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7564: jj1++;
7565: if (cptcovn > 0) {
1.264 brouard 7566: fprintf(fichtm,"\n<p><a name=\"rescov");
7567: for (cpt=1; cpt<=cptcoveff;cpt++){
7568: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7569: }
7570: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7571: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7572: }
7573: fprintf(fichtm,"\"</a>");
7574:
1.222 brouard 7575: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7576: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7577: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7578: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7579: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7580: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7581: }
1.237 brouard 7582: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7583: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7584: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7585: }
7586:
1.230 brouard 7587: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.321 brouard 7588: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7589: if(invalidvarcomb[k1]){
7590: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7591: printf("\nCombination (%d) ignored because no cases \n",k1);
7592: continue;
7593: }
7594: }
7595: /* aij, bij */
1.259 brouard 7596: 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 7597: <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 7598: /* Pij */
1.241 brouard 7599: 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> \
7600: <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 7601: /* Quasi-incidences */
7602: 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 7603: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7604: 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 7605: 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> \
7606: <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 7607: /* Survival functions (period) in state j */
7608: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7609: 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);
7610: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7611: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7612: }
7613: /* State specific survival functions (period) */
7614: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7615: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7616: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7617: <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);
7618: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7619: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7620: }
1.288 brouard 7621: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7622: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7623: 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);
7624: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"P_"),subdirf2(optionfilefiname,"P_"));
7625: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7626: }
1.296 brouard 7627: if(prevbcast==1){
1.288 brouard 7628: /* Backward prevalence in each health state */
1.222 brouard 7629: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7630: 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> \
1.241 brouard 7631: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"PB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 7632: }
1.217 brouard 7633: }
1.222 brouard 7634: if(prevfcast==1){
1.288 brouard 7635: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7636: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7637: 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);
7638: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7639: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7640: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7641: }
7642: }
1.296 brouard 7643: if(prevbcast==1){
1.268 brouard 7644: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7645: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7646: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7647: 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 \
7648: 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 7649: 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);
7650: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7651: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7652: }
7653: }
1.220 brouard 7654:
1.222 brouard 7655: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7656: 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);
7657: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7658: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7659: }
7660: /* } /\* end i1 *\/ */
7661: }/* End k1 */
7662: fprintf(fichtm,"</ul>");
1.126 brouard 7663:
1.222 brouard 7664: fprintf(fichtm,"\
1.126 brouard 7665: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7666: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7667: - 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 7668: But because parameters are usually highly correlated (a higher incidence of disability \
7669: and a higher incidence of recovery can give very close observed transition) it might \
7670: be very useful to look not only at linear confidence intervals estimated from the \
7671: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7672: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7673: covariance matrix of the one-step probabilities. \
7674: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7675:
1.222 brouard 7676: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7677: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7678: fprintf(fichtm,"\
1.126 brouard 7679: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7680: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7681:
1.222 brouard 7682: fprintf(fichtm,"\
1.126 brouard 7683: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7684: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7685: fprintf(fichtm,"\
1.126 brouard 7686: - 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): \
7687: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7688: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7689: fprintf(fichtm,"\
1.126 brouard 7690: - (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): \
7691: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7692: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7693: fprintf(fichtm,"\
1.288 brouard 7694: - 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 7695: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7696: fprintf(fichtm,"\
1.128 brouard 7697: - 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 7698: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7699: fprintf(fichtm,"\
1.288 brouard 7700: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7701: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7702:
7703: /* if(popforecast==1) fprintf(fichtm,"\n */
7704: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7705: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7706: /* <br>",fileres,fileres,fileres,fileres); */
7707: /* else */
7708: /* fprintf(fichtm,"\n No population forecast: popforecast = %d (instead of 1) or stepm = %d (instead of 1) or model=%s (instead of .)<br><br></li>\n",popforecast, stepm, model); */
1.222 brouard 7709: fflush(fichtm);
1.126 brouard 7710:
1.225 brouard 7711: m=pow(2,cptcoveff);
1.222 brouard 7712: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7713:
1.317 brouard 7714: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7715:
7716: jj1=0;
7717:
7718: fprintf(fichtm," \n<ul>");
7719: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7720: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7721: if(m != 1 && TKresult[nres]!= k1)
7722: continue;
7723: jj1++;
7724: if (cptcovn > 0) {
7725: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
7726: for (cpt=1; cpt<=cptcoveff;cpt++){
7727: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7728: }
7729: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7730: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7731: }
7732: fprintf(fichtm,"\">");
7733:
7734: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7735: fprintf(fichtm,"************ Results for covariates");
7736: for (cpt=1; cpt<=cptcoveff;cpt++){
7737: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7738: }
7739: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7740: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7741: }
7742: if(invalidvarcomb[k1]){
7743: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7744: continue;
7745: }
7746: fprintf(fichtm,"</a></li>");
7747: } /* cptcovn >0 */
7748: }
7749: fprintf(fichtm," \n</ul>");
7750:
1.222 brouard 7751: jj1=0;
1.237 brouard 7752:
1.241 brouard 7753: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7754: for(k1=1; k1<=m;k1++){
1.253 brouard 7755: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7756: continue;
1.222 brouard 7757: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7758: jj1++;
1.126 brouard 7759: if (cptcovn > 0) {
1.317 brouard 7760: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
7761: for (cpt=1; cpt<=cptcoveff;cpt++){
7762: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7763: }
7764: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7765: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7766: }
7767: fprintf(fichtm,"\"</a>");
7768:
1.126 brouard 7769: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.317 brouard 7770: for (cpt=1; cpt<=cptcoveff;cpt++){ /**< cptcoveff number of variables */
1.237 brouard 7771: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
1.317 brouard 7772: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
1.237 brouard 7773: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 7774: }
1.237 brouard 7775: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7776: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7777: }
7778:
1.321 brouard 7779: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 7780:
1.222 brouard 7781: if(invalidvarcomb[k1]){
7782: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7783: continue;
7784: }
1.126 brouard 7785: }
7786: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7787: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7788: 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);
7789: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7790: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7791: }
7792: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7793: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7794: true period expectancies (those weighted with period prevalences are also\
7795: drawn in addition to the population based expectancies computed using\
1.314 brouard 7796: 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);
7797: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7798: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7799: /* } /\* end i1 *\/ */
7800: }/* End k1 */
1.241 brouard 7801: }/* End nres */
1.222 brouard 7802: fprintf(fichtm,"</ul>");
7803: fflush(fichtm);
1.126 brouard 7804: }
7805:
7806: /******************* Gnuplot file **************/
1.296 brouard 7807: 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 7808:
7809: char dirfileres[132],optfileres[132];
1.264 brouard 7810: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7811: int cpt=0,k1=0,i=0,k=0,j=0,jk=0,k2=0,k3=0,k4=0,ij=0, ijp=0, l=0;
1.211 brouard 7812: int lv=0, vlv=0, kl=0;
1.130 brouard 7813: int ng=0;
1.201 brouard 7814: int vpopbased;
1.223 brouard 7815: int ioffset; /* variable offset for columns */
1.270 brouard 7816: int iyearc=1; /* variable column for year of projection */
7817: int iagec=1; /* variable column for age of projection */
1.235 brouard 7818: int nres=0; /* Index of resultline */
1.266 brouard 7819: int istart=1; /* For starting graphs in projections */
1.219 brouard 7820:
1.126 brouard 7821: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7822: /* printf("Problem with file %s",optionfilegnuplot); */
7823: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7824: /* } */
7825:
7826: /*#ifdef windows */
7827: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7828: /*#endif */
1.225 brouard 7829: m=pow(2,cptcoveff);
1.126 brouard 7830:
1.274 brouard 7831: /* diagram of the model */
7832: fprintf(ficgp,"\n#Diagram of the model \n");
7833: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7834: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7835: 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);
7836:
7837: fprintf(ficgp,"\n#Centripete arrows (turning in other direction (1-i) instead of (i-1)) \nset for [i=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);
7838: fprintf(ficgp,"\n#show arrow\nunset label\n");
7839: 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);
7840: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7841: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7842: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7843: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7844:
1.202 brouard 7845: /* Contribution to likelihood */
7846: /* Plot the probability implied in the likelihood */
1.223 brouard 7847: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7848: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7849: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7850: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7851: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7852: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7853: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7854: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7855: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7856: 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));
7857: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7858: 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));
7859: for (i=1; i<= nlstate ; i ++) {
7860: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7861: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7862: 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);
7863: for (j=2; j<= nlstate+ndeath ; j ++) {
7864: 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);
7865: }
7866: fprintf(ficgp,";\nset out; unset ylabel;\n");
7867: }
7868: /* 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 */
7869: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7870: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7871: fprintf(ficgp,"\nset out;unset log\n");
7872: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7873:
1.126 brouard 7874: strcpy(dirfileres,optionfilefiname);
7875: strcpy(optfileres,"vpl");
1.223 brouard 7876: /* 1eme*/
1.238 brouard 7877: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7878: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7879: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7880: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7881: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7882: continue;
7883: /* We are interested in selected combination by the resultline */
1.246 brouard 7884: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7885: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7886: strcpy(gplotlabel,"(");
1.238 brouard 7887: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
1.332 brouard 7888: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the value of the covariate corresponding to k1 combination *\/ */
7889: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 7890: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7891: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7892: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7893: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7894: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7895: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7896: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7897: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7898: }
7899: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7900: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7901: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7902: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7903: }
7904: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7905: /* printf("\n#\n"); */
1.238 brouard 7906: fprintf(ficgp,"\n#\n");
7907: if(invalidvarcomb[k1]){
1.260 brouard 7908: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7909: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7910: continue;
7911: }
1.235 brouard 7912:
1.241 brouard 7913: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7914: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7915: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
1.321 brouard 7916: fprintf(ficgp,"set title \"Alive state %d %s model=%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 7917: 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);
7918: /* 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); */
7919: /* k1-1 error should be nres-1*/
1.238 brouard 7920: for (i=1; i<= nlstate ; i ++) {
7921: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7922: else fprintf(ficgp," %%*lf (%%*lf)");
7923: }
1.288 brouard 7924: 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 7925: for (i=1; i<= nlstate ; i ++) {
7926: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7927: else fprintf(ficgp," %%*lf (%%*lf)");
7928: }
1.260 brouard 7929: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7930: for (i=1; i<= nlstate ; i ++) {
7931: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7932: else fprintf(ficgp," %%*lf (%%*lf)");
7933: }
1.265 brouard 7934: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7935:
7936: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7937: if(cptcoveff ==0){
1.271 brouard 7938: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7939: }else{
7940: kl=0;
7941: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 7942: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
7943: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 7944: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7945: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7946: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7947: vlv= nbcode[Tvaraff[k]][lv];
7948: kl++;
7949: /* 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 *\/ */
7950: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7951: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7952: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7953: if(k==cptcoveff){
7954: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7955: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7956: }else{
7957: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7958: kl++;
7959: }
7960: } /* end covariate */
7961: } /* end if no covariate */
7962:
1.296 brouard 7963: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7964: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7965: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7966: if(cptcoveff ==0){
1.245 brouard 7967: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7968: }else{
7969: kl=0;
7970: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 7971: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
7972: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 7973: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7974: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7975: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 7976: /* vlv= nbcode[Tvaraff[k]][lv]; */
7977: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 7978: kl++;
1.238 brouard 7979: /* 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 *\/ */
7980: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7981: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7982: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7983: if(k==cptcoveff){
1.245 brouard 7984: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7985: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7986: }else{
1.332 brouard 7987: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 7988: kl++;
7989: }
7990: } /* end covariate */
7991: } /* end if no covariate */
1.296 brouard 7992: if(prevbcast == 1){
1.268 brouard 7993: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7994: /* k1-1 error should be nres-1*/
7995: for (i=1; i<= nlstate ; i ++) {
7996: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7997: else fprintf(ficgp," %%*lf (%%*lf)");
7998: }
1.271 brouard 7999: 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 8000: for (i=1; i<= nlstate ; i ++) {
8001: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8002: else fprintf(ficgp," %%*lf (%%*lf)");
8003: }
1.276 brouard 8004: 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 8005: for (i=1; i<= nlstate ; i ++) {
8006: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8007: else fprintf(ficgp," %%*lf (%%*lf)");
8008: }
1.274 brouard 8009: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 8010: } /* end if backprojcast */
1.296 brouard 8011: } /* end if prevbcast */
1.276 brouard 8012: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
8013: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 8014: } /* nres */
1.201 brouard 8015: } /* k1 */
8016: } /* cpt */
1.235 brouard 8017:
8018:
1.126 brouard 8019: /*2 eme*/
1.238 brouard 8020: for (k1=1; k1<= m ; k1 ++){
8021: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8022: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 8023: continue;
8024: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 8025: strcpy(gplotlabel,"(");
1.238 brouard 8026: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8027: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8028: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.223 brouard 8029: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8030: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8031: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8032: /* vlv= nbcode[Tvaraff[k]][lv]; */
8033: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 8034: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8035: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 8036: }
1.237 brouard 8037: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 8038: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8039: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 8040: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8041: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 8042: }
1.264 brouard 8043: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8044: fprintf(ficgp,"\n#\n");
1.223 brouard 8045: if(invalidvarcomb[k1]){
8046: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8047: continue;
8048: }
1.219 brouard 8049:
1.241 brouard 8050: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 8051: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 8052: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
8053: if(vpopbased==0){
1.238 brouard 8054: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 8055: }else
1.238 brouard 8056: fprintf(ficgp,"\nreplot ");
8057: for (i=1; i<= nlstate+1 ; i ++) {
8058: k=2*i;
1.261 brouard 8059: 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 8060: for (j=1; j<= nlstate+1 ; j ++) {
8061: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8062: else fprintf(ficgp," %%*lf (%%*lf)");
8063: }
8064: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
8065: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 8066: 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 8067: for (j=1; j<= nlstate+1 ; j ++) {
8068: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8069: else fprintf(ficgp," %%*lf (%%*lf)");
8070: }
8071: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 8072: 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 8073: for (j=1; j<= nlstate+1 ; j ++) {
8074: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8075: else fprintf(ficgp," %%*lf (%%*lf)");
8076: }
8077: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
8078: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
8079: } /* state */
8080: } /* vpopbased */
1.264 brouard 8081: 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 8082: } /* end nres */
8083: } /* k1 end 2 eme*/
8084:
8085:
8086: /*3eme*/
8087: for (k1=1; k1<= m ; k1 ++){
8088: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8089: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 8090: continue;
8091:
1.332 brouard 8092: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 8093: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 8094: strcpy(gplotlabel,"(");
1.238 brouard 8095: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8096: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8097: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.238 brouard 8098: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8099: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8100: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8101: /* vlv= nbcode[Tvaraff[k]][lv]; */
8102: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.238 brouard 8103: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8104: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 8105: }
8106: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8107: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]);
8108: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]);
1.238 brouard 8109: }
1.264 brouard 8110: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8111: fprintf(ficgp,"\n#\n");
8112: if(invalidvarcomb[k1]){
8113: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8114: continue;
8115: }
8116:
8117: /* k=2+nlstate*(2*cpt-2); */
8118: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 8119: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 8120: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 8121: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 8122: 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 8123: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8124: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8125: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
8126: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8127: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8128: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 8129:
1.238 brouard 8130: */
8131: for (i=1; i< nlstate ; i ++) {
1.261 brouard 8132: 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 8133: /* 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 8134:
1.238 brouard 8135: }
1.261 brouard 8136: 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 8137: }
1.264 brouard 8138: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 8139: } /* end nres */
8140: } /* end kl 3eme */
1.126 brouard 8141:
1.223 brouard 8142: /* 4eme */
1.201 brouard 8143: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 8144: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
8145: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8146: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 8147: continue;
1.238 brouard 8148: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 8149: strcpy(gplotlabel,"(");
1.238 brouard 8150: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
8151: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8152: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
8153: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
1.238 brouard 8154: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8155: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8156: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8157: /* vlv= nbcode[Tvaraff[k]][lv]; */
8158: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.238 brouard 8159: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8160: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 8161: }
8162: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8163: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8164: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.238 brouard 8165: }
1.264 brouard 8166: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8167: fprintf(ficgp,"\n#\n");
8168: if(invalidvarcomb[k1]){
8169: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8170: continue;
1.223 brouard 8171: }
1.238 brouard 8172:
1.241 brouard 8173: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 8174: 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 8175: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8176: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8177: k=3;
8178: for (i=1; i<= nlstate ; i ++){
8179: if(i==1){
8180: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8181: }else{
8182: fprintf(ficgp,", '' ");
8183: }
8184: l=(nlstate+ndeath)*(i-1)+1;
8185: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8186: for (j=2; j<= nlstate+ndeath ; j ++)
8187: fprintf(ficgp,"+$%d",k+l+j-1);
8188: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
8189: } /* nlstate */
1.264 brouard 8190: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8191: } /* end cpt state*/
8192: } /* end nres */
8193: } /* end covariate k1 */
8194:
1.220 brouard 8195: /* 5eme */
1.201 brouard 8196: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 8197: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
8198: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8199: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 8200: continue;
1.238 brouard 8201: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 8202: strcpy(gplotlabel,"(");
1.238 brouard 8203: 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);
8204: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8205: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
8206: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
1.238 brouard 8207: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8208: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8209: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8210: /* vlv= nbcode[Tvaraff[k]][lv]; */
8211: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.238 brouard 8212: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8213: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 8214: }
8215: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8216: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8217: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.238 brouard 8218: }
1.264 brouard 8219: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8220: fprintf(ficgp,"\n#\n");
8221: if(invalidvarcomb[k1]){
8222: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8223: continue;
8224: }
1.227 brouard 8225:
1.241 brouard 8226: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8227: 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 8228: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8229: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8230: k=3;
8231: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8232: if(j==1)
8233: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8234: else
8235: fprintf(ficgp,", '' ");
8236: l=(nlstate+ndeath)*(cpt-1) +j;
8237: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8238: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8239: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8240: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8241: } /* nlstate */
8242: fprintf(ficgp,", '' ");
8243: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8244: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8245: l=(nlstate+ndeath)*(cpt-1) +j;
8246: if(j < nlstate)
8247: fprintf(ficgp,"$%d +",k+l);
8248: else
8249: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8250: }
1.264 brouard 8251: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8252: } /* end cpt state*/
8253: } /* end covariate */
8254: } /* end nres */
1.227 brouard 8255:
1.220 brouard 8256: /* 6eme */
1.202 brouard 8257: /* CV preval stable (period) for each covariate */
1.237 brouard 8258: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8259: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8260: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8261: continue;
1.255 brouard 8262: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8263: strcpy(gplotlabel,"(");
1.288 brouard 8264: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 8265: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8266: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8267: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8268: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8269: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8270: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8271: /* vlv= nbcode[Tvaraff[k]][lv]; */
8272: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8273: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8274: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 8275: }
1.237 brouard 8276: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8277: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8278: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8279: }
1.264 brouard 8280: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8281: fprintf(ficgp,"\n#\n");
1.223 brouard 8282: if(invalidvarcomb[k1]){
1.227 brouard 8283: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8284: continue;
1.223 brouard 8285: }
1.227 brouard 8286:
1.241 brouard 8287: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8288: 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 8289: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8290: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8291: k=3; /* Offset */
1.255 brouard 8292: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8293: if(i==1)
8294: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8295: else
8296: fprintf(ficgp,", '' ");
1.255 brouard 8297: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8298: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8299: for (j=2; j<= nlstate ; j ++)
8300: fprintf(ficgp,"+$%d",k+l+j-1);
8301: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8302: } /* nlstate */
1.264 brouard 8303: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8304: } /* end cpt state*/
8305: } /* end covariate */
1.227 brouard 8306:
8307:
1.220 brouard 8308: /* 7eme */
1.296 brouard 8309: if(prevbcast == 1){
1.288 brouard 8310: /* CV backward prevalence for each covariate */
1.237 brouard 8311: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8312: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8313: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8314: continue;
1.268 brouard 8315: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8316: strcpy(gplotlabel,"(");
1.288 brouard 8317: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8318: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8319: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8320: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8321: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8322: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 8323: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8324: /* vlv= nbcode[Tvaraff[k]][lv]; */
8325: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8326: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8327: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8328: }
1.237 brouard 8329: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8330: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8331: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8332: }
1.264 brouard 8333: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8334: fprintf(ficgp,"\n#\n");
8335: if(invalidvarcomb[k1]){
8336: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8337: continue;
8338: }
8339:
1.241 brouard 8340: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8341: 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 8342: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8343: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8344: k=3; /* Offset */
1.268 brouard 8345: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8346: if(i==1)
8347: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8348: else
8349: fprintf(ficgp,", '' ");
8350: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8351: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8352: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8353: /* 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 8354: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8355: /* for (j=2; j<= nlstate ; j ++) */
8356: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8357: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8358: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8359: } /* nlstate */
1.264 brouard 8360: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8361: } /* end cpt state*/
8362: } /* end covariate */
1.296 brouard 8363: } /* End if prevbcast */
1.218 brouard 8364:
1.223 brouard 8365: /* 8eme */
1.218 brouard 8366: if(prevfcast==1){
1.288 brouard 8367: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8368:
1.237 brouard 8369: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8370: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8371: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8372: continue;
1.211 brouard 8373: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8374: strcpy(gplotlabel,"(");
1.288 brouard 8375: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8376: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
1.332 brouard 8377: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8378: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8379: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8380: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8381: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8382: /* vlv= nbcode[Tvaraff[k]][lv]; */
8383: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8384: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8385: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8386: }
1.237 brouard 8387: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8388: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8389: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8390: }
1.264 brouard 8391: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8392: fprintf(ficgp,"\n#\n");
8393: if(invalidvarcomb[k1]){
8394: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8395: continue;
8396: }
8397:
8398: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8399: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8400: 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 8401: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8402: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8403:
8404: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8405: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8406: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8407: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8408: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8409: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8410: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8411: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8412: if(i==istart){
1.227 brouard 8413: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8414: }else{
8415: fprintf(ficgp,",\\\n '' ");
8416: }
8417: if(cptcoveff ==0){ /* No covariate */
8418: ioffset=2; /* Age is in 2 */
8419: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8420: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8421: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8422: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8423: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8424: if(i==nlstate+1){
1.270 brouard 8425: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8426: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8427: fprintf(ficgp,",\\\n '' ");
8428: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8429: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8430: offyear, \
1.268 brouard 8431: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8432: }else
1.227 brouard 8433: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8434: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8435: }else{ /* more than 2 covariates */
1.270 brouard 8436: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8437: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8438: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8439: iyearc=ioffset-1;
8440: iagec=ioffset;
1.227 brouard 8441: fprintf(ficgp," u %d:(",ioffset);
8442: kl=0;
8443: strcpy(gplotcondition,"(");
8444: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8445: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8446: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8447: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8448: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8449: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8450: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8451: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8452: kl++;
8453: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8454: kl++;
8455: if(k <cptcoveff && cptcoveff>1)
8456: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8457: }
8458: strcpy(gplotcondition+strlen(gplotcondition),")");
8459: /* 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 *\/ */
8460: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8461: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8462: /* '' 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*/
8463: if(i==nlstate+1){
1.270 brouard 8464: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8465: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8466: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8467: fprintf(ficgp," u %d:(",iagec);
8468: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8469: iyearc, iagec, offyear, \
8470: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8471: /* '' 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 8472: }else{
8473: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8474: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8475: }
8476: } /* end if covariate */
8477: } /* nlstate */
1.264 brouard 8478: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8479: } /* end cpt state*/
8480: } /* end covariate */
8481: } /* End if prevfcast */
1.227 brouard 8482:
1.296 brouard 8483: if(prevbcast==1){
1.268 brouard 8484: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8485:
8486: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8487: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8488: if(m != 1 && TKresult[nres]!= k1)
8489: continue;
8490: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8491: strcpy(gplotlabel,"(");
8492: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
8493: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
1.332 brouard 8494: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8495: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.268 brouard 8496: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8497: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8498: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8499: /* vlv= nbcode[Tvaraff[k]][lv]; */
8500: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.268 brouard 8501: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8502: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8503: }
8504: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8505: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8506: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.268 brouard 8507: }
8508: strcpy(gplotlabel+strlen(gplotlabel),")");
8509: fprintf(ficgp,"\n#\n");
8510: if(invalidvarcomb[k1]){
8511: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8512: continue;
8513: }
8514:
8515: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8516: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8517: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8518: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8519: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8520:
8521: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8522: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8523: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8524: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8525: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8526: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8527: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8528: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8529: if(i==istart){
8530: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8531: }else{
8532: fprintf(ficgp,",\\\n '' ");
8533: }
8534: if(cptcoveff ==0){ /* No covariate */
8535: ioffset=2; /* Age is in 2 */
8536: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8537: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8538: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8539: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8540: fprintf(ficgp," u %d:(", ioffset);
8541: if(i==nlstate+1){
1.270 brouard 8542: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8543: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8544: fprintf(ficgp,",\\\n '' ");
8545: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8546: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8547: offbyear, \
8548: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8549: }else
8550: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8551: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8552: }else{ /* more than 2 covariates */
1.270 brouard 8553: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8554: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8555: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8556: iyearc=ioffset-1;
8557: iagec=ioffset;
1.268 brouard 8558: fprintf(ficgp," u %d:(",ioffset);
8559: kl=0;
8560: strcpy(gplotcondition,"(");
8561: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8562: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8563: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.268 brouard 8564: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8565: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8566: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8567: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8568: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.268 brouard 8569: kl++;
8570: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8571: kl++;
8572: if(k <cptcoveff && cptcoveff>1)
8573: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8574: }
8575: strcpy(gplotcondition+strlen(gplotcondition),")");
8576: /* 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 *\/ */
8577: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8578: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8579: /* '' 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*/
8580: if(i==nlstate+1){
1.270 brouard 8581: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8582: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8583: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8584: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8585: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8586: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8587: iyearc,iagec,offbyear, \
8588: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8589: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8590: }else{
8591: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8592: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8593: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8594: }
8595: } /* end if covariate */
8596: } /* nlstate */
8597: fprintf(ficgp,"\nset out; unset label;\n");
8598: } /* end cpt state*/
8599: } /* end covariate */
1.296 brouard 8600: } /* End if prevbcast */
1.268 brouard 8601:
1.227 brouard 8602:
1.238 brouard 8603: /* 9eme writing MLE parameters */
8604: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8605: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8606: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8607: for(k=1; k <=(nlstate+ndeath); k++){
8608: if (k != i) {
1.227 brouard 8609: fprintf(ficgp,"# current state %d\n",k);
8610: for(j=1; j <=ncovmodel; j++){
8611: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8612: jk++;
8613: }
8614: fprintf(ficgp,"\n");
1.126 brouard 8615: }
8616: }
1.223 brouard 8617: }
1.187 brouard 8618: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8619:
1.145 brouard 8620: /*goto avoid;*/
1.238 brouard 8621: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8622: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8623: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8624: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8625: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8626: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8627: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8628: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8629: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8630: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8631: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8632: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8633: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8634: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8635: fprintf(ficgp,"#\n");
1.223 brouard 8636: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8637: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8638: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8639: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8640: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8641: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8642: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8643: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8644: continue;
1.264 brouard 8645: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8646: strcpy(gplotlabel,"(");
1.276 brouard 8647: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8648: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
1.332 brouard 8649: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8650: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.264 brouard 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 */
1.332 brouard 8654: /* vlv= nbcode[Tvaraff[k]][lv]; */
8655: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.264 brouard 8656: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8657: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8658: }
1.237 brouard 8659: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 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]]);
1.237 brouard 8662: }
1.264 brouard 8663: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8664: fprintf(ficgp,"\n#\n");
1.264 brouard 8665: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8666: fprintf(ficgp,"\nset key outside ");
8667: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8668: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8669: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8670: if (ng==1){
8671: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8672: fprintf(ficgp,"\nunset log y");
8673: }else if (ng==2){
8674: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8675: fprintf(ficgp,"\nset log y");
8676: }else if (ng==3){
8677: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8678: fprintf(ficgp,"\nset log y");
8679: }else
8680: fprintf(ficgp,"\nunset title ");
8681: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8682: i=1;
8683: for(k2=1; k2<=nlstate; k2++) {
8684: k3=i;
8685: for(k=1; k<=(nlstate+ndeath); k++) {
8686: if (k != k2){
8687: switch( ng) {
8688: case 1:
8689: if(nagesqr==0)
8690: fprintf(ficgp," p%d+p%d*x",i,i+1);
8691: else /* nagesqr =1 */
8692: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8693: break;
8694: case 2: /* ng=2 */
8695: if(nagesqr==0)
8696: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8697: else /* nagesqr =1 */
8698: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8699: break;
8700: case 3:
8701: if(nagesqr==0)
8702: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8703: else /* nagesqr =1 */
8704: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8705: break;
8706: }
8707: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8708: ijp=1; /* product no age */
8709: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8710: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8711: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 8712: switch(Typevar[j]){
8713: case 1:
8714: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8715: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
8716: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8717: if(DummyV[j]==0){/* Bug valgrind */
8718: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8719: }else{ /* quantitative */
8720: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8721: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8722: }
8723: ij++;
1.268 brouard 8724: }
1.237 brouard 8725: }
1.329 brouard 8726: }
8727: break;
8728: case 2:
8729: if(cptcovprod >0){
8730: if(j==Tprod[ijp]) { /* */
8731: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8732: if(ijp <=cptcovprod) { /* Product */
8733: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8734: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8735: /* 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)]); */
8736: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8737: }else{ /* Vn is dummy and Vm is quanti */
8738: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8739: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8740: }
8741: }else{ /* Vn*Vm Vn is quanti */
8742: if(DummyV[Tvard[ijp][2]]==0){
8743: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8744: }else{ /* Both quanti */
8745: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8746: }
1.268 brouard 8747: }
1.329 brouard 8748: ijp++;
1.237 brouard 8749: }
1.329 brouard 8750: } /* end Tprod */
8751: }
8752: break;
8753: case 0:
8754: /* simple covariate */
1.264 brouard 8755: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8756: if(Dummy[j]==0){
8757: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8758: }else{ /* quantitative */
8759: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8760: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8761: }
1.329 brouard 8762: /* end simple */
8763: break;
8764: default:
8765: break;
8766: } /* end switch */
1.237 brouard 8767: } /* end j */
1.329 brouard 8768: }else{ /* k=k2 */
8769: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
8770: fprintf(ficgp," (1.");i=i-ncovmodel;
8771: }else
8772: i=i-ncovmodel;
1.223 brouard 8773: }
1.227 brouard 8774:
1.223 brouard 8775: if(ng != 1){
8776: fprintf(ficgp,")/(1");
1.227 brouard 8777:
1.264 brouard 8778: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8779: if(nagesqr==0)
1.264 brouard 8780: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8781: else /* nagesqr =1 */
1.264 brouard 8782: 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 8783:
1.223 brouard 8784: ij=1;
1.329 brouard 8785: ijp=1;
8786: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
8787: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
8788: switch(Typevar[j]){
8789: case 1:
8790: if(cptcovage >0){
8791: if(j==Tage[ij]) { /* Bug valgrind */
8792: if(ij <=cptcovage) { /* Bug valgrind */
8793: if(DummyV[j]==0){/* Bug valgrind */
8794: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
8795: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
8796: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
8797: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
8798: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8799: }else{ /* quantitative */
8800: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8801: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8802: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8803: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8804: }
8805: ij++;
8806: }
8807: }
8808: }
8809: break;
8810: case 2:
8811: if(cptcovprod >0){
8812: if(j==Tprod[ijp]) { /* */
8813: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8814: if(ijp <=cptcovprod) { /* Product */
8815: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8816: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8817: /* 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)]); */
8818: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8819: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
8820: }else{ /* Vn is dummy and Vm is quanti */
8821: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8822: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8823: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8824: }
8825: }else{ /* Vn*Vm Vn is quanti */
8826: if(DummyV[Tvard[ijp][2]]==0){
8827: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8828: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
8829: }else{ /* Both quanti */
8830: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8831: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8832: }
8833: }
8834: ijp++;
8835: }
8836: } /* end Tprod */
8837: } /* end if */
8838: break;
8839: case 0:
8840: /* simple covariate */
8841: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
8842: if(Dummy[j]==0){
8843: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8844: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
8845: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8846: }else{ /* quantitative */
8847: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
8848: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
8849: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8850: }
8851: /* end simple */
8852: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
8853: break;
8854: default:
8855: break;
8856: } /* end switch */
1.223 brouard 8857: }
8858: fprintf(ficgp,")");
8859: }
8860: fprintf(ficgp,")");
8861: if(ng ==2)
1.276 brouard 8862: 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 8863: else /* ng= 3 */
1.276 brouard 8864: 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 8865: }else{ /* end ng <> 1 */
1.223 brouard 8866: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8867: 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 8868: }
8869: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8870: fprintf(ficgp,",");
8871: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8872: fprintf(ficgp,",");
8873: i=i+ncovmodel;
8874: } /* end k */
8875: } /* end k2 */
1.276 brouard 8876: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8877: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8878: } /* end k1 */
1.223 brouard 8879: } /* end ng */
8880: /* avoid: */
8881: fflush(ficgp);
1.126 brouard 8882: } /* end gnuplot */
8883:
8884:
8885: /*************** Moving average **************/
1.219 brouard 8886: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8887: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8888:
1.222 brouard 8889: int i, cpt, cptcod;
8890: int modcovmax =1;
8891: int mobilavrange, mob;
8892: int iage=0;
1.288 brouard 8893: int firstA1=0, firstA2=0;
1.222 brouard 8894:
1.266 brouard 8895: double sum=0., sumr=0.;
1.222 brouard 8896: double age;
1.266 brouard 8897: double *sumnewp, *sumnewm, *sumnewmr;
8898: double *agemingood, *agemaxgood;
8899: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8900:
8901:
1.278 brouard 8902: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8903: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8904:
8905: sumnewp = vector(1,ncovcombmax);
8906: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8907: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8908: agemingood = vector(1,ncovcombmax);
1.266 brouard 8909: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8910: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8911: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8912:
8913: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8914: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8915: sumnewp[cptcod]=0.;
1.266 brouard 8916: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8917: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8918: }
8919: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8920:
1.266 brouard 8921: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8922: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8923: else mobilavrange=mobilav;
8924: for (age=bage; age<=fage; age++)
8925: for (i=1; i<=nlstate;i++)
8926: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8927: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8928: /* We keep the original values on the extreme ages bage, fage and for
8929: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8930: we use a 5 terms etc. until the borders are no more concerned.
8931: */
8932: for (mob=3;mob <=mobilavrange;mob=mob+2){
8933: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8934: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8935: sumnewm[cptcod]=0.;
8936: for (i=1; i<=nlstate;i++){
1.222 brouard 8937: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8938: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8939: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8940: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8941: }
8942: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8943: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8944: } /* end i */
8945: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8946: } /* end cptcod */
1.222 brouard 8947: }/* end age */
8948: }/* end mob */
1.266 brouard 8949: }else{
8950: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8951: return -1;
1.266 brouard 8952: }
8953:
8954: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8955: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8956: if(invalidvarcomb[cptcod]){
8957: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8958: continue;
8959: }
1.219 brouard 8960:
1.266 brouard 8961: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8962: sumnewm[cptcod]=0.;
8963: sumnewmr[cptcod]=0.;
8964: for (i=1; i<=nlstate;i++){
8965: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8966: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8967: }
8968: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8969: agemingoodr[cptcod]=age;
8970: }
8971: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8972: agemingood[cptcod]=age;
8973: }
8974: } /* age */
8975: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8976: sumnewm[cptcod]=0.;
1.266 brouard 8977: sumnewmr[cptcod]=0.;
1.222 brouard 8978: for (i=1; i<=nlstate;i++){
8979: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8980: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8981: }
8982: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8983: agemaxgoodr[cptcod]=age;
1.222 brouard 8984: }
8985: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8986: agemaxgood[cptcod]=age;
8987: }
8988: } /* age */
8989: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8990: /* but they will change */
1.288 brouard 8991: firstA1=0;firstA2=0;
1.266 brouard 8992: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8993: sumnewm[cptcod]=0.;
8994: sumnewmr[cptcod]=0.;
8995: for (i=1; i<=nlstate;i++){
8996: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8997: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8998: }
8999: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9000: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9001: agemaxgoodr[cptcod]=age; /* age min */
9002: for (i=1; i<=nlstate;i++)
9003: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9004: }else{ /* bad we change the value with the values of good ages */
9005: for (i=1; i<=nlstate;i++){
9006: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
9007: } /* i */
9008: } /* end bad */
9009: }else{
9010: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9011: agemaxgood[cptcod]=age;
9012: }else{ /* bad we change the value with the values of good ages */
9013: for (i=1; i<=nlstate;i++){
9014: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
9015: } /* i */
9016: } /* end bad */
9017: }/* end else */
9018: sum=0.;sumr=0.;
9019: for (i=1; i<=nlstate;i++){
9020: sum+=mobaverage[(int)age][i][cptcod];
9021: sumr+=probs[(int)age][i][cptcod];
9022: }
9023: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 9024: if(!firstA1){
9025: firstA1=1;
9026: 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);
9027: }
9028: 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 9029: } /* end bad */
9030: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9031: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 9032: if(!firstA2){
9033: firstA2=1;
9034: 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);
9035: }
9036: 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 9037: } /* end bad */
9038: }/* age */
1.266 brouard 9039:
9040: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 9041: sumnewm[cptcod]=0.;
1.266 brouard 9042: sumnewmr[cptcod]=0.;
1.222 brouard 9043: for (i=1; i<=nlstate;i++){
9044: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9045: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9046: }
9047: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9048: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
9049: agemingoodr[cptcod]=age;
9050: for (i=1; i<=nlstate;i++)
9051: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9052: }else{ /* bad we change the value with the values of good ages */
9053: for (i=1; i<=nlstate;i++){
9054: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
9055: } /* i */
9056: } /* end bad */
9057: }else{
9058: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9059: agemingood[cptcod]=age;
9060: }else{ /* bad */
9061: for (i=1; i<=nlstate;i++){
9062: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
9063: } /* i */
9064: } /* end bad */
9065: }/* end else */
9066: sum=0.;sumr=0.;
9067: for (i=1; i<=nlstate;i++){
9068: sum+=mobaverage[(int)age][i][cptcod];
9069: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 9070: }
1.266 brouard 9071: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 9072: 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 9073: } /* end bad */
9074: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9075: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 9076: 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 9077: } /* end bad */
9078: }/* age */
1.266 brouard 9079:
1.222 brouard 9080:
9081: for (age=bage; age<=fage; age++){
1.235 brouard 9082: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 9083: sumnewp[cptcod]=0.;
9084: sumnewm[cptcod]=0.;
9085: for (i=1; i<=nlstate;i++){
9086: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
9087: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9088: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
9089: }
9090: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
9091: }
9092: /* printf("\n"); */
9093: /* } */
1.266 brouard 9094:
1.222 brouard 9095: /* brutal averaging */
1.266 brouard 9096: /* for (i=1; i<=nlstate;i++){ */
9097: /* for (age=1; age<=bage; age++){ */
9098: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
9099: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9100: /* } */
9101: /* for (age=fage; age<=AGESUP; age++){ */
9102: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
9103: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9104: /* } */
9105: /* } /\* end i status *\/ */
9106: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
9107: /* for (age=1; age<=AGESUP; age++){ */
9108: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
9109: /* mobaverage[(int)age][i][cptcod]=0.; */
9110: /* } */
9111: /* } */
1.222 brouard 9112: }/* end cptcod */
1.266 brouard 9113: free_vector(agemaxgoodr,1, ncovcombmax);
9114: free_vector(agemaxgood,1, ncovcombmax);
9115: free_vector(agemingood,1, ncovcombmax);
9116: free_vector(agemingoodr,1, ncovcombmax);
9117: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 9118: free_vector(sumnewm,1, ncovcombmax);
9119: free_vector(sumnewp,1, ncovcombmax);
9120: return 0;
9121: }/* End movingaverage */
1.218 brouard 9122:
1.126 brouard 9123:
1.296 brouard 9124:
1.126 brouard 9125: /************** Forecasting ******************/
1.296 brouard 9126: /* 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)*/
9127: 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){
9128: /* dateintemean, mean date of interviews
9129: dateprojd, year, month, day of starting projection
9130: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 9131: agemin, agemax range of age
9132: dateprev1 dateprev2 range of dates during which prevalence is computed
9133: */
1.296 brouard 9134: /* double anprojd, mprojd, jprojd; */
9135: /* double anprojf, mprojf, jprojf; */
1.267 brouard 9136: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 9137: double agec; /* generic age */
1.296 brouard 9138: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 9139: double *popeffectif,*popcount;
9140: double ***p3mat;
1.218 brouard 9141: /* double ***mobaverage; */
1.126 brouard 9142: char fileresf[FILENAMELENGTH];
9143:
9144: agelim=AGESUP;
1.211 brouard 9145: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9146: in each health status at the date of interview (if between dateprev1 and dateprev2).
9147: We still use firstpass and lastpass as another selection.
9148: */
1.214 brouard 9149: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9150: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 9151:
1.201 brouard 9152: strcpy(fileresf,"F_");
9153: strcat(fileresf,fileresu);
1.126 brouard 9154: if((ficresf=fopen(fileresf,"w"))==NULL) {
9155: printf("Problem with forecast resultfile: %s\n", fileresf);
9156: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
9157: }
1.235 brouard 9158: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
9159: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 9160:
1.225 brouard 9161: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 9162:
9163:
9164: stepsize=(int) (stepm+YEARM-1)/YEARM;
9165: if (stepm<=12) stepsize=1;
9166: if(estepm < stepm){
9167: printf ("Problem %d lower than %d\n",estepm, stepm);
9168: }
1.270 brouard 9169: else{
9170: hstepm=estepm;
9171: }
9172: if(estepm > stepm){ /* Yes every two year */
9173: stepsize=2;
9174: }
1.296 brouard 9175: hstepm=hstepm/stepm;
1.126 brouard 9176:
1.296 brouard 9177:
9178: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9179: /* fractional in yp1 *\/ */
9180: /* aintmean=yp; */
9181: /* yp2=modf((yp1*12),&yp); */
9182: /* mintmean=yp; */
9183: /* yp1=modf((yp2*30.5),&yp); */
9184: /* jintmean=yp; */
9185: /* if(jintmean==0) jintmean=1; */
9186: /* if(mintmean==0) mintmean=1; */
1.126 brouard 9187:
1.296 brouard 9188:
9189: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
9190: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
9191: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 9192: i1=pow(2,cptcoveff);
1.126 brouard 9193: if (cptcovn < 1){i1=1;}
9194:
1.296 brouard 9195: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 9196:
9197: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 9198:
1.126 brouard 9199: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 9200: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9201: 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 9202: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9203: continue;
1.227 brouard 9204: if(invalidvarcomb[k]){
9205: printf("\nCombination (%d) projection ignored because no cases \n",k);
9206: continue;
9207: }
9208: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
9209: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9210: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); */
9211: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 9212: }
1.235 brouard 9213: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 9214: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 9215: }
1.227 brouard 9216: fprintf(ficresf," yearproj age");
9217: for(j=1; j<=nlstate+ndeath;j++){
9218: for(i=1; i<=nlstate;i++)
9219: fprintf(ficresf," p%d%d",i,j);
9220: fprintf(ficresf," wp.%d",j);
9221: }
1.296 brouard 9222: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 9223: fprintf(ficresf,"\n");
1.296 brouard 9224: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 9225: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
9226: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 9227: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
9228: nhstepm = nhstepm/hstepm;
9229: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9230: oldm=oldms;savm=savms;
1.268 brouard 9231: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 9232: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 9233: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 9234: for (h=0; h<=nhstepm; h++){
9235: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 9236: break;
9237: }
9238: }
9239: fprintf(ficresf,"\n");
9240: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9241: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
9242: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* TnsdVar[Tvaraff] correct */
1.296 brouard 9243: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9244:
9245: for(j=1; j<=nlstate+ndeath;j++) {
9246: ppij=0.;
9247: for(i=1; i<=nlstate;i++) {
1.278 brouard 9248: if (mobilav>=1)
9249: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9250: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9251: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9252: }
1.268 brouard 9253: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9254: } /* end i */
9255: fprintf(ficresf," %.3f", ppij);
9256: }/* end j */
1.227 brouard 9257: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9258: } /* end agec */
1.266 brouard 9259: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9260: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9261: } /* end yearp */
9262: } /* end k */
1.219 brouard 9263:
1.126 brouard 9264: fclose(ficresf);
1.215 brouard 9265: printf("End of Computing forecasting \n");
9266: fprintf(ficlog,"End of Computing forecasting\n");
9267:
1.126 brouard 9268: }
9269:
1.269 brouard 9270: /************** Back Forecasting ******************/
1.296 brouard 9271: /* 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){ */
9272: 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){
9273: /* back1, year, month, day of starting backprojection
1.267 brouard 9274: agemin, agemax range of age
9275: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9276: anback2 year of end of backprojection (same day and month as back1).
9277: prevacurrent and prev are prevalences.
1.267 brouard 9278: */
9279: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9280: double agec; /* generic age */
1.302 brouard 9281: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9282: double *popeffectif,*popcount;
9283: double ***p3mat;
9284: /* double ***mobaverage; */
9285: char fileresfb[FILENAMELENGTH];
9286:
1.268 brouard 9287: agelim=AGEINF;
1.267 brouard 9288: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9289: in each health status at the date of interview (if between dateprev1 and dateprev2).
9290: We still use firstpass and lastpass as another selection.
9291: */
9292: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9293: /* firstpass, lastpass, stepm, weightopt, model); */
9294:
9295: /*Do we need to compute prevalence again?*/
9296:
9297: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9298:
9299: strcpy(fileresfb,"FB_");
9300: strcat(fileresfb,fileresu);
9301: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9302: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9303: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9304: }
9305: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9306: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9307:
9308: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9309:
9310:
9311: stepsize=(int) (stepm+YEARM-1)/YEARM;
9312: if (stepm<=12) stepsize=1;
9313: if(estepm < stepm){
9314: printf ("Problem %d lower than %d\n",estepm, stepm);
9315: }
1.270 brouard 9316: else{
9317: hstepm=estepm;
9318: }
9319: if(estepm >= stepm){ /* Yes every two year */
9320: stepsize=2;
9321: }
1.267 brouard 9322:
9323: hstepm=hstepm/stepm;
1.296 brouard 9324: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9325: /* fractional in yp1 *\/ */
9326: /* aintmean=yp; */
9327: /* yp2=modf((yp1*12),&yp); */
9328: /* mintmean=yp; */
9329: /* yp1=modf((yp2*30.5),&yp); */
9330: /* jintmean=yp; */
9331: /* if(jintmean==0) jintmean=1; */
9332: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9333:
9334: i1=pow(2,cptcoveff);
9335: if (cptcovn < 1){i1=1;}
9336:
1.296 brouard 9337: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9338: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9339:
9340: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9341:
9342: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9343: for(k=1; k<=i1;k++){
9344: if(i1 != 1 && TKresult[nres]!= k)
9345: continue;
9346: if(invalidvarcomb[k]){
9347: printf("\nCombination (%d) projection ignored because no cases \n",k);
9348: continue;
9349: }
1.268 brouard 9350: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9351: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9352: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.267 brouard 9353: }
9354: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9355: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9356: }
9357: fprintf(ficresfb," yearbproj age");
9358: for(j=1; j<=nlstate+ndeath;j++){
9359: for(i=1; i<=nlstate;i++)
1.268 brouard 9360: fprintf(ficresfb," b%d%d",i,j);
9361: fprintf(ficresfb," b.%d",j);
1.267 brouard 9362: }
1.296 brouard 9363: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9364: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9365: fprintf(ficresfb,"\n");
1.296 brouard 9366: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9367: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9368: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9369: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9370: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9371: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9372: nhstepm = nhstepm/hstepm;
9373: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9374: oldm=oldms;savm=savms;
1.268 brouard 9375: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9376: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9377: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9378: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9379: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9380: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9381: for (h=0; h<=nhstepm; h++){
1.268 brouard 9382: if (h*hstepm/YEARM*stepm ==-yearp) {
9383: break;
9384: }
9385: }
9386: fprintf(ficresfb,"\n");
9387: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9388: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.296 brouard 9389: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9390: for(i=1; i<=nlstate+ndeath;i++) {
9391: ppij=0.;ppi=0.;
9392: for(j=1; j<=nlstate;j++) {
9393: /* if (mobilav==1) */
1.269 brouard 9394: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9395: ppi=ppi+prevacurrent[(int)agec][j][k];
9396: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9397: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9398: /* else { */
9399: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9400: /* } */
1.268 brouard 9401: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9402: } /* end j */
9403: if(ppi <0.99){
9404: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9405: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9406: }
9407: fprintf(ficresfb," %.3f", ppij);
9408: }/* end j */
1.267 brouard 9409: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9410: } /* end agec */
9411: } /* end yearp */
9412: } /* end k */
1.217 brouard 9413:
1.267 brouard 9414: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9415:
1.267 brouard 9416: fclose(ficresfb);
9417: printf("End of Computing Back forecasting \n");
9418: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9419:
1.267 brouard 9420: }
1.217 brouard 9421:
1.269 brouard 9422: /* Variance of prevalence limit: varprlim */
9423: 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 9424: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9425:
9426: char fileresvpl[FILENAMELENGTH];
9427: FILE *ficresvpl;
9428: double **oldm, **savm;
9429: double **varpl; /* Variances of prevalence limits by age */
9430: int i1, k, nres, j ;
9431:
9432: strcpy(fileresvpl,"VPL_");
9433: strcat(fileresvpl,fileresu);
9434: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9435: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9436: exit(0);
9437: }
1.288 brouard 9438: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9439: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9440:
9441: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9442: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9443:
9444: i1=pow(2,cptcoveff);
9445: if (cptcovn < 1){i1=1;}
9446:
9447: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9448: for(k=1; k<=i1;k++){ /* We find the combination equivalent to result line values of dummies */
1.269 brouard 9449: if(i1 != 1 && TKresult[nres]!= k)
9450: continue;
9451: fprintf(ficresvpl,"\n#****** ");
9452: printf("\n#****** ");
9453: fprintf(ficlog,"\n#****** ");
9454: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9455: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9456: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9457: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.269 brouard 9458: }
9459: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 9460: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9461: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9462: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.269 brouard 9463: }
9464: fprintf(ficresvpl,"******\n");
9465: printf("******\n");
9466: fprintf(ficlog,"******\n");
9467:
9468: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9469: oldm=oldms;savm=savms;
9470: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9471: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9472: /*}*/
9473: }
9474:
9475: fclose(ficresvpl);
1.288 brouard 9476: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9477: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9478:
9479: }
9480: /* Variance of back prevalence: varbprlim */
9481: 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){
9482: /*------- Variance of back (stable) prevalence------*/
9483:
9484: char fileresvbl[FILENAMELENGTH];
9485: FILE *ficresvbl;
9486:
9487: double **oldm, **savm;
9488: double **varbpl; /* Variances of back prevalence limits by age */
9489: int i1, k, nres, j ;
9490:
9491: strcpy(fileresvbl,"VBL_");
9492: strcat(fileresvbl,fileresu);
9493: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9494: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9495: exit(0);
9496: }
9497: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9498: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9499:
9500:
9501: i1=pow(2,cptcoveff);
9502: if (cptcovn < 1){i1=1;}
9503:
9504: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9505: for(k=1; k<=i1;k++){
9506: if(i1 != 1 && TKresult[nres]!= k)
9507: continue;
9508: fprintf(ficresvbl,"\n#****** ");
9509: printf("\n#****** ");
9510: fprintf(ficlog,"\n#****** ");
9511: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9512: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9513: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9514: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.269 brouard 9515: }
9516: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 9517: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9518: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9519: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.269 brouard 9520: }
9521: fprintf(ficresvbl,"******\n");
9522: printf("******\n");
9523: fprintf(ficlog,"******\n");
9524:
9525: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9526: oldm=oldms;savm=savms;
9527:
9528: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9529: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9530: /*}*/
9531: }
9532:
9533: fclose(ficresvbl);
9534: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9535: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9536:
9537: } /* End of varbprlim */
9538:
1.126 brouard 9539: /************** Forecasting *****not tested NB*************/
1.227 brouard 9540: /* 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 9541:
1.227 brouard 9542: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9543: /* int *popage; */
9544: /* double calagedatem, agelim, kk1, kk2; */
9545: /* double *popeffectif,*popcount; */
9546: /* double ***p3mat,***tabpop,***tabpopprev; */
9547: /* /\* double ***mobaverage; *\/ */
9548: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9549:
1.227 brouard 9550: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9551: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9552: /* agelim=AGESUP; */
9553: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9554:
1.227 brouard 9555: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9556:
9557:
1.227 brouard 9558: /* strcpy(filerespop,"POP_"); */
9559: /* strcat(filerespop,fileresu); */
9560: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9561: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9562: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9563: /* } */
9564: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9565: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9566:
1.227 brouard 9567: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9568:
1.227 brouard 9569: /* /\* if (mobilav!=0) { *\/ */
9570: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9571: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9572: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9573: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9574: /* /\* } *\/ */
9575: /* /\* } *\/ */
1.126 brouard 9576:
1.227 brouard 9577: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9578: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9579:
1.227 brouard 9580: /* agelim=AGESUP; */
1.126 brouard 9581:
1.227 brouard 9582: /* hstepm=1; */
9583: /* hstepm=hstepm/stepm; */
1.218 brouard 9584:
1.227 brouard 9585: /* if (popforecast==1) { */
9586: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9587: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9588: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9589: /* } */
9590: /* popage=ivector(0,AGESUP); */
9591: /* popeffectif=vector(0,AGESUP); */
9592: /* popcount=vector(0,AGESUP); */
1.126 brouard 9593:
1.227 brouard 9594: /* i=1; */
9595: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9596:
1.227 brouard 9597: /* imx=i; */
9598: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9599: /* } */
1.218 brouard 9600:
1.227 brouard 9601: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9602: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9603: /* k=k+1; */
9604: /* fprintf(ficrespop,"\n#******"); */
9605: /* for(j=1;j<=cptcoveff;j++) { */
9606: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9607: /* } */
9608: /* fprintf(ficrespop,"******\n"); */
9609: /* fprintf(ficrespop,"# Age"); */
9610: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9611: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9612:
1.227 brouard 9613: /* for (cpt=0; cpt<=0;cpt++) { */
9614: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9615:
1.227 brouard 9616: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9617: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9618: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9619:
1.227 brouard 9620: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9621: /* oldm=oldms;savm=savms; */
9622: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9623:
1.227 brouard 9624: /* for (h=0; h<=nhstepm; h++){ */
9625: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9626: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9627: /* } */
9628: /* for(j=1; j<=nlstate+ndeath;j++) { */
9629: /* kk1=0.;kk2=0; */
9630: /* for(i=1; i<=nlstate;i++) { */
9631: /* if (mobilav==1) */
9632: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9633: /* else { */
9634: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9635: /* } */
9636: /* } */
9637: /* if (h==(int)(calagedatem+12*cpt)){ */
9638: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9639: /* /\*fprintf(ficrespop," %.3f", kk1); */
9640: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9641: /* } */
9642: /* } */
9643: /* for(i=1; i<=nlstate;i++){ */
9644: /* kk1=0.; */
9645: /* for(j=1; j<=nlstate;j++){ */
9646: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9647: /* } */
9648: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9649: /* } */
1.218 brouard 9650:
1.227 brouard 9651: /* if (h==(int)(calagedatem+12*cpt)) */
9652: /* for(j=1; j<=nlstate;j++) */
9653: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9654: /* } */
9655: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9656: /* } */
9657: /* } */
1.218 brouard 9658:
1.227 brouard 9659: /* /\******\/ */
1.218 brouard 9660:
1.227 brouard 9661: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9662: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9663: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9664: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9665: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9666:
1.227 brouard 9667: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9668: /* oldm=oldms;savm=savms; */
9669: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9670: /* for (h=0; h<=nhstepm; h++){ */
9671: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9672: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9673: /* } */
9674: /* for(j=1; j<=nlstate+ndeath;j++) { */
9675: /* kk1=0.;kk2=0; */
9676: /* for(i=1; i<=nlstate;i++) { */
9677: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9678: /* } */
9679: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9680: /* } */
9681: /* } */
9682: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9683: /* } */
9684: /* } */
9685: /* } */
9686: /* } */
1.218 brouard 9687:
1.227 brouard 9688: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9689:
1.227 brouard 9690: /* if (popforecast==1) { */
9691: /* free_ivector(popage,0,AGESUP); */
9692: /* free_vector(popeffectif,0,AGESUP); */
9693: /* free_vector(popcount,0,AGESUP); */
9694: /* } */
9695: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9696: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9697: /* fclose(ficrespop); */
9698: /* } /\* End of popforecast *\/ */
1.218 brouard 9699:
1.126 brouard 9700: int fileappend(FILE *fichier, char *optionfich)
9701: {
9702: if((fichier=fopen(optionfich,"a"))==NULL) {
9703: printf("Problem with file: %s\n", optionfich);
9704: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9705: return (0);
9706: }
9707: fflush(fichier);
9708: return (1);
9709: }
9710:
9711:
9712: /**************** function prwizard **********************/
9713: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9714: {
9715:
9716: /* Wizard to print covariance matrix template */
9717:
1.164 brouard 9718: char ca[32], cb[32];
9719: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9720: int numlinepar;
9721:
9722: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9723: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9724: for(i=1; i <=nlstate; i++){
9725: jj=0;
9726: for(j=1; j <=nlstate+ndeath; j++){
9727: if(j==i) continue;
9728: jj++;
9729: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9730: printf("%1d%1d",i,j);
9731: fprintf(ficparo,"%1d%1d",i,j);
9732: for(k=1; k<=ncovmodel;k++){
9733: /* printf(" %lf",param[i][j][k]); */
9734: /* fprintf(ficparo," %lf",param[i][j][k]); */
9735: printf(" 0.");
9736: fprintf(ficparo," 0.");
9737: }
9738: printf("\n");
9739: fprintf(ficparo,"\n");
9740: }
9741: }
9742: printf("# Scales (for hessian or gradient estimation)\n");
9743: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9744: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9745: for(i=1; i <=nlstate; i++){
9746: jj=0;
9747: for(j=1; j <=nlstate+ndeath; j++){
9748: if(j==i) continue;
9749: jj++;
9750: fprintf(ficparo,"%1d%1d",i,j);
9751: printf("%1d%1d",i,j);
9752: fflush(stdout);
9753: for(k=1; k<=ncovmodel;k++){
9754: /* printf(" %le",delti3[i][j][k]); */
9755: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9756: printf(" 0.");
9757: fprintf(ficparo," 0.");
9758: }
9759: numlinepar++;
9760: printf("\n");
9761: fprintf(ficparo,"\n");
9762: }
9763: }
9764: printf("# Covariance matrix\n");
9765: /* # 121 Var(a12)\n\ */
9766: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9767: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9768: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9769: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9770: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9771: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9772: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9773: fflush(stdout);
9774: fprintf(ficparo,"# Covariance matrix\n");
9775: /* # 121 Var(a12)\n\ */
9776: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9777: /* # ...\n\ */
9778: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9779:
9780: for(itimes=1;itimes<=2;itimes++){
9781: jj=0;
9782: for(i=1; i <=nlstate; i++){
9783: for(j=1; j <=nlstate+ndeath; j++){
9784: if(j==i) continue;
9785: for(k=1; k<=ncovmodel;k++){
9786: jj++;
9787: ca[0]= k+'a'-1;ca[1]='\0';
9788: if(itimes==1){
9789: printf("#%1d%1d%d",i,j,k);
9790: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9791: }else{
9792: printf("%1d%1d%d",i,j,k);
9793: fprintf(ficparo,"%1d%1d%d",i,j,k);
9794: /* printf(" %.5le",matcov[i][j]); */
9795: }
9796: ll=0;
9797: for(li=1;li <=nlstate; li++){
9798: for(lj=1;lj <=nlstate+ndeath; lj++){
9799: if(lj==li) continue;
9800: for(lk=1;lk<=ncovmodel;lk++){
9801: ll++;
9802: if(ll<=jj){
9803: cb[0]= lk +'a'-1;cb[1]='\0';
9804: if(ll<jj){
9805: if(itimes==1){
9806: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9807: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9808: }else{
9809: printf(" 0.");
9810: fprintf(ficparo," 0.");
9811: }
9812: }else{
9813: if(itimes==1){
9814: printf(" Var(%s%1d%1d)",ca,i,j);
9815: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9816: }else{
9817: printf(" 0.");
9818: fprintf(ficparo," 0.");
9819: }
9820: }
9821: }
9822: } /* end lk */
9823: } /* end lj */
9824: } /* end li */
9825: printf("\n");
9826: fprintf(ficparo,"\n");
9827: numlinepar++;
9828: } /* end k*/
9829: } /*end j */
9830: } /* end i */
9831: } /* end itimes */
9832:
9833: } /* end of prwizard */
9834: /******************* Gompertz Likelihood ******************************/
9835: double gompertz(double x[])
9836: {
1.302 brouard 9837: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9838: int i,n=0; /* n is the size of the sample */
9839:
1.220 brouard 9840: for (i=1;i<=imx ; i++) {
1.126 brouard 9841: sump=sump+weight[i];
9842: /* sump=sump+1;*/
9843: num=num+1;
9844: }
1.302 brouard 9845: L=0.0;
9846: /* agegomp=AGEGOMP; */
1.126 brouard 9847: /* for (i=0; i<=imx; i++)
9848: 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]);*/
9849:
1.302 brouard 9850: for (i=1;i<=imx ; i++) {
9851: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9852: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9853: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9854: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9855: * +
9856: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9857: */
9858: if (wav[i] > 1 || agedc[i] < AGESUP) {
9859: if (cens[i] == 1){
9860: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9861: } else if (cens[i] == 0){
1.126 brouard 9862: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9863: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9864: } else
9865: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9866: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9867: L=L+A*weight[i];
1.126 brouard 9868: /* 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 9869: }
9870: }
1.126 brouard 9871:
1.302 brouard 9872: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9873:
9874: return -2*L*num/sump;
9875: }
9876:
1.136 brouard 9877: #ifdef GSL
9878: /******************* Gompertz_f Likelihood ******************************/
9879: double gompertz_f(const gsl_vector *v, void *params)
9880: {
1.302 brouard 9881: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9882: double *x= (double *) v->data;
9883: int i,n=0; /* n is the size of the sample */
9884:
9885: for (i=0;i<=imx-1 ; i++) {
9886: sump=sump+weight[i];
9887: /* sump=sump+1;*/
9888: num=num+1;
9889: }
9890:
9891:
9892: /* for (i=0; i<=imx; i++)
9893: 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]);*/
9894: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9895: for (i=1;i<=imx ; i++)
9896: {
9897: if (cens[i] == 1 && wav[i]>1)
9898: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9899:
9900: if (cens[i] == 0 && wav[i]>1)
9901: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9902: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9903:
9904: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9905: if (wav[i] > 1 ) { /* ??? */
9906: LL=LL+A*weight[i];
9907: /* 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]);*/
9908: }
9909: }
9910:
9911: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9912: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9913:
9914: return -2*LL*num/sump;
9915: }
9916: #endif
9917:
1.126 brouard 9918: /******************* Printing html file ***********/
1.201 brouard 9919: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9920: int lastpass, int stepm, int weightopt, char model[],\
9921: int imx, double p[],double **matcov,double agemortsup){
9922: int i,k;
9923:
9924: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9925: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9926: for (i=1;i<=2;i++)
9927: 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 9928: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9929: fprintf(fichtm,"</ul>");
9930:
9931: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9932:
9933: 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>");
9934:
9935: for (k=agegomp;k<(agemortsup-2);k++)
9936: 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]);
9937:
9938:
9939: fflush(fichtm);
9940: }
9941:
9942: /******************* Gnuplot file **************/
1.201 brouard 9943: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9944:
9945: char dirfileres[132],optfileres[132];
1.164 brouard 9946:
1.126 brouard 9947: int ng;
9948:
9949:
9950: /*#ifdef windows */
9951: fprintf(ficgp,"cd \"%s\" \n",pathc);
9952: /*#endif */
9953:
9954:
9955: strcpy(dirfileres,optionfilefiname);
9956: strcpy(optfileres,"vpl");
1.199 brouard 9957: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9958: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9959: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9960: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9961: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9962:
9963: }
9964:
1.136 brouard 9965: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9966: {
1.126 brouard 9967:
1.136 brouard 9968: /*-------- data file ----------*/
9969: FILE *fic;
9970: char dummy[]=" ";
1.240 brouard 9971: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9972: int lstra;
1.136 brouard 9973: int linei, month, year,iout;
1.302 brouard 9974: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9975: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9976: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9977: char *stratrunc;
1.223 brouard 9978:
1.240 brouard 9979: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9980: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 9981: for(v=1;v<NCOVMAX;v++){
9982: DummyV[v]=0;
9983: FixedV[v]=0;
9984: }
1.126 brouard 9985:
1.240 brouard 9986: for(v=1; v <=ncovcol;v++){
9987: DummyV[v]=0;
9988: FixedV[v]=0;
9989: }
9990: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9991: DummyV[v]=1;
9992: FixedV[v]=0;
9993: }
9994: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9995: DummyV[v]=0;
9996: FixedV[v]=1;
9997: }
9998: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9999: DummyV[v]=1;
10000: FixedV[v]=1;
10001: }
10002: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
10003: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10004: 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]);
10005: }
1.126 brouard 10006:
1.136 brouard 10007: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 10008: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10009: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 10010: }
1.126 brouard 10011:
1.302 brouard 10012: /* Is it a BOM UTF-8 Windows file? */
10013: /* First data line */
10014: linei=0;
10015: while(fgets(line, MAXLINE, fic)) {
10016: noffset=0;
10017: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10018: {
10019: noffset=noffset+3;
10020: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
10021: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
10022: fflush(ficlog); return 1;
10023: }
10024: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
10025: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
10026: {
10027: noffset=noffset+2;
1.304 brouard 10028: 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);
10029: 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 10030: fflush(ficlog); return 1;
10031: }
10032: else if( line[0] == 0 && line[1] == 0)
10033: {
10034: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10035: noffset=noffset+4;
1.304 brouard 10036: 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);
10037: 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 10038: fflush(ficlog); return 1;
10039: }
10040: } else{
10041: ;/*printf(" Not a BOM file\n");*/
10042: }
10043: /* If line starts with a # it is a comment */
10044: if (line[noffset] == '#') {
10045: linei=linei+1;
10046: break;
10047: }else{
10048: break;
10049: }
10050: }
10051: fclose(fic);
10052: if((fic=fopen(datafile,"r"))==NULL) {
10053: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10054: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
10055: }
10056: /* Not a Bom file */
10057:
1.136 brouard 10058: i=1;
10059: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
10060: linei=linei+1;
10061: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
10062: if(line[j] == '\t')
10063: line[j] = ' ';
10064: }
10065: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
10066: ;
10067: };
10068: line[j+1]=0; /* Trims blanks at end of line */
10069: if(line[0]=='#'){
10070: fprintf(ficlog,"Comment line\n%s\n",line);
10071: printf("Comment line\n%s\n",line);
10072: continue;
10073: }
10074: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 10075: strcpy(line, linetmp);
1.223 brouard 10076:
10077: /* Loops on waves */
10078: for (j=maxwav;j>=1;j--){
10079: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 10080: cutv(stra, strb, line, ' ');
10081: if(strb[0]=='.') { /* Missing value */
10082: lval=-1;
10083: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
10084: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
10085: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
10086: 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);
10087: 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);
10088: return 1;
10089: }
10090: }else{
10091: errno=0;
10092: /* what_kind_of_number(strb); */
10093: dval=strtod(strb,&endptr);
10094: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
10095: /* if(strb != endptr && *endptr == '\0') */
10096: /* dval=dlval; */
10097: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10098: if( strb[0]=='\0' || (*endptr != '\0')){
10099: 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);
10100: 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);
10101: return 1;
10102: }
10103: cotqvar[j][iv][i]=dval;
10104: cotvar[j][ntv+iv][i]=dval;
10105: }
10106: strcpy(line,stra);
1.223 brouard 10107: }/* end loop ntqv */
1.225 brouard 10108:
1.223 brouard 10109: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 10110: cutv(stra, strb, line, ' ');
10111: if(strb[0]=='.') { /* Missing value */
10112: lval=-1;
10113: }else{
10114: errno=0;
10115: lval=strtol(strb,&endptr,10);
10116: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10117: if( strb[0]=='\0' || (*endptr != '\0')){
10118: 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);
10119: 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);
10120: return 1;
10121: }
10122: }
10123: if(lval <-1 || lval >1){
10124: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10125: 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 10126: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10127: For example, for multinomial values like 1, 2 and 3,\n \
10128: build V1=0 V2=0 for the reference value (1),\n \
10129: V1=1 V2=0 for (2) \n \
1.223 brouard 10130: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10131: output of IMaCh is often meaningless.\n \
1.319 brouard 10132: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 10133: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10134: 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 10135: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10136: For example, for multinomial values like 1, 2 and 3,\n \
10137: build V1=0 V2=0 for the reference value (1),\n \
10138: V1=1 V2=0 for (2) \n \
1.223 brouard 10139: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10140: output of IMaCh is often meaningless.\n \
1.319 brouard 10141: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 10142: return 1;
10143: }
10144: cotvar[j][iv][i]=(double)(lval);
10145: strcpy(line,stra);
1.223 brouard 10146: }/* end loop ntv */
1.225 brouard 10147:
1.223 brouard 10148: /* Statuses at wave */
1.137 brouard 10149: cutv(stra, strb, line, ' ');
1.223 brouard 10150: if(strb[0]=='.') { /* Missing value */
1.238 brouard 10151: lval=-1;
1.136 brouard 10152: }else{
1.238 brouard 10153: errno=0;
10154: lval=strtol(strb,&endptr,10);
10155: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10156: if( strb[0]=='\0' || (*endptr != '\0')){
10157: 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);
10158: 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);
10159: return 1;
10160: }
1.136 brouard 10161: }
1.225 brouard 10162:
1.136 brouard 10163: s[j][i]=lval;
1.225 brouard 10164:
1.223 brouard 10165: /* Date of Interview */
1.136 brouard 10166: strcpy(line,stra);
10167: cutv(stra, strb,line,' ');
1.169 brouard 10168: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10169: }
1.169 brouard 10170: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 10171: month=99;
10172: year=9999;
1.136 brouard 10173: }else{
1.225 brouard 10174: 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);
10175: 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);
10176: return 1;
1.136 brouard 10177: }
10178: anint[j][i]= (double) year;
1.302 brouard 10179: mint[j][i]= (double)month;
10180: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
10181: /* 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]); */
10182: /* 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]); */
10183: /* } */
1.136 brouard 10184: strcpy(line,stra);
1.223 brouard 10185: } /* End loop on waves */
1.225 brouard 10186:
1.223 brouard 10187: /* Date of death */
1.136 brouard 10188: cutv(stra, strb,line,' ');
1.169 brouard 10189: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10190: }
1.169 brouard 10191: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 10192: month=99;
10193: year=9999;
10194: }else{
1.141 brouard 10195: 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 10196: 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);
10197: return 1;
1.136 brouard 10198: }
10199: andc[i]=(double) year;
10200: moisdc[i]=(double) month;
10201: strcpy(line,stra);
10202:
1.223 brouard 10203: /* Date of birth */
1.136 brouard 10204: cutv(stra, strb,line,' ');
1.169 brouard 10205: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10206: }
1.169 brouard 10207: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 10208: month=99;
10209: year=9999;
10210: }else{
1.141 brouard 10211: 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);
10212: 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 10213: return 1;
1.136 brouard 10214: }
10215: if (year==9999) {
1.141 brouard 10216: 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);
10217: 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 10218: return 1;
10219:
1.136 brouard 10220: }
10221: annais[i]=(double)(year);
1.302 brouard 10222: moisnais[i]=(double)(month);
10223: for (j=1;j<=maxwav;j++){
10224: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
10225: 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]);
10226: 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]);
10227: }
10228: }
10229:
1.136 brouard 10230: strcpy(line,stra);
1.225 brouard 10231:
1.223 brouard 10232: /* Sample weight */
1.136 brouard 10233: cutv(stra, strb,line,' ');
10234: errno=0;
10235: dval=strtod(strb,&endptr);
10236: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 10237: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
10238: 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 10239: fflush(ficlog);
10240: return 1;
10241: }
10242: weight[i]=dval;
10243: strcpy(line,stra);
1.225 brouard 10244:
1.223 brouard 10245: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10246: cutv(stra, strb, line, ' ');
10247: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10248: lval=-1;
1.311 brouard 10249: coqvar[iv][i]=NAN;
10250: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10251: }else{
1.225 brouard 10252: errno=0;
10253: /* what_kind_of_number(strb); */
10254: dval=strtod(strb,&endptr);
10255: /* if(strb != endptr && *endptr == '\0') */
10256: /* dval=dlval; */
10257: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10258: if( strb[0]=='\0' || (*endptr != '\0')){
10259: 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);
10260: 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);
10261: return 1;
10262: }
10263: coqvar[iv][i]=dval;
1.226 brouard 10264: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10265: }
10266: strcpy(line,stra);
10267: }/* end loop nqv */
1.136 brouard 10268:
1.223 brouard 10269: /* Covariate values */
1.136 brouard 10270: for (j=ncovcol;j>=1;j--){
10271: cutv(stra, strb,line,' ');
1.223 brouard 10272: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10273: lval=-1;
1.136 brouard 10274: }else{
1.225 brouard 10275: errno=0;
10276: lval=strtol(strb,&endptr,10);
10277: if( strb[0]=='\0' || (*endptr != '\0')){
10278: 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);
10279: 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);
10280: return 1;
10281: }
1.136 brouard 10282: }
10283: if(lval <-1 || lval >1){
1.225 brouard 10284: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10285: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10286: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10287: For example, for multinomial values like 1, 2 and 3,\n \
10288: build V1=0 V2=0 for the reference value (1),\n \
10289: V1=1 V2=0 for (2) \n \
1.136 brouard 10290: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10291: output of IMaCh is often meaningless.\n \
1.136 brouard 10292: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10293: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10294: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10295: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10296: For example, for multinomial values like 1, 2 and 3,\n \
10297: build V1=0 V2=0 for the reference value (1),\n \
10298: V1=1 V2=0 for (2) \n \
1.136 brouard 10299: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10300: output of IMaCh is often meaningless.\n \
1.136 brouard 10301: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10302: return 1;
1.136 brouard 10303: }
10304: covar[j][i]=(double)(lval);
10305: strcpy(line,stra);
10306: }
10307: lstra=strlen(stra);
1.225 brouard 10308:
1.136 brouard 10309: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10310: stratrunc = &(stra[lstra-9]);
10311: num[i]=atol(stratrunc);
10312: }
10313: else
10314: num[i]=atol(stra);
10315: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10316: 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;}*/
10317:
10318: i=i+1;
10319: } /* End loop reading data */
1.225 brouard 10320:
1.136 brouard 10321: *imax=i-1; /* Number of individuals */
10322: fclose(fic);
1.225 brouard 10323:
1.136 brouard 10324: return (0);
1.164 brouard 10325: /* endread: */
1.225 brouard 10326: printf("Exiting readdata: ");
10327: fclose(fic);
10328: return (1);
1.223 brouard 10329: }
1.126 brouard 10330:
1.234 brouard 10331: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10332: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10333: while (*p2 == ' ')
1.234 brouard 10334: p2++;
10335: /* while ((*p1++ = *p2++) !=0) */
10336: /* ; */
10337: /* do */
10338: /* while (*p2 == ' ') */
10339: /* p2++; */
10340: /* while (*p1++ == *p2++); */
10341: *stri=p2;
1.145 brouard 10342: }
10343:
1.330 brouard 10344: int decoderesult( char resultline[], int nres)
1.230 brouard 10345: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10346: {
1.235 brouard 10347: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10348: char resultsav[MAXLINE];
1.330 brouard 10349: /* int resultmodel[MAXLINE]; */
1.334 brouard 10350: /* int modelresult[MAXLINE]; */
1.230 brouard 10351: char stra[80], strb[80], strc[80], strd[80],stre[80];
10352:
1.234 brouard 10353: removefirstspace(&resultline);
1.332 brouard 10354: printf("decoderesult:%s\n",resultline);
1.230 brouard 10355:
1.332 brouard 10356: strcpy(resultsav,resultline);
10357: printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline);
1.230 brouard 10358: if (strlen(resultsav) >1){
1.334 brouard 10359: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 10360: }
1.253 brouard 10361: if(j == 0){ /* Resultline but no = */
10362: TKresult[nres]=0; /* Combination for the nresult and the model */
10363: return (0);
10364: }
1.234 brouard 10365: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.334 brouard 10366: 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);
10367: 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 10368: /* return 1;*/
1.234 brouard 10369: }
1.334 brouard 10370: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 10371: if(nbocc(resultsav,'=') >1){
1.318 brouard 10372: 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 10373: /* 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 10374: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 10375: /* If a blank, then strc="V4=" and strd='\0' */
10376: if(strc[0]=='\0'){
10377: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
10378: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
10379: return 1;
10380: }
1.234 brouard 10381: }else
10382: cutl(strc,strd,resultsav,'=');
1.318 brouard 10383: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10384:
1.230 brouard 10385: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10386: 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 10387: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10388: /* cptcovsel++; */
10389: if (nbocc(stra,'=') >0)
10390: strcpy(resultsav,stra); /* and analyzes it */
10391: }
1.235 brouard 10392: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10393: /* 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 10394: 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 10395: if(Typevar[k1]==0){ /* Single covariate in model */
10396: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 10397: match=0;
1.318 brouard 10398: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10399: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10400: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10401: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10402: break;
10403: }
10404: }
10405: if(match == 0){
1.332 brouard 10406: printf("Error in result line (Dummy single): V%d is missing in result: %s according to model=%s. Tvar[k1=%d]=%d is different from Tvarsel[k2=%d]=%d.\n",Tvar[k1], resultline, model,k1, Tvar[k1], k2, Tvarsel[k2]);
10407: fprintf(ficlog,"Error in result line (Dummy single): V%d is missing in result: %s according to model=%s\n",Tvar[k1], resultline, model);
1.310 brouard 10408: return 1;
1.234 brouard 10409: }
1.332 brouard 10410: }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*/
10411: /* We feed resultmodel[k1]=k2; */
10412: match=0;
10413: 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 */
10414: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10415: 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 10416: resultmodel[nres][k1]=k2; /* Added here */
10417: printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]);
10418: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10419: break;
10420: }
10421: }
10422: if(match == 0){
10423: printf("Error in result line (Product with age): V%d is missing in result: %s according to model=%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
1.333 brouard 10424: fprintf(ficlog,"Error in result line (Product with age): V%d is missing in result: %s according to model=%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
1.332 brouard 10425: return 1;
10426: }
10427: }else if(Typevar[k1]==2){ /* Product No age We want to get the position in the resultline of the product in the model line*/
10428: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
10429: match=0;
10430: 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]);
10431: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10432: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10433: /* modelresult[k2]=k1; */
10434: printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]);
10435: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10436: }
10437: }
10438: if(match == 0){
10439: printf("Error in result line (Product without age first variable): V%d is missing in result: %s according to model=%s\n",Tvardk[k1][1], resultline, model);
1.333 brouard 10440: fprintf(ficlog,"Error in result line (Product without age first variable): V%d is missing in result: %s according to model=%s\n",Tvardk[k1][1], resultline, model);
1.332 brouard 10441: return 1;
10442: }
10443: match=0;
10444: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10445: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10446: /* modelresult[k2]=k1;*/
10447: printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]);
10448: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10449: break;
10450: }
10451: }
10452: if(match == 0){
10453: printf("Error in result line (Product without age second variable): V%d is missing in result: %s according to model=%s\n",Tvardk[k1][2], resultline, model);
1.333 brouard 10454: fprintf(ficlog,"Error in result line (Product without age second variable): V%d is missing in result : %s according to model=%s\n",Tvardk[k1][2], resultline, model);
1.332 brouard 10455: return 1;
10456: }
10457: }/* End of testing */
1.333 brouard 10458: }/* End loop cptcovt */
1.235 brouard 10459: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10460: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 10461: 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)
10462: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 10463: match=0;
1.318 brouard 10464: 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 10465: if(Typevar[k1]==0){ /* Single only */
1.237 brouard 10466: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 brouard 10467: 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 10468: 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 10469: ++match;
10470: }
10471: }
10472: }
10473: if(match == 0){
1.332 brouard 10474: printf("Error in result line: variable V%d is missing in model; result: %s, model=%s\n",Tvarsel[k2], resultline, model);
10475: fprintf(ficlog,"Error in result line: variable V%d is missing in model; result: %s, model=%s\n",Tvarsel[k2], resultline, model);
1.310 brouard 10476: return 1;
1.234 brouard 10477: }else if(match > 1){
10478: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 10479: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
10480: return 1;
1.234 brouard 10481: }
10482: }
1.334 brouard 10483: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 10484: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10485: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 10486: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10487: /* 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*/
10488: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 10489: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10490: /* 1 0 0 0 */
10491: /* 2 1 0 0 */
10492: /* 3 0 1 0 */
1.330 brouard 10493: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 10494: /* 5 0 0 1 */
1.330 brouard 10495: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 10496: /* 7 0 1 1 */
10497: /* 8 1 1 1 */
1.237 brouard 10498: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10499: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10500: /* V5*age V5 known which value for nres? */
10501: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 10502: 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.
10503: * loop on position k1 in the MODEL LINE */
1.331 brouard 10504: /* k counting number of combination of single dummies in the equation model */
10505: /* k4 counting single dummies in the equation model */
10506: /* k4q counting single quantitatives in the equation model */
1.334 brouard 10507: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Dummy and Single, k1 is sorting according to MODEL, but k3 to resultline */
10508: /* 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 10509: /* 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 10510: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 10511: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
10512: /* k3 is the position in the nres result line of the k1th variable of the model equation */
10513: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
10514: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
10515: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 10516: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 10517: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 10518: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 10519: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
10520: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
10521: 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 10522: 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 10523: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 10524: /* Tinvresult[nres][4]=1 */
1.334 brouard 10525: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
10526: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
10527: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10528: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 10529: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 10530: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.332 brouard 10531: 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 10532: k4++;;
1.331 brouard 10533: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 10534: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 10535: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 10536: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 10537: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
10538: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
10539: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 10540: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
10541: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10542: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
10543: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10544: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10545: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 10546: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 10547: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 10548: precov[nres][k1]=Tvalsel[k3q];
10549: 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 10550: k4q++;;
1.331 brouard 10551: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
10552: /* Tvar[k1]; */ /* Age variable */
1.332 brouard 10553: /* Wrong we want the value of variable name Tvar[k1] */
10554:
10555: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
1.331 brouard 10556: 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 10557: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
1.332 brouard 10558: precov[nres][k1]=Tvalsel[k3];
10559: 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 10560: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.332 brouard 10561: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.331 brouard 10562: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 10563: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
1.332 brouard 10564: precov[nres][k1]=Tvalsel[k3q];
1.334 brouard 10565: 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 10566: }else if(Typevar[k1]==2 ){ /* For product quant or dummy (not with age) */
1.332 brouard 10567: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
10568: 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 10569: }else{
1.332 brouard 10570: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
10571: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 10572: }
10573: }
1.234 brouard 10574:
1.334 brouard 10575: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 10576: return (0);
10577: }
1.235 brouard 10578:
1.230 brouard 10579: int decodemodel( char model[], int lastobs)
10580: /**< This routine decodes the model and returns:
1.224 brouard 10581: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10582: * - nagesqr = 1 if age*age in the model, otherwise 0.
10583: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10584: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10585: * - cptcovage number of covariates with age*products =2
10586: * - cptcovs number of simple covariates
10587: * - 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
10588: * which is a new column after the 9 (ncovcol) variables.
1.319 brouard 10589: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10590: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10591: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10592: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10593: */
1.319 brouard 10594: /* 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 10595: {
1.238 brouard 10596: int i, j, k, ks, v;
1.227 brouard 10597: int j1, k1, k2, k3, k4;
1.136 brouard 10598: char modelsav[80];
1.145 brouard 10599: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10600: char *strpt;
1.136 brouard 10601:
1.145 brouard 10602: /*removespace(model);*/
1.136 brouard 10603: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10604: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10605: if (strstr(model,"AGE") !=0){
1.192 brouard 10606: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10607: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10608: return 1;
10609: }
1.141 brouard 10610: if (strstr(model,"v") !=0){
10611: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
10612: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
10613: return 1;
10614: }
1.187 brouard 10615: strcpy(modelsav,model);
10616: if ((strpt=strstr(model,"age*age")) !=0){
10617: printf(" strpt=%s, model=%s\n",strpt, model);
10618: if(strpt != model){
1.234 brouard 10619: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10620: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10621: corresponding column of parameters.\n",model);
1.234 brouard 10622: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10623: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10624: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10625: return 1;
1.225 brouard 10626: }
1.187 brouard 10627: nagesqr=1;
10628: if (strstr(model,"+age*age") !=0)
1.234 brouard 10629: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10630: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10631: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10632: else
1.234 brouard 10633: substrchaine(modelsav, model, "age*age");
1.187 brouard 10634: }else
10635: nagesqr=0;
10636: if (strlen(modelsav) >1){
10637: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10638: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10639: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10640: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10641: * cst, age and age*age
10642: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10643: /* including age products which are counted in cptcovage.
10644: * but the covariates which are products must be treated
10645: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10646: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10647: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10648:
10649:
1.187 brouard 10650: /* Design
10651: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10652: * < ncovcol=8 >
10653: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10654: * k= 1 2 3 4 5 6 7 8
10655: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10656: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10657: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10658: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10659: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10660: * Tage[++cptcovage]=k
10661: * if products, new covar are created after ncovcol with k1
10662: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10663: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10664: * 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
10665: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10666: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10667: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10668: * < ncovcol=8 >
10669: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10670: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10671: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10672: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10673: * p Tprod[1]@2={ 6, 5}
10674: *p Tvard[1][1]@4= {7, 8, 5, 6}
10675: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10676: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10677: *How to reorganize? Tvars(orted)
1.187 brouard 10678: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10679: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
10680: * {2, 1, 4, 8, 5, 6, 3, 7}
10681: * Struct []
10682: */
1.225 brouard 10683:
1.187 brouard 10684: /* This loop fills the array Tvar from the string 'model'.*/
10685: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
10686: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
10687: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
10688: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
10689: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
10690: /* k=1 Tvar[1]=2 (from V2) */
10691: /* k=5 Tvar[5] */
10692: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 10693: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 10694: /* } */
1.198 brouard 10695: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 10696: /*
10697: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 10698: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
10699: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
10700: }
1.187 brouard 10701: cptcovage=0;
1.319 brouard 10702: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
10703: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
10704: 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" */
10705: if (nbocc(modelsav,'+')==0)
10706: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 10707: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
10708: /*scanf("%d",i);*/
1.319 brouard 10709: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
10710: 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 10711: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
10712: /* covar is not filled and then is empty */
10713: cptcovprod--;
10714: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 10715: 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 10716: Typevar[k]=1; /* 1 for age product */
1.319 brouard 10717: cptcovage++; /* Counts the number of covariates which include age as a product */
10718: 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 10719: /*printf("stre=%s ", stre);*/
10720: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
10721: cptcovprod--;
10722: cutl(stre,strb,strc,'V');
10723: Tvar[k]=atoi(stre);
10724: Typevar[k]=1; /* 1 for age product */
10725: cptcovage++;
10726: Tage[cptcovage]=k;
10727: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
10728: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
10729: cptcovn++;
10730: cptcovprodnoage++;k1++;
10731: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10732: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10733: because this model-covariate is a construction we invent a new column
10734: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.335 ! brouard 10735: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
1.319 brouard 10736: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
10737: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=4 etc */
1.335 ! brouard 10738: /* Please remark that the new variables are model dependent */
! 10739: /* If we have 4 variable but the model uses only 3, like in
! 10740: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
! 10741: * k= 1 2 3 4 5 6 7 8
! 10742: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
! 10743: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
! 10744: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
! 10745: */
1.234 brouard 10746: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10747: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10748: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 10749: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 10750: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 brouard 10751: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 10752: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 brouard 10753: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 10754: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10755: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10756: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10757: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10758: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10759: for (i=1; i<=lastobs;i++){
10760: /* Computes the new covariate which is a product of
10761: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10762: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10763: }
10764: } /* End age is not in the model */
10765: } /* End if model includes a product */
1.319 brouard 10766: else { /* not a product */
1.234 brouard 10767: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10768: /* scanf("%d",i);*/
10769: cutl(strd,strc,strb,'V');
10770: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10771: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10772: Tvar[k]=atoi(strd);
10773: Typevar[k]=0; /* 0 for simple covariates */
10774: }
10775: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10776: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10777: scanf("%d",i);*/
1.187 brouard 10778: } /* end of loop + on total covariates */
10779: } /* end if strlen(modelsave == 0) age*age might exist */
10780: } /* end if strlen(model == 0) */
1.136 brouard 10781:
10782: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10783: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10784:
1.136 brouard 10785: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10786: printf("cptcovprod=%d ", cptcovprod);
10787: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10788: scanf("%d ",i);*/
10789:
10790:
1.230 brouard 10791: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10792: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10793: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10794: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10795: k = 1 2 3 4 5 6 7 8 9
10796: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 10797: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 10798: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10799: Dummy[k] 1 0 0 0 3 1 1 2 3
10800: Tmodelind[combination of covar]=k;
1.225 brouard 10801: */
10802: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10803: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10804: /* 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 10805: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 10806: printf("Model=1+age+%s\n\
1.227 brouard 10807: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10808: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10809: 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 10810: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 10811: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10812: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10813: 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);
1.285 brouard 10814: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10815: for(k=1, ncovf=0, nsd=0, nsq=0, ncovv=0, ncova=0, ncoveff=0, nqfveff=0, ntveff=0, nqtveff=0;k<=cptcovt; k++){ /* or cptocvt */
10816: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10817: Fixed[k]= 0;
10818: Dummy[k]= 0;
1.225 brouard 10819: ncoveff++;
1.232 brouard 10820: ncovf++;
1.234 brouard 10821: nsd++;
10822: modell[k].maintype= FTYPE;
10823: TvarsD[nsd]=Tvar[k];
10824: TvarsDind[nsd]=k;
1.330 brouard 10825: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 10826: TvarF[ncovf]=Tvar[k];
10827: TvarFind[ncovf]=k;
10828: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10829: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10830: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10831: Fixed[k]= 0;
10832: Dummy[k]= 0;
10833: ncoveff++;
10834: ncovf++;
10835: modell[k].maintype= FTYPE;
10836: TvarF[ncovf]=Tvar[k];
1.330 brouard 10837: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 10838: TvarFind[ncovf]=k;
1.230 brouard 10839: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10840: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10841: }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 10842: Fixed[k]= 0;
10843: Dummy[k]= 1;
1.230 brouard 10844: nqfveff++;
1.234 brouard 10845: modell[k].maintype= FTYPE;
10846: modell[k].subtype= FQ;
10847: nsq++;
1.334 brouard 10848: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
10849: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 10850: ncovf++;
1.234 brouard 10851: TvarF[ncovf]=Tvar[k];
10852: TvarFind[ncovf]=k;
1.231 brouard 10853: 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 10854: 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 10855: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10856: Fixed[k]= 1;
10857: Dummy[k]= 0;
1.225 brouard 10858: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10859: modell[k].maintype= VTYPE;
10860: modell[k].subtype= VD;
10861: nsd++;
10862: TvarsD[nsd]=Tvar[k];
10863: TvarsDind[nsd]=k;
1.330 brouard 10864: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 10865: ncovv++; /* Only simple time varying variables */
10866: TvarV[ncovv]=Tvar[k];
1.242 brouard 10867: 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 10868: 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 */
10869: 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 10870: 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);
10871: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10872: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10873: Fixed[k]= 1;
10874: Dummy[k]= 1;
10875: nqtveff++;
10876: modell[k].maintype= VTYPE;
10877: modell[k].subtype= VQ;
10878: ncovv++; /* Only simple time varying variables */
10879: nsq++;
1.334 brouard 10880: 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) */
10881: 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 10882: TvarV[ncovv]=Tvar[k];
1.242 brouard 10883: 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 10884: 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 */
10885: 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 10886: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10887: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10888: 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);
1.228 brouard 10889: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10890: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10891: ncova++;
10892: TvarA[ncova]=Tvar[k];
10893: TvarAind[ncova]=k;
1.231 brouard 10894: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10895: Fixed[k]= 2;
10896: Dummy[k]= 2;
10897: modell[k].maintype= ATYPE;
10898: modell[k].subtype= APFD;
10899: /* ncoveff++; */
1.227 brouard 10900: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10901: Fixed[k]= 2;
10902: Dummy[k]= 3;
10903: modell[k].maintype= ATYPE;
10904: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10905: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10906: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10907: Fixed[k]= 3;
10908: Dummy[k]= 2;
10909: modell[k].maintype= ATYPE;
10910: modell[k].subtype= APVD; /* Product age * varying dummy */
10911: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10912: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10913: Fixed[k]= 3;
10914: Dummy[k]= 3;
10915: modell[k].maintype= ATYPE;
10916: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10917: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10918: }
10919: }else if (Typevar[k] == 2) { /* product without age */
10920: k1=Tposprod[k];
10921: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10922: if(Tvard[k1][2] <=ncovcol){
10923: Fixed[k]= 1;
10924: Dummy[k]= 0;
10925: modell[k].maintype= FTYPE;
10926: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10927: ncovf++; /* Fixed variables without age */
10928: TvarF[ncovf]=Tvar[k];
10929: TvarFind[ncovf]=k;
10930: }else if(Tvard[k1][2] <=ncovcol+nqv){
10931: Fixed[k]= 0; /* or 2 ?*/
10932: Dummy[k]= 1;
10933: modell[k].maintype= FTYPE;
10934: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10935: ncovf++; /* Varying variables without age */
10936: TvarF[ncovf]=Tvar[k];
10937: TvarFind[ncovf]=k;
10938: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10939: Fixed[k]= 1;
10940: Dummy[k]= 0;
10941: modell[k].maintype= VTYPE;
10942: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10943: ncovv++; /* Varying variables without age */
10944: TvarV[ncovv]=Tvar[k];
10945: TvarVind[ncovv]=k;
10946: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10947: Fixed[k]= 1;
10948: Dummy[k]= 1;
10949: modell[k].maintype= VTYPE;
10950: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10951: ncovv++; /* Varying variables without age */
10952: TvarV[ncovv]=Tvar[k];
10953: TvarVind[ncovv]=k;
10954: }
1.227 brouard 10955: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10956: if(Tvard[k1][2] <=ncovcol){
10957: Fixed[k]= 0; /* or 2 ?*/
10958: Dummy[k]= 1;
10959: modell[k].maintype= FTYPE;
10960: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10961: ncovf++; /* Fixed variables without age */
10962: TvarF[ncovf]=Tvar[k];
10963: TvarFind[ncovf]=k;
10964: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10965: Fixed[k]= 1;
10966: Dummy[k]= 1;
10967: modell[k].maintype= VTYPE;
10968: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10969: ncovv++; /* Varying variables without age */
10970: TvarV[ncovv]=Tvar[k];
10971: TvarVind[ncovv]=k;
10972: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10973: Fixed[k]= 1;
10974: Dummy[k]= 1;
10975: modell[k].maintype= VTYPE;
10976: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10977: ncovv++; /* Varying variables without age */
10978: TvarV[ncovv]=Tvar[k];
10979: TvarVind[ncovv]=k;
10980: ncovv++; /* Varying variables without age */
10981: TvarV[ncovv]=Tvar[k];
10982: TvarVind[ncovv]=k;
10983: }
1.227 brouard 10984: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10985: if(Tvard[k1][2] <=ncovcol){
10986: Fixed[k]= 1;
10987: Dummy[k]= 1;
10988: modell[k].maintype= VTYPE;
10989: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10990: ncovv++; /* Varying variables without age */
10991: TvarV[ncovv]=Tvar[k];
10992: TvarVind[ncovv]=k;
10993: }else if(Tvard[k1][2] <=ncovcol+nqv){
10994: Fixed[k]= 1;
10995: Dummy[k]= 1;
10996: modell[k].maintype= VTYPE;
10997: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10998: ncovv++; /* Varying variables without age */
10999: TvarV[ncovv]=Tvar[k];
11000: TvarVind[ncovv]=k;
11001: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11002: Fixed[k]= 1;
11003: Dummy[k]= 0;
11004: modell[k].maintype= VTYPE;
11005: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
11006: ncovv++; /* Varying variables without age */
11007: TvarV[ncovv]=Tvar[k];
11008: TvarVind[ncovv]=k;
11009: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11010: Fixed[k]= 1;
11011: Dummy[k]= 1;
11012: modell[k].maintype= VTYPE;
11013: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
11014: ncovv++; /* Varying variables without age */
11015: TvarV[ncovv]=Tvar[k];
11016: TvarVind[ncovv]=k;
11017: }
1.227 brouard 11018: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 11019: if(Tvard[k1][2] <=ncovcol){
11020: Fixed[k]= 1;
11021: Dummy[k]= 1;
11022: modell[k].maintype= VTYPE;
11023: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
11024: ncovv++; /* Varying variables without age */
11025: TvarV[ncovv]=Tvar[k];
11026: TvarVind[ncovv]=k;
11027: }else if(Tvard[k1][2] <=ncovcol+nqv){
11028: Fixed[k]= 1;
11029: Dummy[k]= 1;
11030: modell[k].maintype= VTYPE;
11031: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
11032: ncovv++; /* Varying variables without age */
11033: TvarV[ncovv]=Tvar[k];
11034: TvarVind[ncovv]=k;
11035: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11036: Fixed[k]= 1;
11037: Dummy[k]= 1;
11038: modell[k].maintype= VTYPE;
11039: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
11040: ncovv++; /* Varying variables without age */
11041: TvarV[ncovv]=Tvar[k];
11042: TvarVind[ncovv]=k;
11043: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11044: Fixed[k]= 1;
11045: Dummy[k]= 1;
11046: modell[k].maintype= VTYPE;
11047: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
11048: ncovv++; /* Varying variables without age */
11049: TvarV[ncovv]=Tvar[k];
11050: TvarVind[ncovv]=k;
11051: }
1.227 brouard 11052: }else{
1.240 brouard 11053: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11054: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11055: } /*end k1*/
1.225 brouard 11056: }else{
1.226 brouard 11057: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
11058: 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 11059: }
1.227 brouard 11060: 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]);
1.231 brouard 11061: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 11062: 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]);
11063: }
11064: /* Searching for doublons in the model */
11065: for(k1=1; k1<= cptcovt;k1++){
11066: for(k2=1; k2 <k1;k2++){
1.285 brouard 11067: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
11068: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 11069: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
11070: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 11071: printf("Error duplication in the model=%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]);
11072: fprintf(ficlog,"Error duplication in the model=%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 11073: return(1);
11074: }
11075: }else if (Typevar[k1] ==2){
11076: k3=Tposprod[k1];
11077: k4=Tposprod[k2];
11078: 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])) ){
11079: printf("Error duplication in the model=%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]]);
11080: fprintf(ficlog,"Error duplication in the model=%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);
11081: return(1);
11082: }
11083: }
1.227 brouard 11084: }
11085: }
1.225 brouard 11086: }
11087: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
11088: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 11089: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
11090: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 11091: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 11092: /*endread:*/
1.225 brouard 11093: printf("Exiting decodemodel: ");
11094: return (1);
1.136 brouard 11095: }
11096:
1.169 brouard 11097: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 11098: {/* Check ages at death */
1.136 brouard 11099: int i, m;
1.218 brouard 11100: int firstone=0;
11101:
1.136 brouard 11102: for (i=1; i<=imx; i++) {
11103: for(m=2; (m<= maxwav); m++) {
11104: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
11105: anint[m][i]=9999;
1.216 brouard 11106: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
11107: s[m][i]=-1;
1.136 brouard 11108: }
11109: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 11110: *nberr = *nberr + 1;
1.218 brouard 11111: if(firstone == 0){
11112: firstone=1;
1.260 brouard 11113: 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 11114: }
1.262 brouard 11115: 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 11116: s[m][i]=-1; /* Droping the death status */
1.136 brouard 11117: }
11118: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 11119: (*nberr)++;
1.259 brouard 11120: 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 11121: 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 11122: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 11123: }
11124: }
11125: }
11126:
11127: for (i=1; i<=imx; i++) {
11128: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
11129: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 11130: 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 11131: if (s[m][i] >= nlstate+1) {
1.169 brouard 11132: if(agedc[i]>0){
11133: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 11134: agev[m][i]=agedc[i];
1.214 brouard 11135: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 11136: }else {
1.136 brouard 11137: if ((int)andc[i]!=9999){
11138: nbwarn++;
11139: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
11140: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
11141: agev[m][i]=-1;
11142: }
11143: }
1.169 brouard 11144: } /* agedc > 0 */
1.214 brouard 11145: } /* end if */
1.136 brouard 11146: else if(s[m][i] !=9){ /* Standard case, age in fractional
11147: years but with the precision of a month */
11148: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
11149: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
11150: agev[m][i]=1;
11151: else if(agev[m][i] < *agemin){
11152: *agemin=agev[m][i];
11153: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
11154: }
11155: else if(agev[m][i] >*agemax){
11156: *agemax=agev[m][i];
1.156 brouard 11157: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 11158: }
11159: /*agev[m][i]=anint[m][i]-annais[i];*/
11160: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 11161: } /* en if 9*/
1.136 brouard 11162: else { /* =9 */
1.214 brouard 11163: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 11164: agev[m][i]=1;
11165: s[m][i]=-1;
11166: }
11167: }
1.214 brouard 11168: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 11169: agev[m][i]=1;
1.214 brouard 11170: else{
11171: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11172: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11173: agev[m][i]=0;
11174: }
11175: } /* End for lastpass */
11176: }
1.136 brouard 11177:
11178: for (i=1; i<=imx; i++) {
11179: for(m=firstpass; (m<=lastpass); m++){
11180: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 11181: (*nberr)++;
1.136 brouard 11182: 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);
11183: 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);
11184: return 1;
11185: }
11186: }
11187: }
11188:
11189: /*for (i=1; i<=imx; i++){
11190: for (m=firstpass; (m<lastpass); m++){
11191: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
11192: }
11193:
11194: }*/
11195:
11196:
1.139 brouard 11197: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
11198: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 11199:
11200: return (0);
1.164 brouard 11201: /* endread:*/
1.136 brouard 11202: printf("Exiting calandcheckages: ");
11203: return (1);
11204: }
11205:
1.172 brouard 11206: #if defined(_MSC_VER)
11207: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11208: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11209: //#include "stdafx.h"
11210: //#include <stdio.h>
11211: //#include <tchar.h>
11212: //#include <windows.h>
11213: //#include <iostream>
11214: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
11215:
11216: LPFN_ISWOW64PROCESS fnIsWow64Process;
11217:
11218: BOOL IsWow64()
11219: {
11220: BOOL bIsWow64 = FALSE;
11221:
11222: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
11223: // (HANDLE, PBOOL);
11224:
11225: //LPFN_ISWOW64PROCESS fnIsWow64Process;
11226:
11227: HMODULE module = GetModuleHandle(_T("kernel32"));
11228: const char funcName[] = "IsWow64Process";
11229: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
11230: GetProcAddress(module, funcName);
11231:
11232: if (NULL != fnIsWow64Process)
11233: {
11234: if (!fnIsWow64Process(GetCurrentProcess(),
11235: &bIsWow64))
11236: //throw std::exception("Unknown error");
11237: printf("Unknown error\n");
11238: }
11239: return bIsWow64 != FALSE;
11240: }
11241: #endif
1.177 brouard 11242:
1.191 brouard 11243: void syscompilerinfo(int logged)
1.292 brouard 11244: {
11245: #include <stdint.h>
11246:
11247: /* #include "syscompilerinfo.h"*/
1.185 brouard 11248: /* command line Intel compiler 32bit windows, XP compatible:*/
11249: /* /GS /W3 /Gy
11250: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
11251: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
11252: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 11253: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
11254: */
11255: /* 64 bits */
1.185 brouard 11256: /*
11257: /GS /W3 /Gy
11258: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
11259: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
11260: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
11261: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
11262: /* Optimization are useless and O3 is slower than O2 */
11263: /*
11264: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
11265: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
11266: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
11267: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
11268: */
1.186 brouard 11269: /* Link is */ /* /OUT:"visual studio
1.185 brouard 11270: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
11271: /PDB:"visual studio
11272: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
11273: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
11274: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
11275: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
11276: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
11277: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
11278: uiAccess='false'"
11279: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
11280: /NOLOGO /TLBID:1
11281: */
1.292 brouard 11282:
11283:
1.177 brouard 11284: #if defined __INTEL_COMPILER
1.178 brouard 11285: #if defined(__GNUC__)
11286: struct utsname sysInfo; /* For Intel on Linux and OS/X */
11287: #endif
1.177 brouard 11288: #elif defined(__GNUC__)
1.179 brouard 11289: #ifndef __APPLE__
1.174 brouard 11290: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 11291: #endif
1.177 brouard 11292: struct utsname sysInfo;
1.178 brouard 11293: int cross = CROSS;
11294: if (cross){
11295: printf("Cross-");
1.191 brouard 11296: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 11297: }
1.174 brouard 11298: #endif
11299:
1.191 brouard 11300: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 11301: #if defined(__clang__)
1.191 brouard 11302: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 11303: #endif
11304: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 11305: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 11306: #endif
11307: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 11308: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 11309: #endif
11310: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 11311: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 11312: #endif
11313: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 11314: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 11315: #endif
11316: #if defined(_MSC_VER)
1.191 brouard 11317: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 11318: #endif
11319: #if defined(__PGI)
1.191 brouard 11320: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 11321: #endif
11322: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 11323: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 11324: #endif
1.191 brouard 11325: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 11326:
1.167 brouard 11327: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
11328: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
11329: // Windows (x64 and x86)
1.191 brouard 11330: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 11331: #elif __unix__ // all unices, not all compilers
11332: // Unix
1.191 brouard 11333: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 11334: #elif __linux__
11335: // linux
1.191 brouard 11336: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 11337: #elif __APPLE__
1.174 brouard 11338: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 11339: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11340: #endif
11341:
11342: /* __MINGW32__ */
11343: /* __CYGWIN__ */
11344: /* __MINGW64__ */
11345: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11346: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11347: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11348: /* _WIN64 // Defined for applications for Win64. */
11349: /* _M_X64 // Defined for compilations that target x64 processors. */
11350: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11351:
1.167 brouard 11352: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11353: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11354: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11355: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11356: #else
1.191 brouard 11357: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11358: #endif
11359:
1.169 brouard 11360: #if defined(__GNUC__)
11361: # if defined(__GNUC_PATCHLEVEL__)
11362: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11363: + __GNUC_MINOR__ * 100 \
11364: + __GNUC_PATCHLEVEL__)
11365: # else
11366: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11367: + __GNUC_MINOR__ * 100)
11368: # endif
1.174 brouard 11369: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11370: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11371:
11372: if (uname(&sysInfo) != -1) {
11373: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11374: 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 11375: }
11376: else
11377: perror("uname() error");
1.179 brouard 11378: //#ifndef __INTEL_COMPILER
11379: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11380: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11381: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11382: #endif
1.169 brouard 11383: #endif
1.172 brouard 11384:
1.286 brouard 11385: // void main ()
1.172 brouard 11386: // {
1.169 brouard 11387: #if defined(_MSC_VER)
1.174 brouard 11388: if (IsWow64()){
1.191 brouard 11389: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11390: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11391: }
11392: else{
1.191 brouard 11393: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11394: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11395: }
1.172 brouard 11396: // printf("\nPress Enter to continue...");
11397: // getchar();
11398: // }
11399:
1.169 brouard 11400: #endif
11401:
1.167 brouard 11402:
1.219 brouard 11403: }
1.136 brouard 11404:
1.219 brouard 11405: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11406: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 11407: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 11408: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11409: /* double ftolpl = 1.e-10; */
1.180 brouard 11410: double age, agebase, agelim;
1.203 brouard 11411: double tot;
1.180 brouard 11412:
1.202 brouard 11413: strcpy(filerespl,"PL_");
11414: strcat(filerespl,fileresu);
11415: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11416: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11417: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11418: }
1.288 brouard 11419: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11420: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11421: pstamp(ficrespl);
1.288 brouard 11422: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11423: fprintf(ficrespl,"#Age ");
11424: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11425: fprintf(ficrespl,"\n");
1.180 brouard 11426:
1.219 brouard 11427: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11428:
1.219 brouard 11429: agebase=ageminpar;
11430: agelim=agemaxpar;
1.180 brouard 11431:
1.227 brouard 11432: /* i1=pow(2,ncoveff); */
1.234 brouard 11433: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11434: if (cptcovn < 1){i1=1;}
1.180 brouard 11435:
1.238 brouard 11436: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
11437: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 11438: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11439: continue;
1.235 brouard 11440:
1.238 brouard 11441: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11442: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11443: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11444: /* k=k+1; */
11445: /* to clean */
1.332 brouard 11446: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11447: fprintf(ficrespl,"#******");
11448: printf("#******");
11449: fprintf(ficlog,"#******");
11450: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
1.332 brouard 11451: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
11452: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* Here problem for varying dummy*/
11453: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11454: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11455: }
11456: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11457: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11458: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11459: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11460: }
11461: fprintf(ficrespl,"******\n");
11462: printf("******\n");
11463: fprintf(ficlog,"******\n");
11464: if(invalidvarcomb[k]){
11465: printf("\nCombination (%d) ignored because no case \n",k);
11466: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
11467: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
11468: continue;
11469: }
1.219 brouard 11470:
1.238 brouard 11471: fprintf(ficrespl,"#Age ");
11472: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 11473: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11474: }
11475: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
11476: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 11477:
1.238 brouard 11478: for (age=agebase; age<=agelim; age++){
11479: /* for (age=agebase; age<=agebase; age++){ */
11480: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
11481: fprintf(ficrespl,"%.0f ",age );
11482: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11483: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11484: tot=0.;
11485: for(i=1; i<=nlstate;i++){
11486: tot += prlim[i][i];
11487: fprintf(ficrespl," %.5f", prlim[i][i]);
11488: }
11489: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
11490: } /* Age */
11491: /* was end of cptcod */
11492: } /* cptcov */
11493: } /* nres */
1.219 brouard 11494: return 0;
1.180 brouard 11495: }
11496:
1.218 brouard 11497: 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 11498: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 11499:
11500: /* Computes the back prevalence limit for any combination of covariate values
11501: * at any age between ageminpar and agemaxpar
11502: */
1.235 brouard 11503: int i, j, k, i1, nres=0 ;
1.217 brouard 11504: /* double ftolpl = 1.e-10; */
11505: double age, agebase, agelim;
11506: double tot;
1.218 brouard 11507: /* double ***mobaverage; */
11508: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 11509:
11510: strcpy(fileresplb,"PLB_");
11511: strcat(fileresplb,fileresu);
11512: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 11513: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
11514: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 11515: }
1.288 brouard 11516: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
11517: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 11518: pstamp(ficresplb);
1.288 brouard 11519: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 11520: fprintf(ficresplb,"#Age ");
11521: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
11522: fprintf(ficresplb,"\n");
11523:
1.218 brouard 11524:
11525: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
11526:
11527: agebase=ageminpar;
11528: agelim=agemaxpar;
11529:
11530:
1.227 brouard 11531: i1=pow(2,cptcoveff);
1.218 brouard 11532: if (cptcovn < 1){i1=1;}
1.227 brouard 11533:
1.238 brouard 11534: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11535: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11536: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11537: continue;
1.332 brouard 11538: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11539: fprintf(ficresplb,"#******");
11540: printf("#******");
11541: fprintf(ficlog,"#******");
11542: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
1.332 brouard 11543: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11544: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11545: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11546: }
11547: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 11548: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
11549: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
11550: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 11551: }
11552: fprintf(ficresplb,"******\n");
11553: printf("******\n");
11554: fprintf(ficlog,"******\n");
11555: if(invalidvarcomb[k]){
11556: printf("\nCombination (%d) ignored because no cases \n",k);
11557: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
11558: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
11559: continue;
11560: }
1.218 brouard 11561:
1.238 brouard 11562: fprintf(ficresplb,"#Age ");
11563: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 11564: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11565: }
11566: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
11567: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 11568:
11569:
1.238 brouard 11570: for (age=agebase; age<=agelim; age++){
11571: /* for (age=agebase; age<=agebase; age++){ */
11572: if(mobilavproj > 0){
11573: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11574: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11575: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11576: }else if (mobilavproj == 0){
11577: 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);
11578: 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);
11579: exit(1);
11580: }else{
11581: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11582: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11583: /* printf("TOTOT\n"); */
11584: /* exit(1); */
1.238 brouard 11585: }
11586: fprintf(ficresplb,"%.0f ",age );
11587: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11588: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11589: tot=0.;
11590: for(i=1; i<=nlstate;i++){
11591: tot += bprlim[i][i];
11592: fprintf(ficresplb," %.5f", bprlim[i][i]);
11593: }
11594: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11595: } /* Age */
11596: /* was end of cptcod */
1.255 brouard 11597: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 11598: } /* end of any combination */
11599: } /* end of nres */
1.218 brouard 11600: /* hBijx(p, bage, fage); */
11601: /* fclose(ficrespijb); */
11602:
11603: return 0;
1.217 brouard 11604: }
1.218 brouard 11605:
1.180 brouard 11606: int hPijx(double *p, int bage, int fage){
11607: /*------------- h Pij x at various ages ------------*/
11608:
11609: int stepsize;
11610: int agelim;
11611: int hstepm;
11612: int nhstepm;
1.235 brouard 11613: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11614:
11615: double agedeb;
11616: double ***p3mat;
11617:
1.201 brouard 11618: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 11619: if((ficrespij=fopen(filerespij,"w"))==NULL) {
11620: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
11621: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
11622: }
11623: printf("Computing pij: result on file '%s' \n", filerespij);
11624: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
11625:
11626: stepsize=(int) (stepm+YEARM-1)/YEARM;
11627: /*if (stepm<=24) stepsize=2;*/
11628:
11629: agelim=AGESUP;
11630: hstepm=stepsize*YEARM; /* Every year of age */
11631: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 11632:
1.180 brouard 11633: /* hstepm=1; aff par mois*/
11634: pstamp(ficrespij);
11635: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 11636: i1= pow(2,cptcoveff);
1.218 brouard 11637: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11638: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11639: /* k=k+1; */
1.235 brouard 11640: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11641: for(k=1; k<=i1;k++){
1.253 brouard 11642: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11643: continue;
1.183 brouard 11644: fprintf(ficrespij,"\n#****** ");
1.227 brouard 11645: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11646: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 11647: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11648: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11649: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11650: }
1.183 brouard 11651: fprintf(ficrespij,"******\n");
11652:
11653: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
11654: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
11655: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
11656:
11657: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 11658:
1.183 brouard 11659: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11660: oldm=oldms;savm=savms;
1.235 brouard 11661: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 11662: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
11663: for(i=1; i<=nlstate;i++)
11664: for(j=1; j<=nlstate+ndeath;j++)
11665: fprintf(ficrespij," %1d-%1d",i,j);
11666: fprintf(ficrespij,"\n");
11667: for (h=0; h<=nhstepm; h++){
11668: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11669: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 11670: for(i=1; i<=nlstate;i++)
11671: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 11672: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 11673: fprintf(ficrespij,"\n");
11674: }
1.183 brouard 11675: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11676: fprintf(ficrespij,"\n");
11677: }
1.180 brouard 11678: /*}*/
11679: }
1.218 brouard 11680: return 0;
1.180 brouard 11681: }
1.218 brouard 11682:
11683: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 11684: /*------------- h Bij x at various ages ------------*/
11685:
11686: int stepsize;
1.218 brouard 11687: /* int agelim; */
11688: int ageminl;
1.217 brouard 11689: int hstepm;
11690: int nhstepm;
1.238 brouard 11691: int h, i, i1, j, k, nres;
1.218 brouard 11692:
1.217 brouard 11693: double agedeb;
11694: double ***p3mat;
1.218 brouard 11695:
11696: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
11697: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
11698: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11699: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11700: }
11701: printf("Computing pij back: result on file '%s' \n", filerespijb);
11702: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
11703:
11704: stepsize=(int) (stepm+YEARM-1)/YEARM;
11705: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 11706:
1.218 brouard 11707: /* agelim=AGESUP; */
1.289 brouard 11708: ageminl=AGEINF; /* was 30 */
1.218 brouard 11709: hstepm=stepsize*YEARM; /* Every year of age */
11710: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11711:
11712: /* hstepm=1; aff par mois*/
11713: pstamp(ficrespijb);
1.255 brouard 11714: 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 11715: i1= pow(2,cptcoveff);
1.218 brouard 11716: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11717: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11718: /* k=k+1; */
1.238 brouard 11719: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11720: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11721: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11722: continue;
11723: fprintf(ficrespijb,"\n#****** ");
11724: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11725: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11726: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 11727: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 11728: }
11729: fprintf(ficrespijb,"******\n");
1.264 brouard 11730: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 11731: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
11732: continue;
11733: }
11734:
11735: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
11736: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
11737: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 11738: 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 */
11739: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 11740:
11741: /* nhstepm=nhstepm*YEARM; aff par mois*/
11742:
1.266 brouard 11743: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
11744: /* and memory limitations if stepm is small */
11745:
1.238 brouard 11746: /* oldm=oldms;savm=savms; */
11747: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.325 brouard 11748: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
1.238 brouard 11749: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 11750: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 11751: for(i=1; i<=nlstate;i++)
11752: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 11753: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 11754: fprintf(ficrespijb,"\n");
1.238 brouard 11755: for (h=0; h<=nhstepm; h++){
11756: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11757: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
11758: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
11759: for(i=1; i<=nlstate;i++)
11760: for(j=1; j<=nlstate+ndeath;j++)
1.325 brouard 11761: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.238 brouard 11762: fprintf(ficrespijb,"\n");
11763: }
11764: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11765: fprintf(ficrespijb,"\n");
11766: } /* end age deb */
11767: } /* end combination */
11768: } /* end nres */
1.218 brouard 11769: return 0;
11770: } /* hBijx */
1.217 brouard 11771:
1.180 brouard 11772:
1.136 brouard 11773: /***********************************************/
11774: /**************** Main Program *****************/
11775: /***********************************************/
11776:
11777: int main(int argc, char *argv[])
11778: {
11779: #ifdef GSL
11780: const gsl_multimin_fminimizer_type *T;
11781: size_t iteri = 0, it;
11782: int rval = GSL_CONTINUE;
11783: int status = GSL_SUCCESS;
11784: double ssval;
11785: #endif
11786: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11787: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11788: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11789: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11790: int jj, ll, li, lj, lk;
1.136 brouard 11791: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11792: int num_filled;
1.136 brouard 11793: int itimes;
11794: int NDIM=2;
11795: int vpopbased=0;
1.235 brouard 11796: int nres=0;
1.258 brouard 11797: int endishere=0;
1.277 brouard 11798: int noffset=0;
1.274 brouard 11799: int ncurrv=0; /* Temporary variable */
11800:
1.164 brouard 11801: char ca[32], cb[32];
1.136 brouard 11802: /* FILE *fichtm; *//* Html File */
11803: /* FILE *ficgp;*/ /*Gnuplot File */
11804: struct stat info;
1.191 brouard 11805: double agedeb=0.;
1.194 brouard 11806:
11807: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11808: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11809:
1.165 brouard 11810: double fret;
1.191 brouard 11811: double dum=0.; /* Dummy variable */
1.136 brouard 11812: double ***p3mat;
1.218 brouard 11813: /* double ***mobaverage; */
1.319 brouard 11814: double wald;
1.164 brouard 11815:
11816: char line[MAXLINE];
1.197 brouard 11817: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11818:
1.234 brouard 11819: char modeltemp[MAXLINE];
1.332 brouard 11820: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 11821:
1.136 brouard 11822: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11823: char *tok, *val; /* pathtot */
1.334 brouard 11824: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.195 brouard 11825: int c, h , cpt, c2;
1.191 brouard 11826: int jl=0;
11827: int i1, j1, jk, stepsize=0;
1.194 brouard 11828: int count=0;
11829:
1.164 brouard 11830: int *tab;
1.136 brouard 11831: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11832: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11833: /* double anprojf, mprojf, jprojf; */
11834: /* double jintmean,mintmean,aintmean; */
11835: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11836: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11837: double yrfproj= 10.0; /* Number of years of forward projections */
11838: double yrbproj= 10.0; /* Number of years of backward projections */
11839: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11840: int mobilav=0,popforecast=0;
1.191 brouard 11841: int hstepm=0, nhstepm=0;
1.136 brouard 11842: int agemortsup;
11843: float sumlpop=0.;
11844: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11845: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11846:
1.191 brouard 11847: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11848: double ftolpl=FTOL;
11849: double **prlim;
1.217 brouard 11850: double **bprlim;
1.317 brouard 11851: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
11852: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11853: double ***paramstart; /* Matrix of starting parameter values */
11854: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11855: double **matcov; /* Matrix of covariance */
1.203 brouard 11856: double **hess; /* Hessian matrix */
1.136 brouard 11857: double ***delti3; /* Scale */
11858: double *delti; /* Scale */
11859: double ***eij, ***vareij;
11860: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11861:
1.136 brouard 11862: double *epj, vepp;
1.164 brouard 11863:
1.273 brouard 11864: double dateprev1, dateprev2;
1.296 brouard 11865: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11866: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11867:
1.217 brouard 11868:
1.136 brouard 11869: double **ximort;
1.145 brouard 11870: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11871: int *dcwave;
11872:
1.164 brouard 11873: char z[1]="c";
1.136 brouard 11874:
11875: /*char *strt;*/
11876: char strtend[80];
1.126 brouard 11877:
1.164 brouard 11878:
1.126 brouard 11879: /* setlocale (LC_ALL, ""); */
11880: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11881: /* textdomain (PACKAGE); */
11882: /* setlocale (LC_CTYPE, ""); */
11883: /* setlocale (LC_MESSAGES, ""); */
11884:
11885: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11886: rstart_time = time(NULL);
11887: /* (void) gettimeofday(&start_time,&tzp);*/
11888: start_time = *localtime(&rstart_time);
1.126 brouard 11889: curr_time=start_time;
1.157 brouard 11890: /*tml = *localtime(&start_time.tm_sec);*/
11891: /* strcpy(strstart,asctime(&tml)); */
11892: strcpy(strstart,asctime(&start_time));
1.126 brouard 11893:
11894: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11895: /* tp.tm_sec = tp.tm_sec +86400; */
11896: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11897: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11898: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11899: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11900: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11901: /* strt=asctime(&tmg); */
11902: /* printf("Time(after) =%s",strstart); */
11903: /* (void) time (&time_value);
11904: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11905: * tm = *localtime(&time_value);
11906: * strstart=asctime(&tm);
11907: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11908: */
11909:
11910: nberr=0; /* Number of errors and warnings */
11911: nbwarn=0;
1.184 brouard 11912: #ifdef WIN32
11913: _getcwd(pathcd, size);
11914: #else
1.126 brouard 11915: getcwd(pathcd, size);
1.184 brouard 11916: #endif
1.191 brouard 11917: syscompilerinfo(0);
1.196 brouard 11918: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11919: if(argc <=1){
11920: printf("\nEnter the parameter file name: ");
1.205 brouard 11921: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11922: printf("ERROR Empty parameter file name\n");
11923: goto end;
11924: }
1.126 brouard 11925: i=strlen(pathr);
11926: if(pathr[i-1]=='\n')
11927: pathr[i-1]='\0';
1.156 brouard 11928: i=strlen(pathr);
1.205 brouard 11929: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11930: pathr[i-1]='\0';
1.205 brouard 11931: }
11932: i=strlen(pathr);
11933: if( i==0 ){
11934: printf("ERROR Empty parameter file name\n");
11935: goto end;
11936: }
11937: for (tok = pathr; tok != NULL; ){
1.126 brouard 11938: printf("Pathr |%s|\n",pathr);
11939: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11940: printf("val= |%s| pathr=%s\n",val,pathr);
11941: strcpy (pathtot, val);
11942: if(pathr[0] == '\0') break; /* Dirty */
11943: }
11944: }
1.281 brouard 11945: else if (argc<=2){
11946: strcpy(pathtot,argv[1]);
11947: }
1.126 brouard 11948: else{
11949: strcpy(pathtot,argv[1]);
1.281 brouard 11950: strcpy(z,argv[2]);
11951: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11952: }
11953: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11954: /*cygwin_split_path(pathtot,path,optionfile);
11955: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11956: /* cutv(path,optionfile,pathtot,'\\');*/
11957:
11958: /* Split argv[0], imach program to get pathimach */
11959: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11960: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11961: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11962: /* strcpy(pathimach,argv[0]); */
11963: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11964: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11965: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11966: #ifdef WIN32
11967: _chdir(path); /* Can be a relative path */
11968: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11969: #else
1.126 brouard 11970: chdir(path); /* Can be a relative path */
1.184 brouard 11971: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11972: #endif
11973: printf("Current directory %s!\n",pathcd);
1.126 brouard 11974: strcpy(command,"mkdir ");
11975: strcat(command,optionfilefiname);
11976: if((outcmd=system(command)) != 0){
1.169 brouard 11977: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11978: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11979: /* fclose(ficlog); */
11980: /* exit(1); */
11981: }
11982: /* if((imk=mkdir(optionfilefiname))<0){ */
11983: /* perror("mkdir"); */
11984: /* } */
11985:
11986: /*-------- arguments in the command line --------*/
11987:
1.186 brouard 11988: /* Main Log file */
1.126 brouard 11989: strcat(filelog, optionfilefiname);
11990: strcat(filelog,".log"); /* */
11991: if((ficlog=fopen(filelog,"w"))==NULL) {
11992: printf("Problem with logfile %s\n",filelog);
11993: goto end;
11994: }
11995: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11996: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11997: fprintf(ficlog,"\nEnter the parameter file name: \n");
11998: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11999: path=%s \n\
12000: optionfile=%s\n\
12001: optionfilext=%s\n\
1.156 brouard 12002: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 12003:
1.197 brouard 12004: syscompilerinfo(1);
1.167 brouard 12005:
1.126 brouard 12006: printf("Local time (at start):%s",strstart);
12007: fprintf(ficlog,"Local time (at start): %s",strstart);
12008: fflush(ficlog);
12009: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 12010: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 12011:
12012: /* */
12013: strcpy(fileres,"r");
12014: strcat(fileres, optionfilefiname);
1.201 brouard 12015: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 12016: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 12017: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 12018:
1.186 brouard 12019: /* Main ---------arguments file --------*/
1.126 brouard 12020:
12021: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 12022: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
12023: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 12024: fflush(ficlog);
1.149 brouard 12025: /* goto end; */
12026: exit(70);
1.126 brouard 12027: }
12028:
12029: strcpy(filereso,"o");
1.201 brouard 12030: strcat(filereso,fileresu);
1.126 brouard 12031: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
12032: printf("Problem with Output resultfile: %s\n", filereso);
12033: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
12034: fflush(ficlog);
12035: goto end;
12036: }
1.278 brouard 12037: /*-------- Rewriting parameter file ----------*/
12038: strcpy(rfileres,"r"); /* "Rparameterfile */
12039: strcat(rfileres,optionfilefiname); /* Parameter file first name */
12040: strcat(rfileres,"."); /* */
12041: strcat(rfileres,optionfilext); /* Other files have txt extension */
12042: if((ficres =fopen(rfileres,"w"))==NULL) {
12043: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
12044: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
12045: fflush(ficlog);
12046: goto end;
12047: }
12048: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 12049:
1.278 brouard 12050:
1.126 brouard 12051: /* Reads comments: lines beginning with '#' */
12052: numlinepar=0;
1.277 brouard 12053: /* Is it a BOM UTF-8 Windows file? */
12054: /* First parameter line */
1.197 brouard 12055: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 12056: noffset=0;
12057: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12058: {
12059: noffset=noffset+3;
12060: printf("# File is an UTF8 Bom.\n"); // 0xBF
12061: }
1.302 brouard 12062: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12063: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 12064: {
12065: noffset=noffset+2;
12066: printf("# File is an UTF16BE BOM file\n");
12067: }
12068: else if( line[0] == 0 && line[1] == 0)
12069: {
12070: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12071: noffset=noffset+4;
12072: printf("# File is an UTF16BE BOM file\n");
12073: }
12074: } else{
12075: ;/*printf(" Not a BOM file\n");*/
12076: }
12077:
1.197 brouard 12078: /* If line starts with a # it is a comment */
1.277 brouard 12079: if (line[noffset] == '#') {
1.197 brouard 12080: numlinepar++;
12081: fputs(line,stdout);
12082: fputs(line,ficparo);
1.278 brouard 12083: fputs(line,ficres);
1.197 brouard 12084: fputs(line,ficlog);
12085: continue;
12086: }else
12087: break;
12088: }
12089: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
12090: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
12091: if (num_filled != 5) {
12092: printf("Should be 5 parameters\n");
1.283 brouard 12093: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 12094: }
1.126 brouard 12095: numlinepar++;
1.197 brouard 12096: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 12097: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12098: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12099: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 12100: }
12101: /* Second parameter line */
12102: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 12103: /* while(fscanf(ficpar,"%[^\n]", line)) { */
12104: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 12105: if (line[0] == '#') {
12106: numlinepar++;
1.283 brouard 12107: printf("%s",line);
12108: fprintf(ficres,"%s",line);
12109: fprintf(ficparo,"%s",line);
12110: fprintf(ficlog,"%s",line);
1.197 brouard 12111: continue;
12112: }else
12113: break;
12114: }
1.223 brouard 12115: 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", \
12116: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
12117: if (num_filled != 11) {
12118: 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 12119: printf("but line=%s\n",line);
1.283 brouard 12120: 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");
12121: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 12122: }
1.286 brouard 12123: if( lastpass > maxwav){
12124: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12125: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12126: fflush(ficlog);
12127: goto end;
12128: }
12129: 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 12130: 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 12131: 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 12132: 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 12133: }
1.203 brouard 12134: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 12135: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 12136: /* Third parameter line */
12137: while(fgets(line, MAXLINE, ficpar)) {
12138: /* If line starts with a # it is a comment */
12139: if (line[0] == '#') {
12140: numlinepar++;
1.283 brouard 12141: printf("%s",line);
12142: fprintf(ficres,"%s",line);
12143: fprintf(ficparo,"%s",line);
12144: fprintf(ficlog,"%s",line);
1.197 brouard 12145: continue;
12146: }else
12147: break;
12148: }
1.201 brouard 12149: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 12150: if (num_filled != 1){
1.302 brouard 12151: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
12152: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 12153: model[0]='\0';
12154: goto end;
12155: }
12156: else{
12157: if (model[0]=='+'){
12158: for(i=1; i<=strlen(model);i++)
12159: modeltemp[i-1]=model[i];
1.201 brouard 12160: strcpy(model,modeltemp);
1.197 brouard 12161: }
12162: }
1.199 brouard 12163: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 12164: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 12165: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
12166: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
12167: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 12168: }
12169: /* 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); */
12170: /* numlinepar=numlinepar+3; /\* In general *\/ */
12171: /* 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 12172: /* 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); */
12173: /* 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 12174: fflush(ficlog);
1.190 brouard 12175: /* if(model[0]=='#'|| model[0]== '\0'){ */
12176: if(model[0]=='#'){
1.279 brouard 12177: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
12178: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
12179: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 12180: if(mle != -1){
1.279 brouard 12181: 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 12182: exit(1);
12183: }
12184: }
1.126 brouard 12185: while((c=getc(ficpar))=='#' && c!= EOF){
12186: ungetc(c,ficpar);
12187: fgets(line, MAXLINE, ficpar);
12188: numlinepar++;
1.195 brouard 12189: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
12190: z[0]=line[1];
12191: }
12192: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 12193: fputs(line, stdout);
12194: //puts(line);
1.126 brouard 12195: fputs(line,ficparo);
12196: fputs(line,ficlog);
12197: }
12198: ungetc(c,ficpar);
12199:
12200:
1.290 brouard 12201: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
12202: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
12203: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
12204: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 12205: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
12206: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
12207: v1+v2*age+v2*v3 makes cptcovn = 3
12208: */
12209: if (strlen(model)>1)
1.187 brouard 12210: 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 12211: else
1.187 brouard 12212: ncovmodel=2; /* Constant and age */
1.133 brouard 12213: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
12214: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 12215: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
12216: 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);
12217: 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);
12218: fflush(stdout);
12219: fclose (ficlog);
12220: goto end;
12221: }
1.126 brouard 12222: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12223: delti=delti3[1][1];
12224: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
12225: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 12226: /* We could also provide initial parameters values giving by simple logistic regression
12227: * only one way, that is without matrix product. We will have nlstate maximizations */
12228: /* for(i=1;i<nlstate;i++){ */
12229: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12230: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12231: /* } */
1.126 brouard 12232: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 12233: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
12234: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12235: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12236: fclose (ficparo);
12237: fclose (ficlog);
12238: goto end;
12239: exit(0);
1.220 brouard 12240: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 12241: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 12242: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
12243: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12244: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12245: matcov=matrix(1,npar,1,npar);
1.203 brouard 12246: hess=matrix(1,npar,1,npar);
1.220 brouard 12247: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 12248: /* Read guessed parameters */
1.126 brouard 12249: /* Reads comments: lines beginning with '#' */
12250: while((c=getc(ficpar))=='#' && c!= EOF){
12251: ungetc(c,ficpar);
12252: fgets(line, MAXLINE, ficpar);
12253: numlinepar++;
1.141 brouard 12254: fputs(line,stdout);
1.126 brouard 12255: fputs(line,ficparo);
12256: fputs(line,ficlog);
12257: }
12258: ungetc(c,ficpar);
12259:
12260: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 12261: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 12262: for(i=1; i <=nlstate; i++){
1.234 brouard 12263: j=0;
1.126 brouard 12264: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 12265: if(jj==i) continue;
12266: j++;
1.292 brouard 12267: while((c=getc(ficpar))=='#' && c!= EOF){
12268: ungetc(c,ficpar);
12269: fgets(line, MAXLINE, ficpar);
12270: numlinepar++;
12271: fputs(line,stdout);
12272: fputs(line,ficparo);
12273: fputs(line,ficlog);
12274: }
12275: ungetc(c,ficpar);
1.234 brouard 12276: fscanf(ficpar,"%1d%1d",&i1,&j1);
12277: if ((i1 != i) || (j1 != jj)){
12278: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 12279: It might be a problem of design; if ncovcol and the model are correct\n \
12280: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 12281: exit(1);
12282: }
12283: fprintf(ficparo,"%1d%1d",i1,j1);
12284: if(mle==1)
12285: printf("%1d%1d",i,jj);
12286: fprintf(ficlog,"%1d%1d",i,jj);
12287: for(k=1; k<=ncovmodel;k++){
12288: fscanf(ficpar," %lf",¶m[i][j][k]);
12289: if(mle==1){
12290: printf(" %lf",param[i][j][k]);
12291: fprintf(ficlog," %lf",param[i][j][k]);
12292: }
12293: else
12294: fprintf(ficlog," %lf",param[i][j][k]);
12295: fprintf(ficparo," %lf",param[i][j][k]);
12296: }
12297: fscanf(ficpar,"\n");
12298: numlinepar++;
12299: if(mle==1)
12300: printf("\n");
12301: fprintf(ficlog,"\n");
12302: fprintf(ficparo,"\n");
1.126 brouard 12303: }
12304: }
12305: fflush(ficlog);
1.234 brouard 12306:
1.251 brouard 12307: /* Reads parameters values */
1.126 brouard 12308: p=param[1][1];
1.251 brouard 12309: pstart=paramstart[1][1];
1.126 brouard 12310:
12311: /* Reads comments: lines beginning with '#' */
12312: while((c=getc(ficpar))=='#' && c!= EOF){
12313: ungetc(c,ficpar);
12314: fgets(line, MAXLINE, ficpar);
12315: numlinepar++;
1.141 brouard 12316: fputs(line,stdout);
1.126 brouard 12317: fputs(line,ficparo);
12318: fputs(line,ficlog);
12319: }
12320: ungetc(c,ficpar);
12321:
12322: for(i=1; i <=nlstate; i++){
12323: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 12324: fscanf(ficpar,"%1d%1d",&i1,&j1);
12325: if ( (i1-i) * (j1-j) != 0){
12326: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
12327: exit(1);
12328: }
12329: printf("%1d%1d",i,j);
12330: fprintf(ficparo,"%1d%1d",i1,j1);
12331: fprintf(ficlog,"%1d%1d",i1,j1);
12332: for(k=1; k<=ncovmodel;k++){
12333: fscanf(ficpar,"%le",&delti3[i][j][k]);
12334: printf(" %le",delti3[i][j][k]);
12335: fprintf(ficparo," %le",delti3[i][j][k]);
12336: fprintf(ficlog," %le",delti3[i][j][k]);
12337: }
12338: fscanf(ficpar,"\n");
12339: numlinepar++;
12340: printf("\n");
12341: fprintf(ficparo,"\n");
12342: fprintf(ficlog,"\n");
1.126 brouard 12343: }
12344: }
12345: fflush(ficlog);
1.234 brouard 12346:
1.145 brouard 12347: /* Reads covariance matrix */
1.126 brouard 12348: delti=delti3[1][1];
1.220 brouard 12349:
12350:
1.126 brouard 12351: /* 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 12352:
1.126 brouard 12353: /* Reads comments: lines beginning with '#' */
12354: while((c=getc(ficpar))=='#' && c!= EOF){
12355: ungetc(c,ficpar);
12356: fgets(line, MAXLINE, ficpar);
12357: numlinepar++;
1.141 brouard 12358: fputs(line,stdout);
1.126 brouard 12359: fputs(line,ficparo);
12360: fputs(line,ficlog);
12361: }
12362: ungetc(c,ficpar);
1.220 brouard 12363:
1.126 brouard 12364: matcov=matrix(1,npar,1,npar);
1.203 brouard 12365: hess=matrix(1,npar,1,npar);
1.131 brouard 12366: for(i=1; i <=npar; i++)
12367: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12368:
1.194 brouard 12369: /* Scans npar lines */
1.126 brouard 12370: for(i=1; i <=npar; i++){
1.226 brouard 12371: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12372: if(count != 3){
1.226 brouard 12373: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12374: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12375: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12376: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12377: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12378: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12379: exit(1);
1.220 brouard 12380: }else{
1.226 brouard 12381: if(mle==1)
12382: printf("%1d%1d%d",i1,j1,jk);
12383: }
12384: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12385: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12386: for(j=1; j <=i; j++){
1.226 brouard 12387: fscanf(ficpar," %le",&matcov[i][j]);
12388: if(mle==1){
12389: printf(" %.5le",matcov[i][j]);
12390: }
12391: fprintf(ficlog," %.5le",matcov[i][j]);
12392: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12393: }
12394: fscanf(ficpar,"\n");
12395: numlinepar++;
12396: if(mle==1)
1.220 brouard 12397: printf("\n");
1.126 brouard 12398: fprintf(ficlog,"\n");
12399: fprintf(ficparo,"\n");
12400: }
1.194 brouard 12401: /* End of read covariance matrix npar lines */
1.126 brouard 12402: for(i=1; i <=npar; i++)
12403: for(j=i+1;j<=npar;j++)
1.226 brouard 12404: matcov[i][j]=matcov[j][i];
1.126 brouard 12405:
12406: if(mle==1)
12407: printf("\n");
12408: fprintf(ficlog,"\n");
12409:
12410: fflush(ficlog);
12411:
12412: } /* End of mle != -3 */
1.218 brouard 12413:
1.186 brouard 12414: /* Main data
12415: */
1.290 brouard 12416: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12417: /* num=lvector(1,n); */
12418: /* moisnais=vector(1,n); */
12419: /* annais=vector(1,n); */
12420: /* moisdc=vector(1,n); */
12421: /* andc=vector(1,n); */
12422: /* weight=vector(1,n); */
12423: /* agedc=vector(1,n); */
12424: /* cod=ivector(1,n); */
12425: /* for(i=1;i<=n;i++){ */
12426: num=lvector(firstobs,lastobs);
12427: moisnais=vector(firstobs,lastobs);
12428: annais=vector(firstobs,lastobs);
12429: moisdc=vector(firstobs,lastobs);
12430: andc=vector(firstobs,lastobs);
12431: weight=vector(firstobs,lastobs);
12432: agedc=vector(firstobs,lastobs);
12433: cod=ivector(firstobs,lastobs);
12434: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 12435: num[i]=0;
12436: moisnais[i]=0;
12437: annais[i]=0;
12438: moisdc[i]=0;
12439: andc[i]=0;
12440: agedc[i]=0;
12441: cod[i]=0;
12442: weight[i]=1.0; /* Equal weights, 1 by default */
12443: }
1.290 brouard 12444: mint=matrix(1,maxwav,firstobs,lastobs);
12445: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 12446: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
12447: printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel));
1.126 brouard 12448: tab=ivector(1,NCOVMAX);
1.144 brouard 12449: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 12450: 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 12451:
1.136 brouard 12452: /* Reads data from file datafile */
12453: if (readdata(datafile, firstobs, lastobs, &imx)==1)
12454: goto end;
12455:
12456: /* Calculation of the number of parameters from char model */
1.234 brouard 12457: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 12458: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
12459: k=3 V4 Tvar[k=3]= 4 (from V4)
12460: k=2 V1 Tvar[k=2]= 1 (from V1)
12461: k=1 Tvar[1]=2 (from V2)
1.234 brouard 12462: */
12463:
12464: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
12465: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 12466: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 ! brouard 12467: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 12468: TvarsD=ivector(1,NCOVMAX); /* */
12469: TvarsQind=ivector(1,NCOVMAX); /* */
12470: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 12471: TvarF=ivector(1,NCOVMAX); /* */
12472: TvarFind=ivector(1,NCOVMAX); /* */
12473: TvarV=ivector(1,NCOVMAX); /* */
12474: TvarVind=ivector(1,NCOVMAX); /* */
12475: TvarA=ivector(1,NCOVMAX); /* */
12476: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 12477: TvarFD=ivector(1,NCOVMAX); /* */
12478: TvarFDind=ivector(1,NCOVMAX); /* */
12479: TvarFQ=ivector(1,NCOVMAX); /* */
12480: TvarFQind=ivector(1,NCOVMAX); /* */
12481: TvarVD=ivector(1,NCOVMAX); /* */
12482: TvarVDind=ivector(1,NCOVMAX); /* */
12483: TvarVQ=ivector(1,NCOVMAX); /* */
12484: TvarVQind=ivector(1,NCOVMAX); /* */
12485:
1.230 brouard 12486: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 12487: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 12488: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
12489: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
12490: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 12491: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
12492: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
12493: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
12494: */
12495: /* For model-covariate k tells which data-covariate to use but
12496: because this model-covariate is a construction we invent a new column
12497: ncovcol + k1
12498: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
12499: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 12500: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
12501: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 12502: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
12503: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 12504: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 12505: */
1.145 brouard 12506: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
12507: 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 12508: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
12509: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 brouard 12510: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 12511: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 12512: 4 covariates (3 plus signs)
12513: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 12514: */
12515: for(i=1;i<NCOVMAX;i++)
12516: Tage[i]=0;
1.230 brouard 12517: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 12518: * individual dummy, fixed or varying:
12519: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
12520: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 12521: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
12522: * V1 df, V2 qf, V3 & V4 dv, V5 qv
12523: * Tmodelind[1]@9={9,0,3,2,}*/
12524: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
12525: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 12526: * individual quantitative, fixed or varying:
12527: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
12528: * 3, 1, 0, 0, 0, 0, 0, 0},
12529: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 12530: /* Main decodemodel */
12531:
1.187 brouard 12532:
1.223 brouard 12533: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 12534: goto end;
12535:
1.137 brouard 12536: if((double)(lastobs-imx)/(double)imx > 1.10){
12537: nbwarn++;
12538: 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);
12539: 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);
12540: }
1.136 brouard 12541: /* if(mle==1){*/
1.137 brouard 12542: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
12543: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 12544: }
12545:
12546: /*-calculation of age at interview from date of interview and age at death -*/
12547: agev=matrix(1,maxwav,1,imx);
12548:
12549: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
12550: goto end;
12551:
1.126 brouard 12552:
1.136 brouard 12553: agegomp=(int)agemin;
1.290 brouard 12554: free_vector(moisnais,firstobs,lastobs);
12555: free_vector(annais,firstobs,lastobs);
1.126 brouard 12556: /* free_matrix(mint,1,maxwav,1,n);
12557: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 12558: /* free_vector(moisdc,1,n); */
12559: /* free_vector(andc,1,n); */
1.145 brouard 12560: /* */
12561:
1.126 brouard 12562: wav=ivector(1,imx);
1.214 brouard 12563: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
12564: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
12565: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
12566: 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.*/
12567: bh=imatrix(1,lastpass-firstpass+2,1,imx);
12568: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 12569:
12570: /* Concatenates waves */
1.214 brouard 12571: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
12572: Death is a valid wave (if date is known).
12573: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
12574: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
12575: and mw[mi+1][i]. dh depends on stepm.
12576: */
12577:
1.126 brouard 12578: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 12579: /* Concatenates waves */
1.145 brouard 12580:
1.290 brouard 12581: free_vector(moisdc,firstobs,lastobs);
12582: free_vector(andc,firstobs,lastobs);
1.215 brouard 12583:
1.126 brouard 12584: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12585: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12586: ncodemax[1]=1;
1.145 brouard 12587: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12588: cptcoveff=0;
1.220 brouard 12589: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 ! brouard 12590: 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 12591: }
12592:
12593: ncovcombmax=pow(2,cptcoveff);
12594: invalidvarcomb=ivector(1, ncovcombmax);
12595: for(i=1;i<ncovcombmax;i++)
12596: invalidvarcomb[i]=0;
12597:
1.211 brouard 12598: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12599: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12600: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12601:
1.200 brouard 12602: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12603: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12604: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12605: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12606: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12607: * (currently 0 or 1) in the data.
12608: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12609: * corresponding modality (h,j).
12610: */
12611:
1.145 brouard 12612: h=0;
12613: /*if (cptcovn > 0) */
1.126 brouard 12614: m=pow(2,cptcoveff);
12615:
1.144 brouard 12616: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12617: * For k=4 covariates, h goes from 1 to m=2**k
12618: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12619: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 12620: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
12621: *______________________________ *______________________
12622: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
12623: * 2 2 1 1 1 * 1 0 0 0 1
12624: * 3 i=2 1 2 1 1 * 2 0 0 1 0
12625: * 4 2 2 1 1 * 3 0 0 1 1
12626: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
12627: * 6 2 1 2 1 * 5 0 1 0 1
12628: * 7 i=4 1 2 2 1 * 6 0 1 1 0
12629: * 8 2 2 2 1 * 7 0 1 1 1
12630: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
12631: * 10 2 1 1 2 * 9 1 0 0 1
12632: * 11 i=6 1 2 1 2 * 10 1 0 1 0
12633: * 12 2 2 1 2 * 11 1 0 1 1
12634: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
12635: * 14 2 1 2 2 * 13 1 1 0 1
12636: * 15 i=8 1 2 2 2 * 14 1 1 1 0
12637: * 16 2 2 2 2 * 15 1 1 1 1
12638: */
1.212 brouard 12639: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 12640: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
12641: * and the value of each covariate?
12642: * V1=1, V2=1, V3=2, V4=1 ?
12643: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
12644: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
12645: * In order to get the real value in the data, we use nbcode
12646: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
12647: * We are keeping this crazy system in order to be able (in the future?)
12648: * to have more than 2 values (0 or 1) for a covariate.
12649: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
12650: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
12651: * bbbbbbbb
12652: * 76543210
12653: * h-1 00000101 (6-1=5)
1.219 brouard 12654: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 12655: * &
12656: * 1 00000001 (1)
1.219 brouard 12657: * 00000000 = 1 & ((h-1) >> (k-1))
12658: * +1= 00000001 =1
1.211 brouard 12659: *
12660: * h=14, k=3 => h'=h-1=13, k'=k-1=2
12661: * h' 1101 =2^3+2^2+0x2^1+2^0
12662: * >>k' 11
12663: * & 00000001
12664: * = 00000001
12665: * +1 = 00000010=2 = codtabm(14,3)
12666: * Reverse h=6 and m=16?
12667: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
12668: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
12669: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
12670: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
12671: * V3=decodtabm(14,3,2**4)=2
12672: * h'=13 1101 =2^3+2^2+0x2^1+2^0
12673: *(h-1) >> (j-1) 0011 =13 >> 2
12674: * &1 000000001
12675: * = 000000001
12676: * +1= 000000010 =2
12677: * 2211
12678: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
12679: * V3=2
1.220 brouard 12680: * codtabm and decodtabm are identical
1.211 brouard 12681: */
12682:
1.145 brouard 12683:
12684: free_ivector(Ndum,-1,NCOVMAX);
12685:
12686:
1.126 brouard 12687:
1.186 brouard 12688: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 12689: strcpy(optionfilegnuplot,optionfilefiname);
12690: if(mle==-3)
1.201 brouard 12691: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 12692: strcat(optionfilegnuplot,".gp");
12693:
12694: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
12695: printf("Problem with file %s",optionfilegnuplot);
12696: }
12697: else{
1.204 brouard 12698: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 12699: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 12700: //fprintf(ficgp,"set missing 'NaNq'\n");
12701: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 12702: }
12703: /* fclose(ficgp);*/
1.186 brouard 12704:
12705:
12706: /* Initialisation of --------- index.htm --------*/
1.126 brouard 12707:
12708: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
12709: if(mle==-3)
1.201 brouard 12710: strcat(optionfilehtm,"-MORT_");
1.126 brouard 12711: strcat(optionfilehtm,".htm");
12712: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 12713: printf("Problem with %s \n",optionfilehtm);
12714: exit(0);
1.126 brouard 12715: }
12716:
12717: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
12718: strcat(optionfilehtmcov,"-cov.htm");
12719: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
12720: printf("Problem with %s \n",optionfilehtmcov), exit(0);
12721: }
12722: else{
12723: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
12724: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12725: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 12726: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
12727: }
12728:
1.335 ! brouard 12729: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
! 12730: <title>IMaCh %s</title></head>\n\
! 12731: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
! 12732: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
! 12733: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
! 12734: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
! 12735: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
! 12736:
! 12737: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12738: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 12739: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.335 ! brouard 12740: This file: <a href=\"%s\">%s</a>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 12741: \n\
12742: <hr size=\"2\" color=\"#EC5E5E\">\
12743: <ul><li><h4>Parameter files</h4>\n\
12744: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
12745: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
12746: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
12747: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
12748: - Date and time at start: %s</ul>\n",\
1.335 ! brouard 12749: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 12750: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
12751: fileres,fileres,\
12752: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
12753: fflush(fichtm);
12754:
12755: strcpy(pathr,path);
12756: strcat(pathr,optionfilefiname);
1.184 brouard 12757: #ifdef WIN32
12758: _chdir(optionfilefiname); /* Move to directory named optionfile */
12759: #else
1.126 brouard 12760: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12761: #endif
12762:
1.126 brouard 12763:
1.220 brouard 12764: /* Calculates basic frequencies. Computes observed prevalence at single age
12765: and for any valid combination of covariates
1.126 brouard 12766: and prints on file fileres'p'. */
1.251 brouard 12767: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12768: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12769:
12770: fprintf(fichtm,"\n");
1.286 brouard 12771: 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 12772: ftol, stepm);
12773: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12774: ncurrv=1;
12775: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12776: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12777: ncurrv=i;
12778: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12779: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12780: ncurrv=i;
12781: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12782: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12783: ncurrv=i;
12784: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12785: 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", \
12786: nlstate, ndeath, maxwav, mle, weightopt);
12787:
12788: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12789: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12790:
12791:
1.317 brouard 12792: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12793: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12794: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12795: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12796: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12797: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12798: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12799: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12800: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12801:
1.126 brouard 12802: /* For Powell, parameters are in a vector p[] starting at p[1]
12803: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12804: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12805:
12806: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12807: /* For mortality only */
1.126 brouard 12808: if (mle==-3){
1.136 brouard 12809: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12810: for(i=1;i<=NDIM;i++)
12811: for(j=1;j<=NDIM;j++)
12812: ximort[i][j]=0.;
1.186 brouard 12813: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12814: cens=ivector(firstobs,lastobs);
12815: ageexmed=vector(firstobs,lastobs);
12816: agecens=vector(firstobs,lastobs);
12817: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12818:
1.126 brouard 12819: for (i=1; i<=imx; i++){
12820: dcwave[i]=-1;
12821: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12822: if (s[m][i]>nlstate) {
12823: dcwave[i]=m;
12824: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12825: break;
12826: }
1.126 brouard 12827: }
1.226 brouard 12828:
1.126 brouard 12829: for (i=1; i<=imx; i++) {
12830: if (wav[i]>0){
1.226 brouard 12831: ageexmed[i]=agev[mw[1][i]][i];
12832: j=wav[i];
12833: agecens[i]=1.;
12834:
12835: if (ageexmed[i]> 1 && wav[i] > 0){
12836: agecens[i]=agev[mw[j][i]][i];
12837: cens[i]= 1;
12838: }else if (ageexmed[i]< 1)
12839: cens[i]= -1;
12840: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12841: cens[i]=0 ;
1.126 brouard 12842: }
12843: else cens[i]=-1;
12844: }
12845:
12846: for (i=1;i<=NDIM;i++) {
12847: for (j=1;j<=NDIM;j++)
1.226 brouard 12848: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12849: }
12850:
1.302 brouard 12851: p[1]=0.0268; p[NDIM]=0.083;
12852: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12853:
12854:
1.136 brouard 12855: #ifdef GSL
12856: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12857: #else
1.126 brouard 12858: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12859: #endif
1.201 brouard 12860: strcpy(filerespow,"POW-MORT_");
12861: strcat(filerespow,fileresu);
1.126 brouard 12862: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12863: printf("Problem with resultfile: %s\n", filerespow);
12864: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12865: }
1.136 brouard 12866: #ifdef GSL
12867: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12868: #else
1.126 brouard 12869: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12870: #endif
1.126 brouard 12871: /* for (i=1;i<=nlstate;i++)
12872: for(j=1;j<=nlstate+ndeath;j++)
12873: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12874: */
12875: fprintf(ficrespow,"\n");
1.136 brouard 12876: #ifdef GSL
12877: /* gsl starts here */
12878: T = gsl_multimin_fminimizer_nmsimplex;
12879: gsl_multimin_fminimizer *sfm = NULL;
12880: gsl_vector *ss, *x;
12881: gsl_multimin_function minex_func;
12882:
12883: /* Initial vertex size vector */
12884: ss = gsl_vector_alloc (NDIM);
12885:
12886: if (ss == NULL){
12887: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12888: }
12889: /* Set all step sizes to 1 */
12890: gsl_vector_set_all (ss, 0.001);
12891:
12892: /* Starting point */
1.126 brouard 12893:
1.136 brouard 12894: x = gsl_vector_alloc (NDIM);
12895:
12896: if (x == NULL){
12897: gsl_vector_free(ss);
12898: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12899: }
12900:
12901: /* Initialize method and iterate */
12902: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12903: /* gsl_vector_set(x, 0, 0.0268); */
12904: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12905: gsl_vector_set(x, 0, p[1]);
12906: gsl_vector_set(x, 1, p[2]);
12907:
12908: minex_func.f = &gompertz_f;
12909: minex_func.n = NDIM;
12910: minex_func.params = (void *)&p; /* ??? */
12911:
12912: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12913: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12914:
12915: printf("Iterations beginning .....\n\n");
12916: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12917:
12918: iteri=0;
12919: while (rval == GSL_CONTINUE){
12920: iteri++;
12921: status = gsl_multimin_fminimizer_iterate(sfm);
12922:
12923: if (status) printf("error: %s\n", gsl_strerror (status));
12924: fflush(0);
12925:
12926: if (status)
12927: break;
12928:
12929: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12930: ssval = gsl_multimin_fminimizer_size (sfm);
12931:
12932: if (rval == GSL_SUCCESS)
12933: printf ("converged to a local maximum at\n");
12934:
12935: printf("%5d ", iteri);
12936: for (it = 0; it < NDIM; it++){
12937: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12938: }
12939: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12940: }
12941:
12942: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12943:
12944: gsl_vector_free(x); /* initial values */
12945: gsl_vector_free(ss); /* inital step size */
12946: for (it=0; it<NDIM; it++){
12947: p[it+1]=gsl_vector_get(sfm->x,it);
12948: fprintf(ficrespow," %.12lf", p[it]);
12949: }
12950: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12951: #endif
12952: #ifdef POWELL
12953: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12954: #endif
1.126 brouard 12955: fclose(ficrespow);
12956:
1.203 brouard 12957: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12958:
12959: for(i=1; i <=NDIM; i++)
12960: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12961: matcov[i][j]=matcov[j][i];
1.126 brouard 12962:
12963: printf("\nCovariance matrix\n ");
1.203 brouard 12964: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12965: for(i=1; i <=NDIM; i++) {
12966: for(j=1;j<=NDIM;j++){
1.220 brouard 12967: printf("%f ",matcov[i][j]);
12968: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12969: }
1.203 brouard 12970: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12971: }
12972:
12973: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12974: for (i=1;i<=NDIM;i++) {
1.126 brouard 12975: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12976: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12977: }
1.302 brouard 12978: lsurv=vector(agegomp,AGESUP);
12979: lpop=vector(agegomp,AGESUP);
12980: tpop=vector(agegomp,AGESUP);
1.126 brouard 12981: lsurv[agegomp]=100000;
12982:
12983: for (k=agegomp;k<=AGESUP;k++) {
12984: agemortsup=k;
12985: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12986: }
12987:
12988: for (k=agegomp;k<agemortsup;k++)
12989: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12990:
12991: for (k=agegomp;k<agemortsup;k++){
12992: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12993: sumlpop=sumlpop+lpop[k];
12994: }
12995:
12996: tpop[agegomp]=sumlpop;
12997: for (k=agegomp;k<(agemortsup-3);k++){
12998: /* tpop[k+1]=2;*/
12999: tpop[k+1]=tpop[k]-lpop[k];
13000: }
13001:
13002:
13003: printf("\nAge lx qx dx Lx Tx e(x)\n");
13004: for (k=agegomp;k<(agemortsup-2);k++)
13005: 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]);
13006:
13007:
13008: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 13009: ageminpar=50;
13010: agemaxpar=100;
1.194 brouard 13011: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
13012: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13013: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13014: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
13015: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13016: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13017: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13018: }else{
13019: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
13020: 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 13021: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 13022: }
1.201 brouard 13023: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 13024: stepm, weightopt,\
13025: model,imx,p,matcov,agemortsup);
13026:
1.302 brouard 13027: free_vector(lsurv,agegomp,AGESUP);
13028: free_vector(lpop,agegomp,AGESUP);
13029: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 13030: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 13031: free_ivector(dcwave,firstobs,lastobs);
13032: free_vector(agecens,firstobs,lastobs);
13033: free_vector(ageexmed,firstobs,lastobs);
13034: free_ivector(cens,firstobs,lastobs);
1.220 brouard 13035: #ifdef GSL
1.136 brouard 13036: #endif
1.186 brouard 13037: } /* Endof if mle==-3 mortality only */
1.205 brouard 13038: /* Standard */
13039: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
13040: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13041: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 13042: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 13043: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
13044: for (k=1; k<=npar;k++)
13045: printf(" %d %8.5f",k,p[k]);
13046: printf("\n");
1.205 brouard 13047: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
13048: /* mlikeli uses func not funcone */
1.247 brouard 13049: /* for(i=1;i<nlstate;i++){ */
13050: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
13051: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
13052: /* } */
1.205 brouard 13053: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
13054: }
13055: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
13056: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13057: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
13058: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13059: }
13060: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 13061: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13062: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 ! brouard 13063: /* exit(0); */
1.126 brouard 13064: for (k=1; k<=npar;k++)
13065: printf(" %d %8.5f",k,p[k]);
13066: printf("\n");
13067:
13068: /*--------- results files --------------*/
1.283 brouard 13069: /* 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 13070:
13071:
13072: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13073: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 13074: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13075:
13076: printf("#model= 1 + age ");
13077: fprintf(ficres,"#model= 1 + age ");
13078: fprintf(ficlog,"#model= 1 + age ");
13079: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
13080: </ul>", model);
13081:
13082: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
13083: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
13084: if(nagesqr==1){
13085: printf(" + age*age ");
13086: fprintf(ficres," + age*age ");
13087: fprintf(ficlog," + age*age ");
13088: fprintf(fichtm, "<th>+ age*age</th>");
13089: }
13090: for(j=1;j <=ncovmodel-2;j++){
13091: if(Typevar[j]==0) {
13092: printf(" + V%d ",Tvar[j]);
13093: fprintf(ficres," + V%d ",Tvar[j]);
13094: fprintf(ficlog," + V%d ",Tvar[j]);
13095: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13096: }else if(Typevar[j]==1) {
13097: printf(" + V%d*age ",Tvar[j]);
13098: fprintf(ficres," + V%d*age ",Tvar[j]);
13099: fprintf(ficlog," + V%d*age ",Tvar[j]);
13100: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13101: }else if(Typevar[j]==2) {
13102: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13103: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13104: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13105: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13106: }
13107: }
13108: printf("\n");
13109: fprintf(ficres,"\n");
13110: fprintf(ficlog,"\n");
13111: fprintf(fichtm, "</tr>");
13112: fprintf(fichtm, "\n");
13113:
13114:
1.126 brouard 13115: for(i=1,jk=1; i <=nlstate; i++){
13116: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 13117: if (k != i) {
1.319 brouard 13118: fprintf(fichtm, "<tr>");
1.225 brouard 13119: printf("%d%d ",i,k);
13120: fprintf(ficlog,"%d%d ",i,k);
13121: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 13122: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13123: for(j=1; j <=ncovmodel; j++){
13124: printf("%12.7f ",p[jk]);
13125: fprintf(ficlog,"%12.7f ",p[jk]);
13126: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 13127: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 13128: jk++;
13129: }
13130: printf("\n");
13131: fprintf(ficlog,"\n");
13132: fprintf(ficres,"\n");
1.319 brouard 13133: fprintf(fichtm, "</tr>\n");
1.225 brouard 13134: }
1.126 brouard 13135: }
13136: }
1.319 brouard 13137: /* fprintf(fichtm,"</tr>\n"); */
13138: fprintf(fichtm,"</table>\n");
13139: fprintf(fichtm, "\n");
13140:
1.203 brouard 13141: if(mle != 0){
13142: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 13143: ftolhess=ftol; /* Usually correct */
1.203 brouard 13144: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
13145: 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");
13146: 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 13147: 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 13148: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
13149: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
13150: if(nagesqr==1){
13151: printf(" + age*age ");
13152: fprintf(ficres," + age*age ");
13153: fprintf(ficlog," + age*age ");
13154: fprintf(fichtm, "<th>+ age*age</th>");
13155: }
13156: for(j=1;j <=ncovmodel-2;j++){
13157: if(Typevar[j]==0) {
13158: printf(" + V%d ",Tvar[j]);
13159: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13160: }else if(Typevar[j]==1) {
13161: printf(" + V%d*age ",Tvar[j]);
13162: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13163: }else if(Typevar[j]==2) {
13164: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13165: }
13166: }
13167: fprintf(fichtm, "</tr>\n");
13168:
1.203 brouard 13169: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 13170: for(k=1; k <=(nlstate+ndeath); k++){
13171: if (k != i) {
1.319 brouard 13172: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 13173: printf("%d%d ",i,k);
13174: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 13175: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13176: for(j=1; j <=ncovmodel; j++){
1.319 brouard 13177: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 13178: 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]));
13179: 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 13180: if(fabs(wald) > 1.96){
1.321 brouard 13181: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 13182: }else{
13183: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
13184: }
1.324 brouard 13185: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 13186: 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 13187: jk++;
13188: }
13189: printf("\n");
13190: fprintf(ficlog,"\n");
1.319 brouard 13191: fprintf(fichtm, "</tr>\n");
1.225 brouard 13192: }
13193: }
1.193 brouard 13194: }
1.203 brouard 13195: } /* end of hesscov and Wald tests */
1.319 brouard 13196: fprintf(fichtm,"</table>\n");
1.225 brouard 13197:
1.203 brouard 13198: /* */
1.126 brouard 13199: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
13200: printf("# Scales (for hessian or gradient estimation)\n");
13201: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
13202: for(i=1,jk=1; i <=nlstate; i++){
13203: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 13204: if (j!=i) {
13205: fprintf(ficres,"%1d%1d",i,j);
13206: printf("%1d%1d",i,j);
13207: fprintf(ficlog,"%1d%1d",i,j);
13208: for(k=1; k<=ncovmodel;k++){
13209: printf(" %.5e",delti[jk]);
13210: fprintf(ficlog," %.5e",delti[jk]);
13211: fprintf(ficres," %.5e",delti[jk]);
13212: jk++;
13213: }
13214: printf("\n");
13215: fprintf(ficlog,"\n");
13216: fprintf(ficres,"\n");
13217: }
1.126 brouard 13218: }
13219: }
13220:
13221: 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 13222: if(mle >= 1) /* To big for the screen */
1.126 brouard 13223: 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");
13224: 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");
13225: /* # 121 Var(a12)\n\ */
13226: /* # 122 Cov(b12,a12) Var(b12)\n\ */
13227: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
13228: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
13229: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
13230: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
13231: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
13232: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
13233:
13234:
13235: /* Just to have a covariance matrix which will be more understandable
13236: even is we still don't want to manage dictionary of variables
13237: */
13238: for(itimes=1;itimes<=2;itimes++){
13239: jj=0;
13240: for(i=1; i <=nlstate; i++){
1.225 brouard 13241: for(j=1; j <=nlstate+ndeath; j++){
13242: if(j==i) continue;
13243: for(k=1; k<=ncovmodel;k++){
13244: jj++;
13245: ca[0]= k+'a'-1;ca[1]='\0';
13246: if(itimes==1){
13247: if(mle>=1)
13248: printf("#%1d%1d%d",i,j,k);
13249: fprintf(ficlog,"#%1d%1d%d",i,j,k);
13250: fprintf(ficres,"#%1d%1d%d",i,j,k);
13251: }else{
13252: if(mle>=1)
13253: printf("%1d%1d%d",i,j,k);
13254: fprintf(ficlog,"%1d%1d%d",i,j,k);
13255: fprintf(ficres,"%1d%1d%d",i,j,k);
13256: }
13257: ll=0;
13258: for(li=1;li <=nlstate; li++){
13259: for(lj=1;lj <=nlstate+ndeath; lj++){
13260: if(lj==li) continue;
13261: for(lk=1;lk<=ncovmodel;lk++){
13262: ll++;
13263: if(ll<=jj){
13264: cb[0]= lk +'a'-1;cb[1]='\0';
13265: if(ll<jj){
13266: if(itimes==1){
13267: if(mle>=1)
13268: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13269: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13270: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13271: }else{
13272: if(mle>=1)
13273: printf(" %.5e",matcov[jj][ll]);
13274: fprintf(ficlog," %.5e",matcov[jj][ll]);
13275: fprintf(ficres," %.5e",matcov[jj][ll]);
13276: }
13277: }else{
13278: if(itimes==1){
13279: if(mle>=1)
13280: printf(" Var(%s%1d%1d)",ca,i,j);
13281: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
13282: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
13283: }else{
13284: if(mle>=1)
13285: printf(" %.7e",matcov[jj][ll]);
13286: fprintf(ficlog," %.7e",matcov[jj][ll]);
13287: fprintf(ficres," %.7e",matcov[jj][ll]);
13288: }
13289: }
13290: }
13291: } /* end lk */
13292: } /* end lj */
13293: } /* end li */
13294: if(mle>=1)
13295: printf("\n");
13296: fprintf(ficlog,"\n");
13297: fprintf(ficres,"\n");
13298: numlinepar++;
13299: } /* end k*/
13300: } /*end j */
1.126 brouard 13301: } /* end i */
13302: } /* end itimes */
13303:
13304: fflush(ficlog);
13305: fflush(ficres);
1.225 brouard 13306: while(fgets(line, MAXLINE, ficpar)) {
13307: /* If line starts with a # it is a comment */
13308: if (line[0] == '#') {
13309: numlinepar++;
13310: fputs(line,stdout);
13311: fputs(line,ficparo);
13312: fputs(line,ficlog);
1.299 brouard 13313: fputs(line,ficres);
1.225 brouard 13314: continue;
13315: }else
13316: break;
13317: }
13318:
1.209 brouard 13319: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
13320: /* ungetc(c,ficpar); */
13321: /* fgets(line, MAXLINE, ficpar); */
13322: /* fputs(line,stdout); */
13323: /* fputs(line,ficparo); */
13324: /* } */
13325: /* ungetc(c,ficpar); */
1.126 brouard 13326:
13327: estepm=0;
1.209 brouard 13328: 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 13329:
13330: if (num_filled != 6) {
13331: 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);
13332: 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);
13333: goto end;
13334: }
13335: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
13336: }
13337: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
13338: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
13339:
1.209 brouard 13340: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 13341: if (estepm==0 || estepm < stepm) estepm=stepm;
13342: if (fage <= 2) {
13343: bage = ageminpar;
13344: fage = agemaxpar;
13345: }
13346:
13347: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 13348: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
13349: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 13350:
1.186 brouard 13351: /* Other stuffs, more or less useful */
1.254 brouard 13352: while(fgets(line, MAXLINE, ficpar)) {
13353: /* If line starts with a # it is a comment */
13354: if (line[0] == '#') {
13355: numlinepar++;
13356: fputs(line,stdout);
13357: fputs(line,ficparo);
13358: fputs(line,ficlog);
1.299 brouard 13359: fputs(line,ficres);
1.254 brouard 13360: continue;
13361: }else
13362: break;
13363: }
13364:
13365: 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){
13366:
13367: if (num_filled != 7) {
13368: 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);
13369: 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);
13370: goto end;
13371: }
13372: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13373: 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);
13374: 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);
13375: 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 13376: }
1.254 brouard 13377:
13378: while(fgets(line, MAXLINE, ficpar)) {
13379: /* If line starts with a # it is a comment */
13380: if (line[0] == '#') {
13381: numlinepar++;
13382: fputs(line,stdout);
13383: fputs(line,ficparo);
13384: fputs(line,ficlog);
1.299 brouard 13385: fputs(line,ficres);
1.254 brouard 13386: continue;
13387: }else
13388: break;
1.126 brouard 13389: }
13390:
13391:
13392: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13393: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13394:
1.254 brouard 13395: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13396: if (num_filled != 1) {
13397: 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);
13398: 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);
13399: goto end;
13400: }
13401: printf("pop_based=%d\n",popbased);
13402: fprintf(ficlog,"pop_based=%d\n",popbased);
13403: fprintf(ficparo,"pop_based=%d\n",popbased);
13404: fprintf(ficres,"pop_based=%d\n",popbased);
13405: }
13406:
1.258 brouard 13407: /* Results */
1.332 brouard 13408: /* Value of covariate in each resultine will be compututed (if product) and sorted according to model rank */
13409: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
13410: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 13411: endishere=0;
1.258 brouard 13412: nresult=0;
1.308 brouard 13413: parameterline=0;
1.258 brouard 13414: do{
13415: if(!fgets(line, MAXLINE, ficpar)){
13416: endishere=1;
1.308 brouard 13417: parameterline=15;
1.258 brouard 13418: }else if (line[0] == '#') {
13419: /* If line starts with a # it is a comment */
1.254 brouard 13420: numlinepar++;
13421: fputs(line,stdout);
13422: fputs(line,ficparo);
13423: fputs(line,ficlog);
1.299 brouard 13424: fputs(line,ficres);
1.254 brouard 13425: continue;
1.258 brouard 13426: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13427: parameterline=11;
1.296 brouard 13428: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 13429: parameterline=12;
1.307 brouard 13430: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 13431: parameterline=13;
1.307 brouard 13432: }
1.258 brouard 13433: else{
13434: parameterline=14;
1.254 brouard 13435: }
1.308 brouard 13436: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 13437: case 11:
1.296 brouard 13438: 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)){
13439: 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 13440: 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);
13441: 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);
13442: 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);
13443: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 13444: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
13445: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 13446: prvforecast = 1;
13447: }
13448: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 13449: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13450: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13451: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 13452: prvforecast = 2;
13453: }
13454: else {
13455: 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);
13456: 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);
13457: goto end;
1.258 brouard 13458: }
1.254 brouard 13459: break;
1.258 brouard 13460: case 12:
1.296 brouard 13461: 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)){
13462: 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);
13463: 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);
13464: 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);
13465: 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);
13466: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 13467: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
13468: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 13469: prvbackcast = 1;
13470: }
13471: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 13472: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13473: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13474: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 13475: prvbackcast = 2;
13476: }
13477: else {
13478: 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);
13479: 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);
13480: goto end;
1.258 brouard 13481: }
1.230 brouard 13482: break;
1.258 brouard 13483: case 13:
1.332 brouard 13484: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 13485: nresult++; /* Sum of resultlines */
1.332 brouard 13486: printf("Result %d: result:%s\n",nresult, resultlineori);
13487: /* removefirstspace(&resultlineori); */
13488:
13489: if(strstr(resultlineori,"v") !=0){
13490: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
13491: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
13492: return 1;
13493: }
13494: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
13495: printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori);
1.318 brouard 13496: if(nresult > MAXRESULTLINESPONE-1){
13497: 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);
13498: 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 13499: goto end;
13500: }
1.332 brouard 13501:
1.310 brouard 13502: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 13503: fprintf(ficparo,"result: %s\n",resultline);
13504: fprintf(ficres,"result: %s\n",resultline);
13505: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 13506: } else
13507: goto end;
1.307 brouard 13508: break;
13509: case 14:
13510: printf("Error: Unknown command '%s'\n",line);
13511: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 13512: if(line[0] == ' ' || line[0] == '\n'){
13513: printf("It should not be an empty line '%s'\n",line);
13514: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
13515: }
1.307 brouard 13516: if(ncovmodel >=2 && nresult==0 ){
13517: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
13518: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 13519: }
1.307 brouard 13520: /* goto end; */
13521: break;
1.308 brouard 13522: case 15:
13523: printf("End of resultlines.\n");
13524: fprintf(ficlog,"End of resultlines.\n");
13525: break;
13526: default: /* parameterline =0 */
1.307 brouard 13527: nresult=1;
13528: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 13529: } /* End switch parameterline */
13530: }while(endishere==0); /* End do */
1.126 brouard 13531:
1.230 brouard 13532: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 13533: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 13534:
13535: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 13536: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 13537: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13538: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13539: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 13540: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13541: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13542: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13543: }else{
1.270 brouard 13544: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 13545: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
13546: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
13547: if(prvforecast==1){
13548: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
13549: jprojd=jproj1;
13550: mprojd=mproj1;
13551: anprojd=anproj1;
13552: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
13553: jprojf=jproj2;
13554: mprojf=mproj2;
13555: anprojf=anproj2;
13556: } else if(prvforecast == 2){
13557: dateprojd=dateintmean;
13558: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
13559: dateprojf=dateintmean+yrfproj;
13560: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
13561: }
13562: if(prvbackcast==1){
13563: datebackd=(jback1+12*mback1+365*anback1)/365;
13564: jbackd=jback1;
13565: mbackd=mback1;
13566: anbackd=anback1;
13567: datebackf=(jback2+12*mback2+365*anback2)/365;
13568: jbackf=jback2;
13569: mbackf=mback2;
13570: anbackf=anback2;
13571: } else if(prvbackcast == 2){
13572: datebackd=dateintmean;
13573: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
13574: datebackf=dateintmean-yrbproj;
13575: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
13576: }
13577:
13578: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 13579: }
13580: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 13581: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
13582: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 13583:
1.225 brouard 13584: /*------------ free_vector -------------*/
13585: /* chdir(path); */
1.220 brouard 13586:
1.215 brouard 13587: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
13588: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
13589: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
13590: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 13591: free_lvector(num,firstobs,lastobs);
13592: free_vector(agedc,firstobs,lastobs);
1.126 brouard 13593: /*free_matrix(covar,0,NCOVMAX,1,n);*/
13594: /*free_matrix(covar,1,NCOVMAX,1,n);*/
13595: fclose(ficparo);
13596: fclose(ficres);
1.220 brouard 13597:
13598:
1.186 brouard 13599: /* Other results (useful)*/
1.220 brouard 13600:
13601:
1.126 brouard 13602: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13603: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13604: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 13605: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 13606: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13607: fclose(ficrespl);
13608:
13609: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13610: /*#include "hpijx.h"*/
1.332 brouard 13611: /** 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?*/
13612: /* calls hpxij with combination k */
1.180 brouard 13613: hPijx(p, bage, fage);
1.145 brouard 13614: fclose(ficrespij);
1.227 brouard 13615:
1.220 brouard 13616: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 13617: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 13618: k=1;
1.126 brouard 13619: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13620:
1.269 brouard 13621: /* Prevalence for each covariate combination in probs[age][status][cov] */
13622: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13623: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 13624: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 13625: for(k=1;k<=ncovcombmax;k++)
13626: probs[i][j][k]=0.;
1.269 brouard 13627: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
13628: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 13629: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 13630: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13631: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 13632: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 13633: for(k=1;k<=ncovcombmax;k++)
13634: mobaverages[i][j][k]=0.;
1.219 brouard 13635: mobaverage=mobaverages;
13636: if (mobilav!=0) {
1.235 brouard 13637: printf("Movingaveraging observed prevalence\n");
1.258 brouard 13638: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 13639: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
13640: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
13641: printf(" Error in movingaverage mobilav=%d\n",mobilav);
13642: }
1.269 brouard 13643: } else if (mobilavproj !=0) {
1.235 brouard 13644: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 13645: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 13646: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
13647: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
13648: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
13649: }
1.269 brouard 13650: }else{
13651: printf("Internal error moving average\n");
13652: fflush(stdout);
13653: exit(1);
1.219 brouard 13654: }
13655: }/* end if moving average */
1.227 brouard 13656:
1.126 brouard 13657: /*---------- Forecasting ------------------*/
1.296 brouard 13658: if(prevfcast==1){
13659: /* /\* if(stepm ==1){*\/ */
13660: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13661: /*This done previously after freqsummary.*/
13662: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
13663: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
13664:
13665: /* } else if (prvforecast==2){ */
13666: /* /\* if(stepm ==1){*\/ */
13667: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13668: /* } */
13669: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
13670: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 13671: }
1.269 brouard 13672:
1.296 brouard 13673: /* Prevbcasting */
13674: if(prevbcast==1){
1.219 brouard 13675: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13676: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13677: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13678:
13679: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
13680:
13681: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 13682:
1.219 brouard 13683: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
13684: fclose(ficresplb);
13685:
1.222 brouard 13686: hBijx(p, bage, fage, mobaverage);
13687: fclose(ficrespijb);
1.219 brouard 13688:
1.296 brouard 13689: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
13690: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
13691: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
13692: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
13693: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
13694: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
13695:
13696:
1.269 brouard 13697: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13698:
13699:
1.269 brouard 13700: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 13701: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13702: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13703: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 13704: } /* end Prevbcasting */
1.268 brouard 13705:
1.186 brouard 13706:
13707: /* ------ Other prevalence ratios------------ */
1.126 brouard 13708:
1.215 brouard 13709: free_ivector(wav,1,imx);
13710: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
13711: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
13712: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 13713:
13714:
1.127 brouard 13715: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 13716:
1.201 brouard 13717: strcpy(filerese,"E_");
13718: strcat(filerese,fileresu);
1.126 brouard 13719: if((ficreseij=fopen(filerese,"w"))==NULL) {
13720: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13721: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13722: }
1.208 brouard 13723: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
13724: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 13725:
13726: pstamp(ficreseij);
1.219 brouard 13727:
1.235 brouard 13728: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13729: if (cptcovn < 1){i1=1;}
13730:
13731: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13732: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13733: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13734: continue;
1.219 brouard 13735: fprintf(ficreseij,"\n#****** ");
1.235 brouard 13736: printf("\n#****** ");
1.225 brouard 13737: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 13738: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
13739: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 13740: }
13741: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.334 brouard 13742: printf(" V%d=%f ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
13743: fprintf(ficreseij,"V%d=%f ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 13744: }
13745: fprintf(ficreseij,"******\n");
1.235 brouard 13746: printf("******\n");
1.219 brouard 13747:
13748: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13749: oldm=oldms;savm=savms;
1.330 brouard 13750: /* 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 13751: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 13752:
1.219 brouard 13753: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 13754: }
13755: fclose(ficreseij);
1.208 brouard 13756: printf("done evsij\n");fflush(stdout);
13757: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 13758:
1.218 brouard 13759:
1.227 brouard 13760: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 13761:
1.201 brouard 13762: strcpy(filerest,"T_");
13763: strcat(filerest,fileresu);
1.127 brouard 13764: if((ficrest=fopen(filerest,"w"))==NULL) {
13765: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
13766: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
13767: }
1.208 brouard 13768: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
13769: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 13770: strcpy(fileresstde,"STDE_");
13771: strcat(fileresstde,fileresu);
1.126 brouard 13772: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 13773: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
13774: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 13775: }
1.227 brouard 13776: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
13777: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 13778:
1.201 brouard 13779: strcpy(filerescve,"CVE_");
13780: strcat(filerescve,fileresu);
1.126 brouard 13781: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 13782: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
13783: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 13784: }
1.227 brouard 13785: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
13786: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 13787:
1.201 brouard 13788: strcpy(fileresv,"V_");
13789: strcat(fileresv,fileresu);
1.126 brouard 13790: if((ficresvij=fopen(fileresv,"w"))==NULL) {
13791: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
13792: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
13793: }
1.227 brouard 13794: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
13795: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 13796:
1.235 brouard 13797: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13798: if (cptcovn < 1){i1=1;}
13799:
1.334 brouard 13800: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
13801: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
13802: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
13803: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
13804: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
13805: /* */
13806: 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 13807: continue;
1.321 brouard 13808: printf("\n# model %s \n#****** Result for:", model);
13809: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
13810: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.334 brouard 13811: /* It might not be a good idea to mix dummies and quantitative */
13812: /* 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 *\/ */
13813: 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 */
13814: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
13815: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
13816: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
13817: * (V5 is quanti) V4 and V3 are dummies
13818: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
13819: * l=1 l=2
13820: * k=1 1 1 0 0
13821: * k=2 2 1 1 0
13822: * k=3 [1] [2] 0 1
13823: * k=4 2 2 1 1
13824: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
13825: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
13826: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
13827: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
13828: */
13829: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
13830: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
13831: /* We give up with the combinations!! */
13832: 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 */
13833:
13834: if(Dummy[modelresult[nres][j]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to j in resultline */
13835: printf("V%d=%d ",Tvresult[nres][j],Tresult[nres][j]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
13836: fprintf(ficlog,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
13837: fprintf(ficrest,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
13838: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
13839: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
13840: }else{
13841: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
13842: }
13843: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13844: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13845: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
13846: /* For each selected (single) quantitative value */
13847: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13848: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
13849: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
13850: }else{
13851: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
13852: }
13853: }else{
13854: 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 */
13855: 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 */
13856: exit(1);
13857: }
1.335 ! brouard 13858: } /* End loop for each variable in the resultline */
1.334 brouard 13859: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
13860: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
13861: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
13862: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
13863: /* } */
1.208 brouard 13864: fprintf(ficrest,"******\n");
1.227 brouard 13865: fprintf(ficlog,"******\n");
13866: printf("******\n");
1.208 brouard 13867:
13868: fprintf(ficresstdeij,"\n#****** ");
13869: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 13870: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 13871: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
13872: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13873: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13874: }
13875: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value, TvarsQind gives the position of a quantitative in model equation */
13876: fprintf(ficresstdeij," V%d=%f ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
13877: fprintf(ficrescveij," V%d=%f ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 13878: }
1.208 brouard 13879: fprintf(ficresstdeij,"******\n");
13880: fprintf(ficrescveij,"******\n");
13881:
13882: fprintf(ficresvij,"\n#****** ");
1.238 brouard 13883: /* pstamp(ficresvij); */
1.225 brouard 13884: for(j=1;j<=cptcoveff;j++)
1.335 ! brouard 13885: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
! 13886: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 13887: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 13888: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
13889: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 13890: }
1.208 brouard 13891: fprintf(ficresvij,"******\n");
13892:
13893: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13894: oldm=oldms;savm=savms;
1.235 brouard 13895: printf(" cvevsij ");
13896: fprintf(ficlog, " cvevsij ");
13897: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 13898: printf(" end cvevsij \n ");
13899: fprintf(ficlog, " end cvevsij \n ");
13900:
13901: /*
13902: */
13903: /* goto endfree; */
13904:
13905: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13906: pstamp(ficrest);
13907:
1.269 brouard 13908: epj=vector(1,nlstate+1);
1.208 brouard 13909: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 13910: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
13911: cptcod= 0; /* To be deleted */
13912: printf("varevsij vpopbased=%d \n",vpopbased);
13913: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 13914: 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 13915: 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 ");
13916: if(vpopbased==1)
13917: 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);
13918: else
1.288 brouard 13919: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.335 ! brouard 13920: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 13921: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
13922: fprintf(ficrest,"\n");
13923: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 13924: printf("Computing age specific forward period (stable) prevalences in each health state \n");
13925: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13926: for(age=bage; age <=fage ;age++){
1.235 brouard 13927: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 13928: if (vpopbased==1) {
13929: if(mobilav ==0){
13930: for(i=1; i<=nlstate;i++)
13931: prlim[i][i]=probs[(int)age][i][k];
13932: }else{ /* mobilav */
13933: for(i=1; i<=nlstate;i++)
13934: prlim[i][i]=mobaverage[(int)age][i][k];
13935: }
13936: }
1.219 brouard 13937:
1.227 brouard 13938: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
13939: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
13940: /* printf(" age %4.0f ",age); */
13941: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
13942: for(i=1, epj[j]=0.;i <=nlstate;i++) {
13943: epj[j] += prlim[i][i]*eij[i][j][(int)age];
13944: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
13945: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
13946: }
13947: epj[nlstate+1] +=epj[j];
13948: }
13949: /* printf(" age %4.0f \n",age); */
1.219 brouard 13950:
1.227 brouard 13951: for(i=1, vepp=0.;i <=nlstate;i++)
13952: for(j=1;j <=nlstate;j++)
13953: vepp += vareij[i][j][(int)age];
13954: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
13955: for(j=1;j <=nlstate;j++){
13956: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
13957: }
13958: fprintf(ficrest,"\n");
13959: }
1.208 brouard 13960: } /* End vpopbased */
1.269 brouard 13961: free_vector(epj,1,nlstate+1);
1.208 brouard 13962: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
13963: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 13964: printf("done selection\n");fflush(stdout);
13965: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 13966:
1.335 ! brouard 13967: } /* End k selection or end covariate selection for nres */
1.227 brouard 13968:
13969: printf("done State-specific expectancies\n");fflush(stdout);
13970: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
13971:
1.335 ! brouard 13972: /* variance-covariance of forward period prevalence */
1.269 brouard 13973: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13974:
1.227 brouard 13975:
1.290 brouard 13976: free_vector(weight,firstobs,lastobs);
1.330 brouard 13977: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 13978: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 13979: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
13980: free_matrix(anint,1,maxwav,firstobs,lastobs);
13981: free_matrix(mint,1,maxwav,firstobs,lastobs);
13982: free_ivector(cod,firstobs,lastobs);
1.227 brouard 13983: free_ivector(tab,1,NCOVMAX);
13984: fclose(ficresstdeij);
13985: fclose(ficrescveij);
13986: fclose(ficresvij);
13987: fclose(ficrest);
13988: fclose(ficpar);
13989:
13990:
1.126 brouard 13991: /*---------- End : free ----------------*/
1.219 brouard 13992: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 13993: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
13994: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 13995: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
13996: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 13997: } /* mle==-3 arrives here for freeing */
1.227 brouard 13998: /* endfree:*/
13999: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
14000: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
14001: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 14002: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
14003: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
14004: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
14005: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 14006: free_matrix(matcov,1,npar,1,npar);
14007: free_matrix(hess,1,npar,1,npar);
14008: /*free_vector(delti,1,npar);*/
14009: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14010: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 14011: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 14012: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14013:
14014: free_ivector(ncodemax,1,NCOVMAX);
14015: free_ivector(ncodemaxwundef,1,NCOVMAX);
14016: free_ivector(Dummy,-1,NCOVMAX);
14017: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 14018: free_ivector(DummyV,1,NCOVMAX);
14019: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 14020: free_ivector(Typevar,-1,NCOVMAX);
14021: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 14022: free_ivector(TvarsQ,1,NCOVMAX);
14023: free_ivector(TvarsQind,1,NCOVMAX);
14024: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 14025: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 14026: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 14027: free_ivector(TvarFD,1,NCOVMAX);
14028: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 14029: free_ivector(TvarF,1,NCOVMAX);
14030: free_ivector(TvarFind,1,NCOVMAX);
14031: free_ivector(TvarV,1,NCOVMAX);
14032: free_ivector(TvarVind,1,NCOVMAX);
14033: free_ivector(TvarA,1,NCOVMAX);
14034: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 14035: free_ivector(TvarFQ,1,NCOVMAX);
14036: free_ivector(TvarFQind,1,NCOVMAX);
14037: free_ivector(TvarVD,1,NCOVMAX);
14038: free_ivector(TvarVDind,1,NCOVMAX);
14039: free_ivector(TvarVQ,1,NCOVMAX);
14040: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 14041: free_ivector(Tvarsel,1,NCOVMAX);
14042: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 14043: free_ivector(Tposprod,1,NCOVMAX);
14044: free_ivector(Tprod,1,NCOVMAX);
14045: free_ivector(Tvaraff,1,NCOVMAX);
14046: free_ivector(invalidvarcomb,1,ncovcombmax);
14047: free_ivector(Tage,1,NCOVMAX);
14048: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 14049: free_ivector(TmodelInvind,1,NCOVMAX);
14050: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 14051:
14052: free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
14053:
1.227 brouard 14054: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
14055: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 14056: fflush(fichtm);
14057: fflush(ficgp);
14058:
1.227 brouard 14059:
1.126 brouard 14060: if((nberr >0) || (nbwarn>0)){
1.216 brouard 14061: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
14062: 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 14063: }else{
14064: printf("End of Imach\n");
14065: fprintf(ficlog,"End of Imach\n");
14066: }
14067: printf("See log file on %s\n",filelog);
14068: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 14069: /*(void) gettimeofday(&end_time,&tzp);*/
14070: rend_time = time(NULL);
14071: end_time = *localtime(&rend_time);
14072: /* tml = *localtime(&end_time.tm_sec); */
14073: strcpy(strtend,asctime(&end_time));
1.126 brouard 14074: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
14075: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 14076: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 14077:
1.157 brouard 14078: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
14079: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
14080: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 14081: /* printf("Total time was %d uSec.\n", total_usecs);*/
14082: /* if(fileappend(fichtm,optionfilehtm)){ */
14083: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14084: fclose(fichtm);
14085: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14086: fclose(fichtmcov);
14087: fclose(ficgp);
14088: fclose(ficlog);
14089: /*------ End -----------*/
1.227 brouard 14090:
1.281 brouard 14091:
14092: /* Executes gnuplot */
1.227 brouard 14093:
14094: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 14095: #ifdef WIN32
1.227 brouard 14096: if (_chdir(pathcd) != 0)
14097: printf("Can't move to directory %s!\n",path);
14098: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 14099: #else
1.227 brouard 14100: if(chdir(pathcd) != 0)
14101: printf("Can't move to directory %s!\n", path);
14102: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 14103: #endif
1.126 brouard 14104: printf("Current directory %s!\n",pathcd);
14105: /*strcat(plotcmd,CHARSEPARATOR);*/
14106: sprintf(plotcmd,"gnuplot");
1.157 brouard 14107: #ifdef _WIN32
1.126 brouard 14108: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
14109: #endif
14110: if(!stat(plotcmd,&info)){
1.158 brouard 14111: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14112: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 14113: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 14114: }else
14115: strcpy(pplotcmd,plotcmd);
1.157 brouard 14116: #ifdef __unix
1.126 brouard 14117: strcpy(plotcmd,GNUPLOTPROGRAM);
14118: if(!stat(plotcmd,&info)){
1.158 brouard 14119: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14120: }else
14121: strcpy(pplotcmd,plotcmd);
14122: #endif
14123: }else
14124: strcpy(pplotcmd,plotcmd);
14125:
14126: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 14127: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 14128: strcpy(pplotcmd,plotcmd);
1.227 brouard 14129:
1.126 brouard 14130: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 14131: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 14132: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 14133: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 14134: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 14135: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 14136: strcpy(plotcmd,pplotcmd);
14137: }
1.126 brouard 14138: }
1.158 brouard 14139: printf(" Successful, please wait...");
1.126 brouard 14140: while (z[0] != 'q') {
14141: /* chdir(path); */
1.154 brouard 14142: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 14143: scanf("%s",z);
14144: /* if (z[0] == 'c') system("./imach"); */
14145: if (z[0] == 'e') {
1.158 brouard 14146: #ifdef __APPLE__
1.152 brouard 14147: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 14148: #elif __linux
14149: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 14150: #else
1.152 brouard 14151: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 14152: #endif
14153: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
14154: system(pplotcmd);
1.126 brouard 14155: }
14156: else if (z[0] == 'g') system(plotcmd);
14157: else if (z[0] == 'q') exit(0);
14158: }
1.227 brouard 14159: end:
1.126 brouard 14160: while (z[0] != 'q') {
1.195 brouard 14161: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 14162: scanf("%s",z);
14163: }
1.283 brouard 14164: printf("End\n");
1.282 brouard 14165: exit(0);
1.126 brouard 14166: }
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