Annotation of imach/src/imach.c, revision 1.336
1.336 ! brouard 1: /* $Id: imach.c,v 1.335 2022/08/31 08:23:16 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.336 ! brouard 4: Revision 1.335 2022/08/31 08:23:16 brouard
! 5: Summary: improvements...
! 6:
1.335 brouard 7: Revision 1.334 2022/08/25 09:08:41 brouard
8: Summary: In progress for quantitative
9:
1.334 brouard 10: Revision 1.333 2022/08/21 09:10:30 brouard
11: * src/imach.c (Module): Version 0.99r33 A lot of changes in
12: reassigning covariates: my first idea was that people will always
13: use the first covariate V1 into the model but in fact they are
14: producing data with many covariates and can use an equation model
15: with some of the covariate; it means that in a model V2+V3 instead
16: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
17: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
18: the equation model is restricted to two variables only (V2, V3)
19: and the combination for V2 should be codtabm(k,1) instead of
20: (codtabm(k,2), and the code should be
21: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
22: made. All of these should be simplified once a day like we did in
23: hpxij() for example by using precov[nres] which is computed in
24: decoderesult for each nres of each resultline. Loop should be done
25: on the equation model globally by distinguishing only product with
26: age (which are changing with age) and no more on type of
27: covariates, single dummies, single covariates.
28:
1.333 brouard 29: Revision 1.332 2022/08/21 09:06:25 brouard
30: Summary: Version 0.99r33
31:
32: * src/imach.c (Module): Version 0.99r33 A lot of changes in
33: reassigning covariates: my first idea was that people will always
34: use the first covariate V1 into the model but in fact they are
35: producing data with many covariates and can use an equation model
36: with some of the covariate; it means that in a model V2+V3 instead
37: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
38: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
39: the equation model is restricted to two variables only (V2, V3)
40: and the combination for V2 should be codtabm(k,1) instead of
41: (codtabm(k,2), and the code should be
42: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
43: made. All of these should be simplified once a day like we did in
44: hpxij() for example by using precov[nres] which is computed in
45: decoderesult for each nres of each resultline. Loop should be done
46: on the equation model globally by distinguishing only product with
47: age (which are changing with age) and no more on type of
48: covariates, single dummies, single covariates.
49:
1.332 brouard 50: Revision 1.331 2022/08/07 05:40:09 brouard
51: *** empty log message ***
52:
1.331 brouard 53: Revision 1.330 2022/08/06 07:18:25 brouard
54: Summary: last 0.99r31
55:
56: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
57:
1.330 brouard 58: Revision 1.329 2022/08/03 17:29:54 brouard
59: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
60:
1.329 brouard 61: Revision 1.328 2022/07/27 17:40:48 brouard
62: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
63:
1.328 brouard 64: Revision 1.327 2022/07/27 14:47:35 brouard
65: Summary: Still a problem for one-step probabilities in case of quantitative variables
66:
1.327 brouard 67: Revision 1.326 2022/07/26 17:33:55 brouard
68: Summary: some test with nres=1
69:
1.326 brouard 70: Revision 1.325 2022/07/25 14:27:23 brouard
71: Summary: r30
72:
73: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
74: coredumped, revealed by Feiuno, thank you.
75:
1.325 brouard 76: Revision 1.324 2022/07/23 17:44:26 brouard
77: *** empty log message ***
78:
1.324 brouard 79: Revision 1.323 2022/07/22 12:30:08 brouard
80: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
81:
1.323 brouard 82: Revision 1.322 2022/07/22 12:27:48 brouard
83: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
84:
1.322 brouard 85: Revision 1.321 2022/07/22 12:04:24 brouard
86: Summary: r28
87:
88: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
89:
1.321 brouard 90: Revision 1.320 2022/06/02 05:10:11 brouard
91: *** empty log message ***
92:
1.320 brouard 93: Revision 1.319 2022/06/02 04:45:11 brouard
94: * imach.c (Module): Adding the Wald tests from the log to the main
95: htm for better display of the maximum likelihood estimators.
96:
1.319 brouard 97: Revision 1.318 2022/05/24 08:10:59 brouard
98: * imach.c (Module): Some attempts to find a bug of wrong estimates
99: of confidencce intervals with product in the equation modelC
100:
1.318 brouard 101: Revision 1.317 2022/05/15 15:06:23 brouard
102: * imach.c (Module): Some minor improvements
103:
1.317 brouard 104: Revision 1.316 2022/05/11 15:11:31 brouard
105: Summary: r27
106:
1.316 brouard 107: Revision 1.315 2022/05/11 15:06:32 brouard
108: *** empty log message ***
109:
1.315 brouard 110: Revision 1.314 2022/04/13 17:43:09 brouard
111: * imach.c (Module): Adding link to text data files
112:
1.314 brouard 113: Revision 1.313 2022/04/11 15:57:42 brouard
114: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
115:
1.313 brouard 116: Revision 1.312 2022/04/05 21:24:39 brouard
117: *** empty log message ***
118:
1.312 brouard 119: Revision 1.311 2022/04/05 21:03:51 brouard
120: Summary: Fixed quantitative covariates
121:
122: Fixed covariates (dummy or quantitative)
123: with missing values have never been allowed but are ERRORS and
124: program quits. Standard deviations of fixed covariates were
125: wrongly computed. Mean and standard deviations of time varying
126: covariates are still not computed.
127:
1.311 brouard 128: Revision 1.310 2022/03/17 08:45:53 brouard
129: Summary: 99r25
130:
131: Improving detection of errors: result lines should be compatible with
132: the model.
133:
1.310 brouard 134: Revision 1.309 2021/05/20 12:39:14 brouard
135: Summary: Version 0.99r24
136:
1.309 brouard 137: Revision 1.308 2021/03/31 13:11:57 brouard
138: Summary: Version 0.99r23
139:
140:
141: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
142:
1.308 brouard 143: Revision 1.307 2021/03/08 18:11:32 brouard
144: Summary: 0.99r22 fixed bug on result:
145:
1.307 brouard 146: Revision 1.306 2021/02/20 15:44:02 brouard
147: Summary: Version 0.99r21
148:
149: * imach.c (Module): Fix bug on quitting after result lines!
150: (Module): Version 0.99r21
151:
1.306 brouard 152: Revision 1.305 2021/02/20 15:28:30 brouard
153: * imach.c (Module): Fix bug on quitting after result lines!
154:
1.305 brouard 155: Revision 1.304 2021/02/12 11:34:20 brouard
156: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
157:
1.304 brouard 158: Revision 1.303 2021/02/11 19:50:15 brouard
159: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
160:
1.303 brouard 161: Revision 1.302 2020/02/22 21:00:05 brouard
162: * (Module): imach.c Update mle=-3 (for computing Life expectancy
163: and life table from the data without any state)
164:
1.302 brouard 165: Revision 1.301 2019/06/04 13:51:20 brouard
166: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
167:
1.301 brouard 168: Revision 1.300 2019/05/22 19:09:45 brouard
169: Summary: version 0.99r19 of May 2019
170:
1.300 brouard 171: Revision 1.299 2019/05/22 18:37:08 brouard
172: Summary: Cleaned 0.99r19
173:
1.299 brouard 174: Revision 1.298 2019/05/22 18:19:56 brouard
175: *** empty log message ***
176:
1.298 brouard 177: Revision 1.297 2019/05/22 17:56:10 brouard
178: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
179:
1.297 brouard 180: Revision 1.296 2019/05/20 13:03:18 brouard
181: Summary: Projection syntax simplified
182:
183:
184: We can now start projections, forward or backward, from the mean date
185: of inteviews up to or down to a number of years of projection:
186: prevforecast=1 yearsfproj=15.3 mobil_average=0
187: or
188: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
189: or
190: prevbackcast=1 yearsbproj=12.3 mobil_average=1
191: or
192: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
193:
1.296 brouard 194: Revision 1.295 2019/05/18 09:52:50 brouard
195: Summary: doxygen tex bug
196:
1.295 brouard 197: Revision 1.294 2019/05/16 14:54:33 brouard
198: Summary: There was some wrong lines added
199:
1.294 brouard 200: Revision 1.293 2019/05/09 15:17:34 brouard
201: *** empty log message ***
202:
1.293 brouard 203: Revision 1.292 2019/05/09 14:17:20 brouard
204: Summary: Some updates
205:
1.292 brouard 206: Revision 1.291 2019/05/09 13:44:18 brouard
207: Summary: Before ncovmax
208:
1.291 brouard 209: Revision 1.290 2019/05/09 13:39:37 brouard
210: Summary: 0.99r18 unlimited number of individuals
211:
212: 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.
213:
1.290 brouard 214: Revision 1.289 2018/12/13 09:16:26 brouard
215: Summary: Bug for young ages (<-30) will be in r17
216:
1.289 brouard 217: Revision 1.288 2018/05/02 20:58:27 brouard
218: Summary: Some bugs fixed
219:
1.288 brouard 220: Revision 1.287 2018/05/01 17:57:25 brouard
221: Summary: Bug fixed by providing frequencies only for non missing covariates
222:
1.287 brouard 223: Revision 1.286 2018/04/27 14:27:04 brouard
224: Summary: some minor bugs
225:
1.286 brouard 226: Revision 1.285 2018/04/21 21:02:16 brouard
227: Summary: Some bugs fixed, valgrind tested
228:
1.285 brouard 229: Revision 1.284 2018/04/20 05:22:13 brouard
230: Summary: Computing mean and stdeviation of fixed quantitative variables
231:
1.284 brouard 232: Revision 1.283 2018/04/19 14:49:16 brouard
233: Summary: Some minor bugs fixed
234:
1.283 brouard 235: Revision 1.282 2018/02/27 22:50:02 brouard
236: *** empty log message ***
237:
1.282 brouard 238: Revision 1.281 2018/02/27 19:25:23 brouard
239: Summary: Adding second argument for quitting
240:
1.281 brouard 241: Revision 1.280 2018/02/21 07:58:13 brouard
242: Summary: 0.99r15
243:
244: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
245:
1.280 brouard 246: Revision 1.279 2017/07/20 13:35:01 brouard
247: Summary: temporary working
248:
1.279 brouard 249: Revision 1.278 2017/07/19 14:09:02 brouard
250: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
251:
1.278 brouard 252: Revision 1.277 2017/07/17 08:53:49 brouard
253: Summary: BOM files can be read now
254:
1.277 brouard 255: Revision 1.276 2017/06/30 15:48:31 brouard
256: Summary: Graphs improvements
257:
1.276 brouard 258: Revision 1.275 2017/06/30 13:39:33 brouard
259: Summary: Saito's color
260:
1.275 brouard 261: Revision 1.274 2017/06/29 09:47:08 brouard
262: Summary: Version 0.99r14
263:
1.274 brouard 264: Revision 1.273 2017/06/27 11:06:02 brouard
265: Summary: More documentation on projections
266:
1.273 brouard 267: Revision 1.272 2017/06/27 10:22:40 brouard
268: Summary: Color of backprojection changed from 6 to 5(yellow)
269:
1.272 brouard 270: Revision 1.271 2017/06/27 10:17:50 brouard
271: Summary: Some bug with rint
272:
1.271 brouard 273: Revision 1.270 2017/05/24 05:45:29 brouard
274: *** empty log message ***
275:
1.270 brouard 276: Revision 1.269 2017/05/23 08:39:25 brouard
277: Summary: Code into subroutine, cleanings
278:
1.269 brouard 279: Revision 1.268 2017/05/18 20:09:32 brouard
280: Summary: backprojection and confidence intervals of backprevalence
281:
1.268 brouard 282: Revision 1.267 2017/05/13 10:25:05 brouard
283: Summary: temporary save for backprojection
284:
1.267 brouard 285: Revision 1.266 2017/05/13 07:26:12 brouard
286: Summary: Version 0.99r13 (improvements and bugs fixed)
287:
1.266 brouard 288: Revision 1.265 2017/04/26 16:22:11 brouard
289: Summary: imach 0.99r13 Some bugs fixed
290:
1.265 brouard 291: Revision 1.264 2017/04/26 06:01:29 brouard
292: Summary: Labels in graphs
293:
1.264 brouard 294: Revision 1.263 2017/04/24 15:23:15 brouard
295: Summary: to save
296:
1.263 brouard 297: Revision 1.262 2017/04/18 16:48:12 brouard
298: *** empty log message ***
299:
1.262 brouard 300: Revision 1.261 2017/04/05 10:14:09 brouard
301: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
302:
1.261 brouard 303: Revision 1.260 2017/04/04 17:46:59 brouard
304: Summary: Gnuplot indexations fixed (humm)
305:
1.260 brouard 306: Revision 1.259 2017/04/04 13:01:16 brouard
307: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
308:
1.259 brouard 309: Revision 1.258 2017/04/03 10:17:47 brouard
310: Summary: Version 0.99r12
311:
312: Some cleanings, conformed with updated documentation.
313:
1.258 brouard 314: Revision 1.257 2017/03/29 16:53:30 brouard
315: Summary: Temp
316:
1.257 brouard 317: Revision 1.256 2017/03/27 05:50:23 brouard
318: Summary: Temporary
319:
1.256 brouard 320: Revision 1.255 2017/03/08 16:02:28 brouard
321: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
322:
1.255 brouard 323: Revision 1.254 2017/03/08 07:13:00 brouard
324: Summary: Fixing data parameter line
325:
1.254 brouard 326: Revision 1.253 2016/12/15 11:59:41 brouard
327: Summary: 0.99 in progress
328:
1.253 brouard 329: Revision 1.252 2016/09/15 21:15:37 brouard
330: *** empty log message ***
331:
1.252 brouard 332: Revision 1.251 2016/09/15 15:01:13 brouard
333: Summary: not working
334:
1.251 brouard 335: Revision 1.250 2016/09/08 16:07:27 brouard
336: Summary: continue
337:
1.250 brouard 338: Revision 1.249 2016/09/07 17:14:18 brouard
339: Summary: Starting values from frequencies
340:
1.249 brouard 341: Revision 1.248 2016/09/07 14:10:18 brouard
342: *** empty log message ***
343:
1.248 brouard 344: Revision 1.247 2016/09/02 11:11:21 brouard
345: *** empty log message ***
346:
1.247 brouard 347: Revision 1.246 2016/09/02 08:49:22 brouard
348: *** empty log message ***
349:
1.246 brouard 350: Revision 1.245 2016/09/02 07:25:01 brouard
351: *** empty log message ***
352:
1.245 brouard 353: Revision 1.244 2016/09/02 07:17:34 brouard
354: *** empty log message ***
355:
1.244 brouard 356: Revision 1.243 2016/09/02 06:45:35 brouard
357: *** empty log message ***
358:
1.243 brouard 359: Revision 1.242 2016/08/30 15:01:20 brouard
360: Summary: Fixing a lots
361:
1.242 brouard 362: Revision 1.241 2016/08/29 17:17:25 brouard
363: Summary: gnuplot problem in Back projection to fix
364:
1.241 brouard 365: Revision 1.240 2016/08/29 07:53:18 brouard
366: Summary: Better
367:
1.240 brouard 368: Revision 1.239 2016/08/26 15:51:03 brouard
369: Summary: Improvement in Powell output in order to copy and paste
370:
371: Author:
372:
1.239 brouard 373: Revision 1.238 2016/08/26 14:23:35 brouard
374: Summary: Starting tests of 0.99
375:
1.238 brouard 376: Revision 1.237 2016/08/26 09:20:19 brouard
377: Summary: to valgrind
378:
1.237 brouard 379: Revision 1.236 2016/08/25 10:50:18 brouard
380: *** empty log message ***
381:
1.236 brouard 382: Revision 1.235 2016/08/25 06:59:23 brouard
383: *** empty log message ***
384:
1.235 brouard 385: Revision 1.234 2016/08/23 16:51:20 brouard
386: *** empty log message ***
387:
1.234 brouard 388: Revision 1.233 2016/08/23 07:40:50 brouard
389: Summary: not working
390:
1.233 brouard 391: Revision 1.232 2016/08/22 14:20:21 brouard
392: Summary: not working
393:
1.232 brouard 394: Revision 1.231 2016/08/22 07:17:15 brouard
395: Summary: not working
396:
1.231 brouard 397: Revision 1.230 2016/08/22 06:55:53 brouard
398: Summary: Not working
399:
1.230 brouard 400: Revision 1.229 2016/07/23 09:45:53 brouard
401: Summary: Completing for func too
402:
1.229 brouard 403: Revision 1.228 2016/07/22 17:45:30 brouard
404: Summary: Fixing some arrays, still debugging
405:
1.227 brouard 406: Revision 1.226 2016/07/12 18:42:34 brouard
407: Summary: temp
408:
1.226 brouard 409: Revision 1.225 2016/07/12 08:40:03 brouard
410: Summary: saving but not running
411:
1.225 brouard 412: Revision 1.224 2016/07/01 13:16:01 brouard
413: Summary: Fixes
414:
1.224 brouard 415: Revision 1.223 2016/02/19 09:23:35 brouard
416: Summary: temporary
417:
1.223 brouard 418: Revision 1.222 2016/02/17 08:14:50 brouard
419: Summary: Probably last 0.98 stable version 0.98r6
420:
1.222 brouard 421: Revision 1.221 2016/02/15 23:35:36 brouard
422: Summary: minor bug
423:
1.220 brouard 424: Revision 1.219 2016/02/15 00:48:12 brouard
425: *** empty log message ***
426:
1.219 brouard 427: Revision 1.218 2016/02/12 11:29:23 brouard
428: Summary: 0.99 Back projections
429:
1.218 brouard 430: Revision 1.217 2015/12/23 17:18:31 brouard
431: Summary: Experimental backcast
432:
1.217 brouard 433: Revision 1.216 2015/12/18 17:32:11 brouard
434: Summary: 0.98r4 Warning and status=-2
435:
436: Version 0.98r4 is now:
437: - displaying an error when status is -1, date of interview unknown and date of death known;
438: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
439: Older changes concerning s=-2, dating from 2005 have been supersed.
440:
1.216 brouard 441: Revision 1.215 2015/12/16 08:52:24 brouard
442: Summary: 0.98r4 working
443:
1.215 brouard 444: Revision 1.214 2015/12/16 06:57:54 brouard
445: Summary: temporary not working
446:
1.214 brouard 447: Revision 1.213 2015/12/11 18:22:17 brouard
448: Summary: 0.98r4
449:
1.213 brouard 450: Revision 1.212 2015/11/21 12:47:24 brouard
451: Summary: minor typo
452:
1.212 brouard 453: Revision 1.211 2015/11/21 12:41:11 brouard
454: Summary: 0.98r3 with some graph of projected cross-sectional
455:
456: Author: Nicolas Brouard
457:
1.211 brouard 458: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 459: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 460: Summary: Adding ftolpl parameter
461: Author: N Brouard
462:
463: We had difficulties to get smoothed confidence intervals. It was due
464: to the period prevalence which wasn't computed accurately. The inner
465: parameter ftolpl is now an outer parameter of the .imach parameter
466: file after estepm. If ftolpl is small 1.e-4 and estepm too,
467: computation are long.
468:
1.209 brouard 469: Revision 1.208 2015/11/17 14:31:57 brouard
470: Summary: temporary
471:
1.208 brouard 472: Revision 1.207 2015/10/27 17:36:57 brouard
473: *** empty log message ***
474:
1.207 brouard 475: Revision 1.206 2015/10/24 07:14:11 brouard
476: *** empty log message ***
477:
1.206 brouard 478: Revision 1.205 2015/10/23 15:50:53 brouard
479: Summary: 0.98r3 some clarification for graphs on likelihood contributions
480:
1.205 brouard 481: Revision 1.204 2015/10/01 16:20:26 brouard
482: Summary: Some new graphs of contribution to likelihood
483:
1.204 brouard 484: Revision 1.203 2015/09/30 17:45:14 brouard
485: Summary: looking at better estimation of the hessian
486:
487: Also a better criteria for convergence to the period prevalence And
488: therefore adding the number of years needed to converge. (The
489: prevalence in any alive state shold sum to one
490:
1.203 brouard 491: Revision 1.202 2015/09/22 19:45:16 brouard
492: Summary: Adding some overall graph on contribution to likelihood. Might change
493:
1.202 brouard 494: Revision 1.201 2015/09/15 17:34:58 brouard
495: Summary: 0.98r0
496:
497: - Some new graphs like suvival functions
498: - Some bugs fixed like model=1+age+V2.
499:
1.201 brouard 500: Revision 1.200 2015/09/09 16:53:55 brouard
501: Summary: Big bug thanks to Flavia
502:
503: Even model=1+age+V2. did not work anymore
504:
1.200 brouard 505: Revision 1.199 2015/09/07 14:09:23 brouard
506: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
507:
1.199 brouard 508: Revision 1.198 2015/09/03 07:14:39 brouard
509: Summary: 0.98q5 Flavia
510:
1.198 brouard 511: Revision 1.197 2015/09/01 18:24:39 brouard
512: *** empty log message ***
513:
1.197 brouard 514: Revision 1.196 2015/08/18 23:17:52 brouard
515: Summary: 0.98q5
516:
1.196 brouard 517: Revision 1.195 2015/08/18 16:28:39 brouard
518: Summary: Adding a hack for testing purpose
519:
520: After reading the title, ftol and model lines, if the comment line has
521: a q, starting with #q, the answer at the end of the run is quit. It
522: permits to run test files in batch with ctest. The former workaround was
523: $ echo q | imach foo.imach
524:
1.195 brouard 525: Revision 1.194 2015/08/18 13:32:00 brouard
526: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
527:
1.194 brouard 528: Revision 1.193 2015/08/04 07:17:42 brouard
529: Summary: 0.98q4
530:
1.193 brouard 531: Revision 1.192 2015/07/16 16:49:02 brouard
532: Summary: Fixing some outputs
533:
1.192 brouard 534: Revision 1.191 2015/07/14 10:00:33 brouard
535: Summary: Some fixes
536:
1.191 brouard 537: Revision 1.190 2015/05/05 08:51:13 brouard
538: Summary: Adding digits in output parameters (7 digits instead of 6)
539:
540: Fix 1+age+.
541:
1.190 brouard 542: Revision 1.189 2015/04/30 14:45:16 brouard
543: Summary: 0.98q2
544:
1.189 brouard 545: Revision 1.188 2015/04/30 08:27:53 brouard
546: *** empty log message ***
547:
1.188 brouard 548: Revision 1.187 2015/04/29 09:11:15 brouard
549: *** empty log message ***
550:
1.187 brouard 551: Revision 1.186 2015/04/23 12:01:52 brouard
552: Summary: V1*age is working now, version 0.98q1
553:
554: Some codes had been disabled in order to simplify and Vn*age was
555: working in the optimization phase, ie, giving correct MLE parameters,
556: but, as usual, outputs were not correct and program core dumped.
557:
1.186 brouard 558: Revision 1.185 2015/03/11 13:26:42 brouard
559: Summary: Inclusion of compile and links command line for Intel Compiler
560:
1.185 brouard 561: Revision 1.184 2015/03/11 11:52:39 brouard
562: Summary: Back from Windows 8. Intel Compiler
563:
1.184 brouard 564: Revision 1.183 2015/03/10 20:34:32 brouard
565: Summary: 0.98q0, trying with directest, mnbrak fixed
566:
567: We use directest instead of original Powell test; probably no
568: incidence on the results, but better justifications;
569: We fixed Numerical Recipes mnbrak routine which was wrong and gave
570: wrong results.
571:
1.183 brouard 572: Revision 1.182 2015/02/12 08:19:57 brouard
573: Summary: Trying to keep directest which seems simpler and more general
574: Author: Nicolas Brouard
575:
1.182 brouard 576: Revision 1.181 2015/02/11 23:22:24 brouard
577: Summary: Comments on Powell added
578:
579: Author:
580:
1.181 brouard 581: Revision 1.180 2015/02/11 17:33:45 brouard
582: Summary: Finishing move from main to function (hpijx and prevalence_limit)
583:
1.180 brouard 584: Revision 1.179 2015/01/04 09:57:06 brouard
585: Summary: back to OS/X
586:
1.179 brouard 587: Revision 1.178 2015/01/04 09:35:48 brouard
588: *** empty log message ***
589:
1.178 brouard 590: Revision 1.177 2015/01/03 18:40:56 brouard
591: Summary: Still testing ilc32 on OSX
592:
1.177 brouard 593: Revision 1.176 2015/01/03 16:45:04 brouard
594: *** empty log message ***
595:
1.176 brouard 596: Revision 1.175 2015/01/03 16:33:42 brouard
597: *** empty log message ***
598:
1.175 brouard 599: Revision 1.174 2015/01/03 16:15:49 brouard
600: Summary: Still in cross-compilation
601:
1.174 brouard 602: Revision 1.173 2015/01/03 12:06:26 brouard
603: Summary: trying to detect cross-compilation
604:
1.173 brouard 605: Revision 1.172 2014/12/27 12:07:47 brouard
606: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
607:
1.172 brouard 608: Revision 1.171 2014/12/23 13:26:59 brouard
609: Summary: Back from Visual C
610:
611: Still problem with utsname.h on Windows
612:
1.171 brouard 613: Revision 1.170 2014/12/23 11:17:12 brouard
614: Summary: Cleaning some \%% back to %%
615:
616: The escape was mandatory for a specific compiler (which one?), but too many warnings.
617:
1.170 brouard 618: Revision 1.169 2014/12/22 23:08:31 brouard
619: Summary: 0.98p
620:
621: Outputs some informations on compiler used, OS etc. Testing on different platforms.
622:
1.169 brouard 623: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 624: Summary: update
1.169 brouard 625:
1.168 brouard 626: Revision 1.167 2014/12/22 13:50:56 brouard
627: Summary: Testing uname and compiler version and if compiled 32 or 64
628:
629: Testing on Linux 64
630:
1.167 brouard 631: Revision 1.166 2014/12/22 11:40:47 brouard
632: *** empty log message ***
633:
1.166 brouard 634: Revision 1.165 2014/12/16 11:20:36 brouard
635: Summary: After compiling on Visual C
636:
637: * imach.c (Module): Merging 1.61 to 1.162
638:
1.165 brouard 639: Revision 1.164 2014/12/16 10:52:11 brouard
640: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
641:
642: * imach.c (Module): Merging 1.61 to 1.162
643:
1.164 brouard 644: Revision 1.163 2014/12/16 10:30:11 brouard
645: * imach.c (Module): Merging 1.61 to 1.162
646:
1.163 brouard 647: Revision 1.162 2014/09/25 11:43:39 brouard
648: Summary: temporary backup 0.99!
649:
1.162 brouard 650: Revision 1.1 2014/09/16 11:06:58 brouard
651: Summary: With some code (wrong) for nlopt
652:
653: Author:
654:
655: Revision 1.161 2014/09/15 20:41:41 brouard
656: Summary: Problem with macro SQR on Intel compiler
657:
1.161 brouard 658: Revision 1.160 2014/09/02 09:24:05 brouard
659: *** empty log message ***
660:
1.160 brouard 661: Revision 1.159 2014/09/01 10:34:10 brouard
662: Summary: WIN32
663: Author: Brouard
664:
1.159 brouard 665: Revision 1.158 2014/08/27 17:11:51 brouard
666: *** empty log message ***
667:
1.158 brouard 668: Revision 1.157 2014/08/27 16:26:55 brouard
669: Summary: Preparing windows Visual studio version
670: Author: Brouard
671:
672: In order to compile on Visual studio, time.h is now correct and time_t
673: and tm struct should be used. difftime should be used but sometimes I
674: just make the differences in raw time format (time(&now).
675: Trying to suppress #ifdef LINUX
676: Add xdg-open for __linux in order to open default browser.
677:
1.157 brouard 678: Revision 1.156 2014/08/25 20:10:10 brouard
679: *** empty log message ***
680:
1.156 brouard 681: Revision 1.155 2014/08/25 18:32:34 brouard
682: Summary: New compile, minor changes
683: Author: Brouard
684:
1.155 brouard 685: Revision 1.154 2014/06/20 17:32:08 brouard
686: Summary: Outputs now all graphs of convergence to period prevalence
687:
1.154 brouard 688: Revision 1.153 2014/06/20 16:45:46 brouard
689: Summary: If 3 live state, convergence to period prevalence on same graph
690: Author: Brouard
691:
1.153 brouard 692: Revision 1.152 2014/06/18 17:54:09 brouard
693: Summary: open browser, use gnuplot on same dir than imach if not found in the path
694:
1.152 brouard 695: Revision 1.151 2014/06/18 16:43:30 brouard
696: *** empty log message ***
697:
1.151 brouard 698: Revision 1.150 2014/06/18 16:42:35 brouard
699: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
700: Author: brouard
701:
1.150 brouard 702: Revision 1.149 2014/06/18 15:51:14 brouard
703: Summary: Some fixes in parameter files errors
704: Author: Nicolas Brouard
705:
1.149 brouard 706: Revision 1.148 2014/06/17 17:38:48 brouard
707: Summary: Nothing new
708: Author: Brouard
709:
710: Just a new packaging for OS/X version 0.98nS
711:
1.148 brouard 712: Revision 1.147 2014/06/16 10:33:11 brouard
713: *** empty log message ***
714:
1.147 brouard 715: Revision 1.146 2014/06/16 10:20:28 brouard
716: Summary: Merge
717: Author: Brouard
718:
719: Merge, before building revised version.
720:
1.146 brouard 721: Revision 1.145 2014/06/10 21:23:15 brouard
722: Summary: Debugging with valgrind
723: Author: Nicolas Brouard
724:
725: Lot of changes in order to output the results with some covariates
726: After the Edimburgh REVES conference 2014, it seems mandatory to
727: improve the code.
728: No more memory valgrind error but a lot has to be done in order to
729: continue the work of splitting the code into subroutines.
730: Also, decodemodel has been improved. Tricode is still not
731: optimal. nbcode should be improved. Documentation has been added in
732: the source code.
733:
1.144 brouard 734: Revision 1.143 2014/01/26 09:45:38 brouard
735: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
736:
737: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
738: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
739:
1.143 brouard 740: Revision 1.142 2014/01/26 03:57:36 brouard
741: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
742:
743: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
744:
1.142 brouard 745: Revision 1.141 2014/01/26 02:42:01 brouard
746: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
747:
1.141 brouard 748: Revision 1.140 2011/09/02 10:37:54 brouard
749: Summary: times.h is ok with mingw32 now.
750:
1.140 brouard 751: Revision 1.139 2010/06/14 07:50:17 brouard
752: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
753: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
754:
1.139 brouard 755: Revision 1.138 2010/04/30 18:19:40 brouard
756: *** empty log message ***
757:
1.138 brouard 758: Revision 1.137 2010/04/29 18:11:38 brouard
759: (Module): Checking covariates for more complex models
760: than V1+V2. A lot of change to be done. Unstable.
761:
1.137 brouard 762: Revision 1.136 2010/04/26 20:30:53 brouard
763: (Module): merging some libgsl code. Fixing computation
764: of likelione (using inter/intrapolation if mle = 0) in order to
765: get same likelihood as if mle=1.
766: Some cleaning of code and comments added.
767:
1.136 brouard 768: Revision 1.135 2009/10/29 15:33:14 brouard
769: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
770:
1.135 brouard 771: Revision 1.134 2009/10/29 13:18:53 brouard
772: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
773:
1.134 brouard 774: Revision 1.133 2009/07/06 10:21:25 brouard
775: just nforces
776:
1.133 brouard 777: Revision 1.132 2009/07/06 08:22:05 brouard
778: Many tings
779:
1.132 brouard 780: Revision 1.131 2009/06/20 16:22:47 brouard
781: Some dimensions resccaled
782:
1.131 brouard 783: Revision 1.130 2009/05/26 06:44:34 brouard
784: (Module): Max Covariate is now set to 20 instead of 8. A
785: lot of cleaning with variables initialized to 0. Trying to make
786: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
787:
1.130 brouard 788: Revision 1.129 2007/08/31 13:49:27 lievre
789: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
790:
1.129 lievre 791: Revision 1.128 2006/06/30 13:02:05 brouard
792: (Module): Clarifications on computing e.j
793:
1.128 brouard 794: Revision 1.127 2006/04/28 18:11:50 brouard
795: (Module): Yes the sum of survivors was wrong since
796: imach-114 because nhstepm was no more computed in the age
797: loop. Now we define nhstepma in the age loop.
798: (Module): In order to speed up (in case of numerous covariates) we
799: compute health expectancies (without variances) in a first step
800: and then all the health expectancies with variances or standard
801: deviation (needs data from the Hessian matrices) which slows the
802: computation.
803: In the future we should be able to stop the program is only health
804: expectancies and graph are needed without standard deviations.
805:
1.127 brouard 806: Revision 1.126 2006/04/28 17:23:28 brouard
807: (Module): Yes the sum of survivors was wrong since
808: imach-114 because nhstepm was no more computed in the age
809: loop. Now we define nhstepma in the age loop.
810: Version 0.98h
811:
1.126 brouard 812: Revision 1.125 2006/04/04 15:20:31 lievre
813: Errors in calculation of health expectancies. Age was not initialized.
814: Forecasting file added.
815:
816: Revision 1.124 2006/03/22 17:13:53 lievre
817: Parameters are printed with %lf instead of %f (more numbers after the comma).
818: The log-likelihood is printed in the log file
819:
820: Revision 1.123 2006/03/20 10:52:43 brouard
821: * imach.c (Module): <title> changed, corresponds to .htm file
822: name. <head> headers where missing.
823:
824: * imach.c (Module): Weights can have a decimal point as for
825: English (a comma might work with a correct LC_NUMERIC environment,
826: otherwise the weight is truncated).
827: Modification of warning when the covariates values are not 0 or
828: 1.
829: Version 0.98g
830:
831: Revision 1.122 2006/03/20 09:45:41 brouard
832: (Module): Weights can have a decimal point as for
833: English (a comma might work with a correct LC_NUMERIC environment,
834: otherwise the weight is truncated).
835: Modification of warning when the covariates values are not 0 or
836: 1.
837: Version 0.98g
838:
839: Revision 1.121 2006/03/16 17:45:01 lievre
840: * imach.c (Module): Comments concerning covariates added
841:
842: * imach.c (Module): refinements in the computation of lli if
843: status=-2 in order to have more reliable computation if stepm is
844: not 1 month. Version 0.98f
845:
846: Revision 1.120 2006/03/16 15:10:38 lievre
847: (Module): refinements in the computation of lli if
848: status=-2 in order to have more reliable computation if stepm is
849: not 1 month. Version 0.98f
850:
851: Revision 1.119 2006/03/15 17:42:26 brouard
852: (Module): Bug if status = -2, the loglikelihood was
853: computed as likelihood omitting the logarithm. Version O.98e
854:
855: Revision 1.118 2006/03/14 18:20:07 brouard
856: (Module): varevsij Comments added explaining the second
857: table of variances if popbased=1 .
858: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
859: (Module): Function pstamp added
860: (Module): Version 0.98d
861:
862: Revision 1.117 2006/03/14 17:16:22 brouard
863: (Module): varevsij Comments added explaining the second
864: table of variances if popbased=1 .
865: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
866: (Module): Function pstamp added
867: (Module): Version 0.98d
868:
869: Revision 1.116 2006/03/06 10:29:27 brouard
870: (Module): Variance-covariance wrong links and
871: varian-covariance of ej. is needed (Saito).
872:
873: Revision 1.115 2006/02/27 12:17:45 brouard
874: (Module): One freematrix added in mlikeli! 0.98c
875:
876: Revision 1.114 2006/02/26 12:57:58 brouard
877: (Module): Some improvements in processing parameter
878: filename with strsep.
879:
880: Revision 1.113 2006/02/24 14:20:24 brouard
881: (Module): Memory leaks checks with valgrind and:
882: datafile was not closed, some imatrix were not freed and on matrix
883: allocation too.
884:
885: Revision 1.112 2006/01/30 09:55:26 brouard
886: (Module): Back to gnuplot.exe instead of wgnuplot.exe
887:
888: Revision 1.111 2006/01/25 20:38:18 brouard
889: (Module): Lots of cleaning and bugs added (Gompertz)
890: (Module): Comments can be added in data file. Missing date values
891: can be a simple dot '.'.
892:
893: Revision 1.110 2006/01/25 00:51:50 brouard
894: (Module): Lots of cleaning and bugs added (Gompertz)
895:
896: Revision 1.109 2006/01/24 19:37:15 brouard
897: (Module): Comments (lines starting with a #) are allowed in data.
898:
899: Revision 1.108 2006/01/19 18:05:42 lievre
900: Gnuplot problem appeared...
901: To be fixed
902:
903: Revision 1.107 2006/01/19 16:20:37 brouard
904: Test existence of gnuplot in imach path
905:
906: Revision 1.106 2006/01/19 13:24:36 brouard
907: Some cleaning and links added in html output
908:
909: Revision 1.105 2006/01/05 20:23:19 lievre
910: *** empty log message ***
911:
912: Revision 1.104 2005/09/30 16:11:43 lievre
913: (Module): sump fixed, loop imx fixed, and simplifications.
914: (Module): If the status is missing at the last wave but we know
915: that the person is alive, then we can code his/her status as -2
916: (instead of missing=-1 in earlier versions) and his/her
917: contributions to the likelihood is 1 - Prob of dying from last
918: health status (= 1-p13= p11+p12 in the easiest case of somebody in
919: the healthy state at last known wave). Version is 0.98
920:
921: Revision 1.103 2005/09/30 15:54:49 lievre
922: (Module): sump fixed, loop imx fixed, and simplifications.
923:
924: Revision 1.102 2004/09/15 17:31:30 brouard
925: Add the possibility to read data file including tab characters.
926:
927: Revision 1.101 2004/09/15 10:38:38 brouard
928: Fix on curr_time
929:
930: Revision 1.100 2004/07/12 18:29:06 brouard
931: Add version for Mac OS X. Just define UNIX in Makefile
932:
933: Revision 1.99 2004/06/05 08:57:40 brouard
934: *** empty log message ***
935:
936: Revision 1.98 2004/05/16 15:05:56 brouard
937: New version 0.97 . First attempt to estimate force of mortality
938: directly from the data i.e. without the need of knowing the health
939: state at each age, but using a Gompertz model: log u =a + b*age .
940: This is the basic analysis of mortality and should be done before any
941: other analysis, in order to test if the mortality estimated from the
942: cross-longitudinal survey is different from the mortality estimated
943: from other sources like vital statistic data.
944:
945: The same imach parameter file can be used but the option for mle should be -3.
946:
1.324 brouard 947: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 948: former routines in order to include the new code within the former code.
949:
950: The output is very simple: only an estimate of the intercept and of
951: the slope with 95% confident intervals.
952:
953: Current limitations:
954: A) Even if you enter covariates, i.e. with the
955: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
956: B) There is no computation of Life Expectancy nor Life Table.
957:
958: Revision 1.97 2004/02/20 13:25:42 lievre
959: Version 0.96d. Population forecasting command line is (temporarily)
960: suppressed.
961:
962: Revision 1.96 2003/07/15 15:38:55 brouard
963: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
964: rewritten within the same printf. Workaround: many printfs.
965:
966: Revision 1.95 2003/07/08 07:54:34 brouard
967: * imach.c (Repository):
968: (Repository): Using imachwizard code to output a more meaningful covariance
969: matrix (cov(a12,c31) instead of numbers.
970:
971: Revision 1.94 2003/06/27 13:00:02 brouard
972: Just cleaning
973:
974: Revision 1.93 2003/06/25 16:33:55 brouard
975: (Module): On windows (cygwin) function asctime_r doesn't
976: exist so I changed back to asctime which exists.
977: (Module): Version 0.96b
978:
979: Revision 1.92 2003/06/25 16:30:45 brouard
980: (Module): On windows (cygwin) function asctime_r doesn't
981: exist so I changed back to asctime which exists.
982:
983: Revision 1.91 2003/06/25 15:30:29 brouard
984: * imach.c (Repository): Duplicated warning errors corrected.
985: (Repository): Elapsed time after each iteration is now output. It
986: helps to forecast when convergence will be reached. Elapsed time
987: is stamped in powell. We created a new html file for the graphs
988: concerning matrix of covariance. It has extension -cov.htm.
989:
990: Revision 1.90 2003/06/24 12:34:15 brouard
991: (Module): Some bugs corrected for windows. Also, when
992: mle=-1 a template is output in file "or"mypar.txt with the design
993: of the covariance matrix to be input.
994:
995: Revision 1.89 2003/06/24 12:30:52 brouard
996: (Module): Some bugs corrected for windows. Also, when
997: mle=-1 a template is output in file "or"mypar.txt with the design
998: of the covariance matrix to be input.
999:
1000: Revision 1.88 2003/06/23 17:54:56 brouard
1001: * 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.
1002:
1003: Revision 1.87 2003/06/18 12:26:01 brouard
1004: Version 0.96
1005:
1006: Revision 1.86 2003/06/17 20:04:08 brouard
1007: (Module): Change position of html and gnuplot routines and added
1008: routine fileappend.
1009:
1010: Revision 1.85 2003/06/17 13:12:43 brouard
1011: * imach.c (Repository): Check when date of death was earlier that
1012: current date of interview. It may happen when the death was just
1013: prior to the death. In this case, dh was negative and likelihood
1014: was wrong (infinity). We still send an "Error" but patch by
1015: assuming that the date of death was just one stepm after the
1016: interview.
1017: (Repository): Because some people have very long ID (first column)
1018: we changed int to long in num[] and we added a new lvector for
1019: memory allocation. But we also truncated to 8 characters (left
1020: truncation)
1021: (Repository): No more line truncation errors.
1022:
1023: Revision 1.84 2003/06/13 21:44:43 brouard
1024: * imach.c (Repository): Replace "freqsummary" at a correct
1025: place. It differs from routine "prevalence" which may be called
1026: many times. Probs is memory consuming and must be used with
1027: parcimony.
1028: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1029:
1030: Revision 1.83 2003/06/10 13:39:11 lievre
1031: *** empty log message ***
1032:
1033: Revision 1.82 2003/06/05 15:57:20 brouard
1034: Add log in imach.c and fullversion number is now printed.
1035:
1036: */
1037: /*
1038: Interpolated Markov Chain
1039:
1040: Short summary of the programme:
1041:
1.227 brouard 1042: This program computes Healthy Life Expectancies or State-specific
1043: (if states aren't health statuses) Expectancies from
1044: cross-longitudinal data. Cross-longitudinal data consist in:
1045:
1046: -1- a first survey ("cross") where individuals from different ages
1047: are interviewed on their health status or degree of disability (in
1048: the case of a health survey which is our main interest)
1049:
1050: -2- at least a second wave of interviews ("longitudinal") which
1051: measure each change (if any) in individual health status. Health
1052: expectancies are computed from the time spent in each health state
1053: according to a model. More health states you consider, more time is
1054: necessary to reach the Maximum Likelihood of the parameters involved
1055: in the model. The simplest model is the multinomial logistic model
1056: where pij is the probability to be observed in state j at the second
1057: wave conditional to be observed in state i at the first
1058: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1059: etc , where 'age' is age and 'sex' is a covariate. If you want to
1060: have a more complex model than "constant and age", you should modify
1061: the program where the markup *Covariates have to be included here
1062: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1063: convergence.
1064:
1065: The advantage of this computer programme, compared to a simple
1066: multinomial logistic model, is clear when the delay between waves is not
1067: identical for each individual. Also, if a individual missed an
1068: intermediate interview, the information is lost, but taken into
1069: account using an interpolation or extrapolation.
1070:
1071: hPijx is the probability to be observed in state i at age x+h
1072: conditional to the observed state i at age x. The delay 'h' can be
1073: split into an exact number (nh*stepm) of unobserved intermediate
1074: states. This elementary transition (by month, quarter,
1075: semester or year) is modelled as a multinomial logistic. The hPx
1076: matrix is simply the matrix product of nh*stepm elementary matrices
1077: and the contribution of each individual to the likelihood is simply
1078: hPijx.
1079:
1080: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1081: of the life expectancies. It also computes the period (stable) prevalence.
1082:
1083: Back prevalence and projections:
1.227 brouard 1084:
1085: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1086: double agemaxpar, double ftolpl, int *ncvyearp, double
1087: dateprev1,double dateprev2, int firstpass, int lastpass, int
1088: mobilavproj)
1089:
1090: Computes the back prevalence limit for any combination of
1091: covariate values k at any age between ageminpar and agemaxpar and
1092: returns it in **bprlim. In the loops,
1093:
1094: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1095: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1096:
1097: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1098: Computes for any combination of covariates k and any age between bage and fage
1099: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1100: oldm=oldms;savm=savms;
1.227 brouard 1101:
1.267 brouard 1102: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1103: Computes the transition matrix starting at age 'age' over
1104: 'nhstepm*hstepm*stepm' months (i.e. until
1105: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1106: nhstepm*hstepm matrices.
1107:
1108: Returns p3mat[i][j][h] after calling
1109: p3mat[i][j][h]=matprod2(newm,
1110: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1111: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1112: oldm);
1.226 brouard 1113:
1114: Important routines
1115:
1116: - func (or funcone), computes logit (pij) distinguishing
1117: o fixed variables (single or product dummies or quantitative);
1118: o varying variables by:
1119: (1) wave (single, product dummies, quantitative),
1120: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1121: % fixed dummy (treated) or quantitative (not done because time-consuming);
1122: % varying dummy (not done) or quantitative (not done);
1123: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1124: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1125: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1126: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1127: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1128:
1.226 brouard 1129:
1130:
1.324 brouard 1131: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1132: Institut national d'études démographiques, Paris.
1.126 brouard 1133: This software have been partly granted by Euro-REVES, a concerted action
1134: from the European Union.
1135: It is copyrighted identically to a GNU software product, ie programme and
1136: software can be distributed freely for non commercial use. Latest version
1137: can be accessed at http://euroreves.ined.fr/imach .
1138:
1139: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1140: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1141:
1142: **********************************************************************/
1143: /*
1144: main
1145: read parameterfile
1146: read datafile
1147: concatwav
1148: freqsummary
1149: if (mle >= 1)
1150: mlikeli
1151: print results files
1152: if mle==1
1153: computes hessian
1154: read end of parameter file: agemin, agemax, bage, fage, estepm
1155: begin-prev-date,...
1156: open gnuplot file
1157: open html file
1.145 brouard 1158: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1159: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1160: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1161: freexexit2 possible for memory heap.
1162:
1163: h Pij x | pij_nom ficrestpij
1164: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1165: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1166: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1167:
1168: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1169: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1170: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1171: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1172: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1173:
1.126 brouard 1174: forecasting if prevfcast==1 prevforecast call prevalence()
1175: health expectancies
1176: Variance-covariance of DFLE
1177: prevalence()
1178: movingaverage()
1179: varevsij()
1180: if popbased==1 varevsij(,popbased)
1181: total life expectancies
1182: Variance of period (stable) prevalence
1183: end
1184: */
1185:
1.187 brouard 1186: /* #define DEBUG */
1187: /* #define DEBUGBRENT */
1.203 brouard 1188: /* #define DEBUGLINMIN */
1189: /* #define DEBUGHESS */
1190: #define DEBUGHESSIJ
1.224 brouard 1191: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1192: #define POWELL /* Instead of NLOPT */
1.224 brouard 1193: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1194: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1195: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1196: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1197:
1198: #include <math.h>
1199: #include <stdio.h>
1200: #include <stdlib.h>
1201: #include <string.h>
1.226 brouard 1202: #include <ctype.h>
1.159 brouard 1203:
1204: #ifdef _WIN32
1205: #include <io.h>
1.172 brouard 1206: #include <windows.h>
1207: #include <tchar.h>
1.159 brouard 1208: #else
1.126 brouard 1209: #include <unistd.h>
1.159 brouard 1210: #endif
1.126 brouard 1211:
1212: #include <limits.h>
1213: #include <sys/types.h>
1.171 brouard 1214:
1215: #if defined(__GNUC__)
1216: #include <sys/utsname.h> /* Doesn't work on Windows */
1217: #endif
1218:
1.126 brouard 1219: #include <sys/stat.h>
1220: #include <errno.h>
1.159 brouard 1221: /* extern int errno; */
1.126 brouard 1222:
1.157 brouard 1223: /* #ifdef LINUX */
1224: /* #include <time.h> */
1225: /* #include "timeval.h" */
1226: /* #else */
1227: /* #include <sys/time.h> */
1228: /* #endif */
1229:
1.126 brouard 1230: #include <time.h>
1231:
1.136 brouard 1232: #ifdef GSL
1233: #include <gsl/gsl_errno.h>
1234: #include <gsl/gsl_multimin.h>
1235: #endif
1236:
1.167 brouard 1237:
1.162 brouard 1238: #ifdef NLOPT
1239: #include <nlopt.h>
1240: typedef struct {
1241: double (* function)(double [] );
1242: } myfunc_data ;
1243: #endif
1244:
1.126 brouard 1245: /* #include <libintl.h> */
1246: /* #define _(String) gettext (String) */
1247:
1.251 brouard 1248: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1249:
1250: #define GNUPLOTPROGRAM "gnuplot"
1251: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1252: #define FILENAMELENGTH 256
1.126 brouard 1253:
1254: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1255: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1256:
1.144 brouard 1257: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1258: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1259:
1260: #define NINTERVMAX 8
1.144 brouard 1261: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1262: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1263: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1264: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1265: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1266: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1267: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1268: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1269: /* #define AGESUP 130 */
1.288 brouard 1270: /* #define AGESUP 150 */
1271: #define AGESUP 200
1.268 brouard 1272: #define AGEINF 0
1.218 brouard 1273: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1274: #define AGEBASE 40
1.194 brouard 1275: #define AGEOVERFLOW 1.e20
1.164 brouard 1276: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1277: #ifdef _WIN32
1278: #define DIRSEPARATOR '\\'
1279: #define CHARSEPARATOR "\\"
1280: #define ODIRSEPARATOR '/'
1281: #else
1.126 brouard 1282: #define DIRSEPARATOR '/'
1283: #define CHARSEPARATOR "/"
1284: #define ODIRSEPARATOR '\\'
1285: #endif
1286:
1.336 ! brouard 1287: /* $Id: imach.c,v 1.335 2022/08/31 08:23:16 brouard Exp $ */
1.126 brouard 1288: /* $State: Exp $ */
1.196 brouard 1289: #include "version.h"
1290: char version[]=__IMACH_VERSION__;
1.332 brouard 1291: 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";
1.336 ! brouard 1292: char fullversion[]="$Revision: 1.335 $ $Date: 2022/08/31 08:23:16 $";
1.126 brouard 1293: char strstart[80];
1294: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1295: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1296: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1297: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1298: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 brouard 1299: 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 1300: 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 1301: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1302: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1303: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1304: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 brouard 1305: 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 1306: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1307: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1308: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1309: int nsd=0; /**< Total number of single dummy variables (output) */
1310: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1311: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1312: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1313: int ntveff=0; /**< ntveff number of effective time varying variables */
1314: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1315: int cptcov=0; /* Working variable */
1.334 brouard 1316: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1317: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1318: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1319: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1320: int nlstate=2; /* Number of live states */
1321: int ndeath=1; /* Number of dead states */
1.130 brouard 1322: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1323: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1324: int popbased=0;
1325:
1326: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1327: int maxwav=0; /* Maxim number of waves */
1328: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1329: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1330: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1331: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1332: int mle=1, weightopt=0;
1.126 brouard 1333: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1334: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1335: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1336: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1337: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1338: int selected(int kvar); /* Is covariate kvar selected for printing results */
1339:
1.130 brouard 1340: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1341: double **matprod2(); /* test */
1.126 brouard 1342: double **oldm, **newm, **savm; /* Working pointers to matrices */
1343: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1344: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1345:
1.136 brouard 1346: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1347: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1348: FILE *ficlog, *ficrespow;
1.130 brouard 1349: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1350: double fretone; /* Only one call to likelihood */
1.130 brouard 1351: long ipmx=0; /* Number of contributions */
1.126 brouard 1352: double sw; /* Sum of weights */
1353: char filerespow[FILENAMELENGTH];
1354: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1355: FILE *ficresilk;
1356: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1357: FILE *ficresprobmorprev;
1358: FILE *fichtm, *fichtmcov; /* Html File */
1359: FILE *ficreseij;
1360: char filerese[FILENAMELENGTH];
1361: FILE *ficresstdeij;
1362: char fileresstde[FILENAMELENGTH];
1363: FILE *ficrescveij;
1364: char filerescve[FILENAMELENGTH];
1365: FILE *ficresvij;
1366: char fileresv[FILENAMELENGTH];
1.269 brouard 1367:
1.126 brouard 1368: char title[MAXLINE];
1.234 brouard 1369: char model[MAXLINE]; /**< The model line */
1.217 brouard 1370: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1371: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1372: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1373: char command[FILENAMELENGTH];
1374: int outcmd=0;
1375:
1.217 brouard 1376: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1377: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1378: char filelog[FILENAMELENGTH]; /* Log file */
1379: char filerest[FILENAMELENGTH];
1380: char fileregp[FILENAMELENGTH];
1381: char popfile[FILENAMELENGTH];
1382:
1383: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1384:
1.157 brouard 1385: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1386: /* struct timezone tzp; */
1387: /* extern int gettimeofday(); */
1388: struct tm tml, *gmtime(), *localtime();
1389:
1390: extern time_t time();
1391:
1392: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1393: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1394: struct tm tm;
1395:
1.126 brouard 1396: char strcurr[80], strfor[80];
1397:
1398: char *endptr;
1399: long lval;
1400: double dval;
1401:
1402: #define NR_END 1
1403: #define FREE_ARG char*
1404: #define FTOL 1.0e-10
1405:
1406: #define NRANSI
1.240 brouard 1407: #define ITMAX 200
1408: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1409:
1410: #define TOL 2.0e-4
1411:
1412: #define CGOLD 0.3819660
1413: #define ZEPS 1.0e-10
1414: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1415:
1416: #define GOLD 1.618034
1417: #define GLIMIT 100.0
1418: #define TINY 1.0e-20
1419:
1420: static double maxarg1,maxarg2;
1421: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1422: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1423:
1424: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1425: #define rint(a) floor(a+0.5)
1.166 brouard 1426: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1427: #define mytinydouble 1.0e-16
1.166 brouard 1428: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1429: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1430: /* static double dsqrarg; */
1431: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1432: static double sqrarg;
1433: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1434: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1435: int agegomp= AGEGOMP;
1436:
1437: int imx;
1438: int stepm=1;
1439: /* Stepm, step in month: minimum step interpolation*/
1440:
1441: int estepm;
1442: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1443:
1444: int m,nb;
1445: long *num;
1.197 brouard 1446: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1447: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1448: covariate for which somebody answered excluding
1449: undefined. Usually 2: 0 and 1. */
1450: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1451: covariate for which somebody answered including
1452: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1453: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1454: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1455: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1456: 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 1457: double *ageexmed,*agecens;
1458: double dateintmean=0;
1.296 brouard 1459: double anprojd, mprojd, jprojd; /* For eventual projections */
1460: double anprojf, mprojf, jprojf;
1.126 brouard 1461:
1.296 brouard 1462: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1463: double anbackf, mbackf, jbackf;
1464: double jintmean,mintmean,aintmean;
1.126 brouard 1465: double *weight;
1466: int **s; /* Status */
1.141 brouard 1467: double *agedc;
1.145 brouard 1468: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1469: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1470: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1471: double **coqvar; /* Fixed quantitative covariate nqv */
1472: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1473: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1474: double idx;
1475: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1476: /* Some documentation */
1477: /* Design original data
1478: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1479: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1480: * ntv=3 nqtv=1
1.330 brouard 1481: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1482: * For time varying covariate, quanti or dummies
1483: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1484: * cotvar[wav][ntv+iv][i]= [3+(1 to nqtv)][i]=(V12) quanti
1485: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1486: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1487: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1488: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1489: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1490: * k= 1 2 3 4 5 6 7 8 9 10 11
1491: */
1492: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1493: /* 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
1494: # States 1=Coresidence, 2 Living alone, 3 Institution
1495: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1496: */
1.319 brouard 1497: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1498: /* k 1 2 3 4 5 6 7 8 9 */
1499: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1500: /* fixed or varying), 1 for age product, 2 for*/
1501: /* product */
1502: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1503: /*(single or product without age), 2 dummy*/
1504: /* with age product, 3 quant with age product*/
1505: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1506: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 brouard 1507: /*TnsdVar[Tvar] 1 2 3 */
1.319 brouard 1508: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1509: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1510: /* nsq 1 2 */ /* Counting single quantit tv */
1511: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1512: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1513: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1514: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1515: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1516: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1517: /* 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 1518: /* 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 1519: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1520: /* Type */
1521: /* V 1 2 3 4 5 */
1522: /* F F V V V */
1523: /* D Q D D Q */
1524: /* */
1525: int *TvarsD;
1.330 brouard 1526: int *TnsdVar;
1.234 brouard 1527: int *TvarsDind;
1528: int *TvarsQ;
1529: int *TvarsQind;
1530:
1.318 brouard 1531: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1532: int nresult=0;
1.258 brouard 1533: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1534: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1535: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1536: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1537: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1538: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1539: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1540: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1541: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1542: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1543: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1544:
1545: /* 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
1546: # States 1=Coresidence, 2 Living alone, 3 Institution
1547: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1548: */
1.234 brouard 1549: /* 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 1550: 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 */
1551: 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 */
1552: 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 */
1553: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1554: 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 */
1555: 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 1556: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1557: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1558: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1559: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1560: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1561: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1562: 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 */
1563: 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 */
1564:
1.230 brouard 1565: int *Tvarsel; /**< Selected covariates for output */
1566: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1567: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1568: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1569: 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 1570: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1571: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1572: int *Tage;
1.227 brouard 1573: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1574: 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 1575: 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*/
1576: 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 1577: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1578: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1579: int **Tvard;
1.330 brouard 1580: int **Tvardk;
1.227 brouard 1581: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1582: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1583: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1584: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1585: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1586: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1587: double *lsurv, *lpop, *tpop;
1588:
1.231 brouard 1589: #define FD 1; /* Fixed dummy covariate */
1590: #define FQ 2; /* Fixed quantitative covariate */
1591: #define FP 3; /* Fixed product covariate */
1592: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1593: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1594: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1595: #define VD 10; /* Varying dummy covariate */
1596: #define VQ 11; /* Varying quantitative covariate */
1597: #define VP 12; /* Varying product covariate */
1598: #define VPDD 13; /* Varying product dummy*dummy covariate */
1599: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1600: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1601: #define APFD 16; /* Age product * fixed dummy covariate */
1602: #define APFQ 17; /* Age product * fixed quantitative covariate */
1603: #define APVD 18; /* Age product * varying dummy covariate */
1604: #define APVQ 19; /* Age product * varying quantitative covariate */
1605:
1606: #define FTYPE 1; /* Fixed covariate */
1607: #define VTYPE 2; /* Varying covariate (loop in wave) */
1608: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1609:
1610: struct kmodel{
1611: int maintype; /* main type */
1612: int subtype; /* subtype */
1613: };
1614: struct kmodel modell[NCOVMAX];
1615:
1.143 brouard 1616: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1617: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1618:
1619: /**************** split *************************/
1620: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1621: {
1622: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1623: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1624: */
1625: char *ss; /* pointer */
1.186 brouard 1626: int l1=0, l2=0; /* length counters */
1.126 brouard 1627:
1628: l1 = strlen(path ); /* length of path */
1629: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1630: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1631: if ( ss == NULL ) { /* no directory, so determine current directory */
1632: strcpy( name, path ); /* we got the fullname name because no directory */
1633: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1634: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1635: /* get current working directory */
1636: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1637: #ifdef WIN32
1638: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1639: #else
1640: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1641: #endif
1.126 brouard 1642: return( GLOCK_ERROR_GETCWD );
1643: }
1644: /* got dirc from getcwd*/
1645: printf(" DIRC = %s \n",dirc);
1.205 brouard 1646: } else { /* strip directory from path */
1.126 brouard 1647: ss++; /* after this, the filename */
1648: l2 = strlen( ss ); /* length of filename */
1649: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1650: strcpy( name, ss ); /* save file name */
1651: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1652: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1653: printf(" DIRC2 = %s \n",dirc);
1654: }
1655: /* We add a separator at the end of dirc if not exists */
1656: l1 = strlen( dirc ); /* length of directory */
1657: if( dirc[l1-1] != DIRSEPARATOR ){
1658: dirc[l1] = DIRSEPARATOR;
1659: dirc[l1+1] = 0;
1660: printf(" DIRC3 = %s \n",dirc);
1661: }
1662: ss = strrchr( name, '.' ); /* find last / */
1663: if (ss >0){
1664: ss++;
1665: strcpy(ext,ss); /* save extension */
1666: l1= strlen( name);
1667: l2= strlen(ss)+1;
1668: strncpy( finame, name, l1-l2);
1669: finame[l1-l2]= 0;
1670: }
1671:
1672: return( 0 ); /* we're done */
1673: }
1674:
1675:
1676: /******************************************/
1677:
1678: void replace_back_to_slash(char *s, char*t)
1679: {
1680: int i;
1681: int lg=0;
1682: i=0;
1683: lg=strlen(t);
1684: for(i=0; i<= lg; i++) {
1685: (s[i] = t[i]);
1686: if (t[i]== '\\') s[i]='/';
1687: }
1688: }
1689:
1.132 brouard 1690: char *trimbb(char *out, char *in)
1.137 brouard 1691: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1692: char *s;
1693: s=out;
1694: while (*in != '\0'){
1.137 brouard 1695: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1696: in++;
1697: }
1698: *out++ = *in++;
1699: }
1700: *out='\0';
1701: return s;
1702: }
1703:
1.187 brouard 1704: /* char *substrchaine(char *out, char *in, char *chain) */
1705: /* { */
1706: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1707: /* char *s, *t; */
1708: /* t=in;s=out; */
1709: /* while ((*in != *chain) && (*in != '\0')){ */
1710: /* *out++ = *in++; */
1711: /* } */
1712:
1713: /* /\* *in matches *chain *\/ */
1714: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1715: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1716: /* } */
1717: /* in--; chain--; */
1718: /* while ( (*in != '\0')){ */
1719: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1720: /* *out++ = *in++; */
1721: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1722: /* } */
1723: /* *out='\0'; */
1724: /* out=s; */
1725: /* return out; */
1726: /* } */
1727: char *substrchaine(char *out, char *in, char *chain)
1728: {
1729: /* Substract chain 'chain' from 'in', return and output 'out' */
1730: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1731:
1732: char *strloc;
1733:
1734: strcpy (out, in);
1735: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1736: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1737: if(strloc != NULL){
1738: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1739: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1740: /* strcpy (strloc, strloc +strlen(chain));*/
1741: }
1742: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1743: return out;
1744: }
1745:
1746:
1.145 brouard 1747: char *cutl(char *blocc, char *alocc, char *in, char occ)
1748: {
1.187 brouard 1749: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1750: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1751: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1752: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1753: */
1.160 brouard 1754: char *s, *t;
1.145 brouard 1755: t=in;s=in;
1756: while ((*in != occ) && (*in != '\0')){
1757: *alocc++ = *in++;
1758: }
1759: if( *in == occ){
1760: *(alocc)='\0';
1761: s=++in;
1762: }
1763:
1764: if (s == t) {/* occ not found */
1765: *(alocc-(in-s))='\0';
1766: in=s;
1767: }
1768: while ( *in != '\0'){
1769: *blocc++ = *in++;
1770: }
1771:
1772: *blocc='\0';
1773: return t;
1774: }
1.137 brouard 1775: char *cutv(char *blocc, char *alocc, char *in, char occ)
1776: {
1.187 brouard 1777: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1778: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1779: gives blocc="abcdef2ghi" and alocc="j".
1780: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1781: */
1782: char *s, *t;
1783: t=in;s=in;
1784: while (*in != '\0'){
1785: while( *in == occ){
1786: *blocc++ = *in++;
1787: s=in;
1788: }
1789: *blocc++ = *in++;
1790: }
1791: if (s == t) /* occ not found */
1792: *(blocc-(in-s))='\0';
1793: else
1794: *(blocc-(in-s)-1)='\0';
1795: in=s;
1796: while ( *in != '\0'){
1797: *alocc++ = *in++;
1798: }
1799:
1800: *alocc='\0';
1801: return s;
1802: }
1803:
1.126 brouard 1804: int nbocc(char *s, char occ)
1805: {
1806: int i,j=0;
1807: int lg=20;
1808: i=0;
1809: lg=strlen(s);
1810: for(i=0; i<= lg; i++) {
1.234 brouard 1811: if (s[i] == occ ) j++;
1.126 brouard 1812: }
1813: return j;
1814: }
1815:
1.137 brouard 1816: /* void cutv(char *u,char *v, char*t, char occ) */
1817: /* { */
1818: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1819: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1820: /* gives u="abcdef2ghi" and v="j" *\/ */
1821: /* int i,lg,j,p=0; */
1822: /* i=0; */
1823: /* lg=strlen(t); */
1824: /* for(j=0; j<=lg-1; j++) { */
1825: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1826: /* } */
1.126 brouard 1827:
1.137 brouard 1828: /* for(j=0; j<p; j++) { */
1829: /* (u[j] = t[j]); */
1830: /* } */
1831: /* u[p]='\0'; */
1.126 brouard 1832:
1.137 brouard 1833: /* for(j=0; j<= lg; j++) { */
1834: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1835: /* } */
1836: /* } */
1.126 brouard 1837:
1.160 brouard 1838: #ifdef _WIN32
1839: char * strsep(char **pp, const char *delim)
1840: {
1841: char *p, *q;
1842:
1843: if ((p = *pp) == NULL)
1844: return 0;
1845: if ((q = strpbrk (p, delim)) != NULL)
1846: {
1847: *pp = q + 1;
1848: *q = '\0';
1849: }
1850: else
1851: *pp = 0;
1852: return p;
1853: }
1854: #endif
1855:
1.126 brouard 1856: /********************** nrerror ********************/
1857:
1858: void nrerror(char error_text[])
1859: {
1860: fprintf(stderr,"ERREUR ...\n");
1861: fprintf(stderr,"%s\n",error_text);
1862: exit(EXIT_FAILURE);
1863: }
1864: /*********************** vector *******************/
1865: double *vector(int nl, int nh)
1866: {
1867: double *v;
1868: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1869: if (!v) nrerror("allocation failure in vector");
1870: return v-nl+NR_END;
1871: }
1872:
1873: /************************ free vector ******************/
1874: void free_vector(double*v, int nl, int nh)
1875: {
1876: free((FREE_ARG)(v+nl-NR_END));
1877: }
1878:
1879: /************************ivector *******************************/
1880: int *ivector(long nl,long nh)
1881: {
1882: int *v;
1883: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1884: if (!v) nrerror("allocation failure in ivector");
1885: return v-nl+NR_END;
1886: }
1887:
1888: /******************free ivector **************************/
1889: void free_ivector(int *v, long nl, long nh)
1890: {
1891: free((FREE_ARG)(v+nl-NR_END));
1892: }
1893:
1894: /************************lvector *******************************/
1895: long *lvector(long nl,long nh)
1896: {
1897: long *v;
1898: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1899: if (!v) nrerror("allocation failure in ivector");
1900: return v-nl+NR_END;
1901: }
1902:
1903: /******************free lvector **************************/
1904: void free_lvector(long *v, long nl, long nh)
1905: {
1906: free((FREE_ARG)(v+nl-NR_END));
1907: }
1908:
1909: /******************* imatrix *******************************/
1910: int **imatrix(long nrl, long nrh, long ncl, long nch)
1911: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1912: {
1913: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1914: int **m;
1915:
1916: /* allocate pointers to rows */
1917: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1918: if (!m) nrerror("allocation failure 1 in matrix()");
1919: m += NR_END;
1920: m -= nrl;
1921:
1922:
1923: /* allocate rows and set pointers to them */
1924: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1925: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1926: m[nrl] += NR_END;
1927: m[nrl] -= ncl;
1928:
1929: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1930:
1931: /* return pointer to array of pointers to rows */
1932: return m;
1933: }
1934:
1935: /****************** free_imatrix *************************/
1936: void free_imatrix(m,nrl,nrh,ncl,nch)
1937: int **m;
1938: long nch,ncl,nrh,nrl;
1939: /* free an int matrix allocated by imatrix() */
1940: {
1941: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1942: free((FREE_ARG) (m+nrl-NR_END));
1943: }
1944:
1945: /******************* matrix *******************************/
1946: double **matrix(long nrl, long nrh, long ncl, long nch)
1947: {
1948: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1949: double **m;
1950:
1951: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1952: if (!m) nrerror("allocation failure 1 in matrix()");
1953: m += NR_END;
1954: m -= nrl;
1955:
1956: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1957: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1958: m[nrl] += NR_END;
1959: m[nrl] -= ncl;
1960:
1961: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1962: return m;
1.145 brouard 1963: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1964: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1965: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1966: */
1967: }
1968:
1969: /*************************free matrix ************************/
1970: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1971: {
1972: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1973: free((FREE_ARG)(m+nrl-NR_END));
1974: }
1975:
1976: /******************* ma3x *******************************/
1977: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1978: {
1979: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1980: double ***m;
1981:
1982: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1983: if (!m) nrerror("allocation failure 1 in matrix()");
1984: m += NR_END;
1985: m -= nrl;
1986:
1987: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1988: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1989: m[nrl] += NR_END;
1990: m[nrl] -= ncl;
1991:
1992: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1993:
1994: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1995: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1996: m[nrl][ncl] += NR_END;
1997: m[nrl][ncl] -= nll;
1998: for (j=ncl+1; j<=nch; j++)
1999: m[nrl][j]=m[nrl][j-1]+nlay;
2000:
2001: for (i=nrl+1; i<=nrh; i++) {
2002: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2003: for (j=ncl+1; j<=nch; j++)
2004: m[i][j]=m[i][j-1]+nlay;
2005: }
2006: return m;
2007: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2008: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2009: */
2010: }
2011:
2012: /*************************free ma3x ************************/
2013: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2014: {
2015: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2016: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2017: free((FREE_ARG)(m+nrl-NR_END));
2018: }
2019:
2020: /*************** function subdirf ***********/
2021: char *subdirf(char fileres[])
2022: {
2023: /* Caution optionfilefiname is hidden */
2024: strcpy(tmpout,optionfilefiname);
2025: strcat(tmpout,"/"); /* Add to the right */
2026: strcat(tmpout,fileres);
2027: return tmpout;
2028: }
2029:
2030: /*************** function subdirf2 ***********/
2031: char *subdirf2(char fileres[], char *preop)
2032: {
1.314 brouard 2033: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2034: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2035: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2036: /* Caution optionfilefiname is hidden */
2037: strcpy(tmpout,optionfilefiname);
2038: strcat(tmpout,"/");
2039: strcat(tmpout,preop);
2040: strcat(tmpout,fileres);
2041: return tmpout;
2042: }
2043:
2044: /*************** function subdirf3 ***********/
2045: char *subdirf3(char fileres[], char *preop, char *preop2)
2046: {
2047:
2048: /* Caution optionfilefiname is hidden */
2049: strcpy(tmpout,optionfilefiname);
2050: strcat(tmpout,"/");
2051: strcat(tmpout,preop);
2052: strcat(tmpout,preop2);
2053: strcat(tmpout,fileres);
2054: return tmpout;
2055: }
1.213 brouard 2056:
2057: /*************** function subdirfext ***********/
2058: char *subdirfext(char fileres[], char *preop, char *postop)
2059: {
2060:
2061: strcpy(tmpout,preop);
2062: strcat(tmpout,fileres);
2063: strcat(tmpout,postop);
2064: return tmpout;
2065: }
1.126 brouard 2066:
1.213 brouard 2067: /*************** function subdirfext3 ***********/
2068: char *subdirfext3(char fileres[], char *preop, char *postop)
2069: {
2070:
2071: /* Caution optionfilefiname is hidden */
2072: strcpy(tmpout,optionfilefiname);
2073: strcat(tmpout,"/");
2074: strcat(tmpout,preop);
2075: strcat(tmpout,fileres);
2076: strcat(tmpout,postop);
2077: return tmpout;
2078: }
2079:
1.162 brouard 2080: char *asc_diff_time(long time_sec, char ascdiff[])
2081: {
2082: long sec_left, days, hours, minutes;
2083: days = (time_sec) / (60*60*24);
2084: sec_left = (time_sec) % (60*60*24);
2085: hours = (sec_left) / (60*60) ;
2086: sec_left = (sec_left) %(60*60);
2087: minutes = (sec_left) /60;
2088: sec_left = (sec_left) % (60);
2089: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2090: return ascdiff;
2091: }
2092:
1.126 brouard 2093: /***************** f1dim *************************/
2094: extern int ncom;
2095: extern double *pcom,*xicom;
2096: extern double (*nrfunc)(double []);
2097:
2098: double f1dim(double x)
2099: {
2100: int j;
2101: double f;
2102: double *xt;
2103:
2104: xt=vector(1,ncom);
2105: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2106: f=(*nrfunc)(xt);
2107: free_vector(xt,1,ncom);
2108: return f;
2109: }
2110:
2111: /*****************brent *************************/
2112: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2113: {
2114: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2115: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2116: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2117: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2118: * returned function value.
2119: */
1.126 brouard 2120: int iter;
2121: double a,b,d,etemp;
1.159 brouard 2122: double fu=0,fv,fw,fx;
1.164 brouard 2123: double ftemp=0.;
1.126 brouard 2124: double p,q,r,tol1,tol2,u,v,w,x,xm;
2125: double e=0.0;
2126:
2127: a=(ax < cx ? ax : cx);
2128: b=(ax > cx ? ax : cx);
2129: x=w=v=bx;
2130: fw=fv=fx=(*f)(x);
2131: for (iter=1;iter<=ITMAX;iter++) {
2132: xm=0.5*(a+b);
2133: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2134: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2135: printf(".");fflush(stdout);
2136: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2137: #ifdef DEBUGBRENT
1.126 brouard 2138: 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);
2139: 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);
2140: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2141: #endif
2142: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2143: *xmin=x;
2144: return fx;
2145: }
2146: ftemp=fu;
2147: if (fabs(e) > tol1) {
2148: r=(x-w)*(fx-fv);
2149: q=(x-v)*(fx-fw);
2150: p=(x-v)*q-(x-w)*r;
2151: q=2.0*(q-r);
2152: if (q > 0.0) p = -p;
2153: q=fabs(q);
2154: etemp=e;
2155: e=d;
2156: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2157: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2158: else {
1.224 brouard 2159: d=p/q;
2160: u=x+d;
2161: if (u-a < tol2 || b-u < tol2)
2162: d=SIGN(tol1,xm-x);
1.126 brouard 2163: }
2164: } else {
2165: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2166: }
2167: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2168: fu=(*f)(u);
2169: if (fu <= fx) {
2170: if (u >= x) a=x; else b=x;
2171: SHFT(v,w,x,u)
1.183 brouard 2172: SHFT(fv,fw,fx,fu)
2173: } else {
2174: if (u < x) a=u; else b=u;
2175: if (fu <= fw || w == x) {
1.224 brouard 2176: v=w;
2177: w=u;
2178: fv=fw;
2179: fw=fu;
1.183 brouard 2180: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2181: v=u;
2182: fv=fu;
1.183 brouard 2183: }
2184: }
1.126 brouard 2185: }
2186: nrerror("Too many iterations in brent");
2187: *xmin=x;
2188: return fx;
2189: }
2190:
2191: /****************** mnbrak ***********************/
2192:
2193: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2194: double (*func)(double))
1.183 brouard 2195: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2196: the downhill direction (defined by the function as evaluated at the initial points) and returns
2197: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2198: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2199: */
1.126 brouard 2200: double ulim,u,r,q, dum;
2201: double fu;
1.187 brouard 2202:
2203: double scale=10.;
2204: int iterscale=0;
2205:
2206: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2207: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2208:
2209:
2210: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2211: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2212: /* *bx = *ax - (*ax - *bx)/scale; */
2213: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2214: /* } */
2215:
1.126 brouard 2216: if (*fb > *fa) {
2217: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2218: SHFT(dum,*fb,*fa,dum)
2219: }
1.126 brouard 2220: *cx=(*bx)+GOLD*(*bx-*ax);
2221: *fc=(*func)(*cx);
1.183 brouard 2222: #ifdef DEBUG
1.224 brouard 2223: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2224: 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 2225: #endif
1.224 brouard 2226: 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 2227: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2228: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2229: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2230: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2231: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2232: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2233: fu=(*func)(u);
1.163 brouard 2234: #ifdef DEBUG
2235: /* f(x)=A(x-u)**2+f(u) */
2236: double A, fparabu;
2237: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2238: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2239: 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);
2240: 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 2241: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2242: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2243: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2244: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2245: #endif
1.184 brouard 2246: #ifdef MNBRAKORIGINAL
1.183 brouard 2247: #else
1.191 brouard 2248: /* if (fu > *fc) { */
2249: /* #ifdef DEBUG */
2250: /* printf("mnbrak4 fu > fc \n"); */
2251: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2252: /* #endif */
2253: /* /\* 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 *\\/ *\/ */
2254: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2255: /* dum=u; /\* Shifting c and u *\/ */
2256: /* u = *cx; */
2257: /* *cx = dum; */
2258: /* dum = fu; */
2259: /* fu = *fc; */
2260: /* *fc =dum; */
2261: /* } else { /\* end *\/ */
2262: /* #ifdef DEBUG */
2263: /* printf("mnbrak3 fu < fc \n"); */
2264: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2265: /* #endif */
2266: /* dum=u; /\* Shifting c and u *\/ */
2267: /* u = *cx; */
2268: /* *cx = dum; */
2269: /* dum = fu; */
2270: /* fu = *fc; */
2271: /* *fc =dum; */
2272: /* } */
1.224 brouard 2273: #ifdef DEBUGMNBRAK
2274: double A, fparabu;
2275: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2276: fparabu= *fa - A*(*ax-u)*(*ax-u);
2277: 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);
2278: 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 2279: #endif
1.191 brouard 2280: dum=u; /* Shifting c and u */
2281: u = *cx;
2282: *cx = dum;
2283: dum = fu;
2284: fu = *fc;
2285: *fc =dum;
1.183 brouard 2286: #endif
1.162 brouard 2287: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2288: #ifdef DEBUG
1.224 brouard 2289: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2290: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2291: #endif
1.126 brouard 2292: fu=(*func)(u);
2293: if (fu < *fc) {
1.183 brouard 2294: #ifdef DEBUG
1.224 brouard 2295: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2296: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2297: #endif
2298: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2299: SHFT(*fb,*fc,fu,(*func)(u))
2300: #ifdef DEBUG
2301: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2302: #endif
2303: }
1.162 brouard 2304: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2305: #ifdef DEBUG
1.224 brouard 2306: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2307: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2308: #endif
1.126 brouard 2309: u=ulim;
2310: fu=(*func)(u);
1.183 brouard 2311: } else { /* u could be left to b (if r > q parabola has a maximum) */
2312: #ifdef DEBUG
1.224 brouard 2313: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2314: 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 2315: #endif
1.126 brouard 2316: u=(*cx)+GOLD*(*cx-*bx);
2317: fu=(*func)(u);
1.224 brouard 2318: #ifdef DEBUG
2319: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2320: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2321: #endif
1.183 brouard 2322: } /* end tests */
1.126 brouard 2323: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2324: SHFT(*fa,*fb,*fc,fu)
2325: #ifdef DEBUG
1.224 brouard 2326: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2327: 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 2328: #endif
2329: } /* 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 2330: }
2331:
2332: /*************** linmin ************************/
1.162 brouard 2333: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2334: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2335: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2336: the value of func at the returned location p . This is actually all accomplished by calling the
2337: routines mnbrak and brent .*/
1.126 brouard 2338: int ncom;
2339: double *pcom,*xicom;
2340: double (*nrfunc)(double []);
2341:
1.224 brouard 2342: #ifdef LINMINORIGINAL
1.126 brouard 2343: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2344: #else
2345: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2346: #endif
1.126 brouard 2347: {
2348: double brent(double ax, double bx, double cx,
2349: double (*f)(double), double tol, double *xmin);
2350: double f1dim(double x);
2351: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2352: double *fc, double (*func)(double));
2353: int j;
2354: double xx,xmin,bx,ax;
2355: double fx,fb,fa;
1.187 brouard 2356:
1.203 brouard 2357: #ifdef LINMINORIGINAL
2358: #else
2359: double scale=10., axs, xxs; /* Scale added for infinity */
2360: #endif
2361:
1.126 brouard 2362: ncom=n;
2363: pcom=vector(1,n);
2364: xicom=vector(1,n);
2365: nrfunc=func;
2366: for (j=1;j<=n;j++) {
2367: pcom[j]=p[j];
1.202 brouard 2368: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2369: }
1.187 brouard 2370:
1.203 brouard 2371: #ifdef LINMINORIGINAL
2372: xx=1.;
2373: #else
2374: axs=0.0;
2375: xxs=1.;
2376: do{
2377: xx= xxs;
2378: #endif
1.187 brouard 2379: ax=0.;
2380: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2381: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2382: /* 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)) */
2383: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2384: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2385: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2386: /* 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 2387: #ifdef LINMINORIGINAL
2388: #else
2389: if (fx != fx){
1.224 brouard 2390: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2391: printf("|");
2392: fprintf(ficlog,"|");
1.203 brouard 2393: #ifdef DEBUGLINMIN
1.224 brouard 2394: 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 2395: #endif
2396: }
1.224 brouard 2397: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2398: #endif
2399:
1.191 brouard 2400: #ifdef DEBUGLINMIN
2401: 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 2402: 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 2403: #endif
1.224 brouard 2404: #ifdef LINMINORIGINAL
2405: #else
1.317 brouard 2406: if(fb == fx){ /* Flat function in the direction */
2407: xmin=xx;
1.224 brouard 2408: *flat=1;
1.317 brouard 2409: }else{
1.224 brouard 2410: *flat=0;
2411: #endif
2412: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2413: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2414: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2415: /* fmin = f(p[j] + xmin * xi[j]) */
2416: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2417: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2418: #ifdef DEBUG
1.224 brouard 2419: 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);
2420: 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);
2421: #endif
2422: #ifdef LINMINORIGINAL
2423: #else
2424: }
1.126 brouard 2425: #endif
1.191 brouard 2426: #ifdef DEBUGLINMIN
2427: printf("linmin end ");
1.202 brouard 2428: fprintf(ficlog,"linmin end ");
1.191 brouard 2429: #endif
1.126 brouard 2430: for (j=1;j<=n;j++) {
1.203 brouard 2431: #ifdef LINMINORIGINAL
2432: xi[j] *= xmin;
2433: #else
2434: #ifdef DEBUGLINMIN
2435: if(xxs <1.0)
2436: printf(" before xi[%d]=%12.8f", j,xi[j]);
2437: #endif
2438: 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) */
2439: #ifdef DEBUGLINMIN
2440: if(xxs <1.0)
2441: 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 );
2442: #endif
2443: #endif
1.187 brouard 2444: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2445: }
1.191 brouard 2446: #ifdef DEBUGLINMIN
1.203 brouard 2447: printf("\n");
1.191 brouard 2448: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2449: 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 2450: for (j=1;j<=n;j++) {
1.202 brouard 2451: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2452: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2453: if(j % ncovmodel == 0){
1.191 brouard 2454: printf("\n");
1.202 brouard 2455: fprintf(ficlog,"\n");
2456: }
1.191 brouard 2457: }
1.203 brouard 2458: #else
1.191 brouard 2459: #endif
1.126 brouard 2460: free_vector(xicom,1,n);
2461: free_vector(pcom,1,n);
2462: }
2463:
2464:
2465: /*************** powell ************************/
1.162 brouard 2466: /*
1.317 brouard 2467: Minimization of a function func of n variables. Input consists in an initial starting point
2468: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2469: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2470: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2471: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2472: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2473: */
1.224 brouard 2474: #ifdef LINMINORIGINAL
2475: #else
2476: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2477: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2478: #endif
1.126 brouard 2479: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2480: double (*func)(double []))
2481: {
1.224 brouard 2482: #ifdef LINMINORIGINAL
2483: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2484: double (*func)(double []));
1.224 brouard 2485: #else
1.241 brouard 2486: void linmin(double p[], double xi[], int n, double *fret,
2487: double (*func)(double []),int *flat);
1.224 brouard 2488: #endif
1.239 brouard 2489: int i,ibig,j,jk,k;
1.126 brouard 2490: double del,t,*pt,*ptt,*xit;
1.181 brouard 2491: double directest;
1.126 brouard 2492: double fp,fptt;
2493: double *xits;
2494: int niterf, itmp;
2495:
2496: pt=vector(1,n);
2497: ptt=vector(1,n);
2498: xit=vector(1,n);
2499: xits=vector(1,n);
2500: *fret=(*func)(p);
2501: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2502: rcurr_time = time(NULL);
1.126 brouard 2503: for (*iter=1;;++(*iter)) {
2504: ibig=0;
2505: del=0.0;
1.157 brouard 2506: rlast_time=rcurr_time;
2507: /* (void) gettimeofday(&curr_time,&tzp); */
2508: rcurr_time = time(NULL);
2509: curr_time = *localtime(&rcurr_time);
1.324 brouard 2510: 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);
2511: 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 2512: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2513: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2514: for (i=1;i<=n;i++) {
1.126 brouard 2515: fprintf(ficrespow," %.12lf", p[i]);
2516: }
1.239 brouard 2517: fprintf(ficrespow,"\n");fflush(ficrespow);
2518: printf("\n#model= 1 + age ");
2519: fprintf(ficlog,"\n#model= 1 + age ");
2520: if(nagesqr==1){
1.241 brouard 2521: printf(" + age*age ");
2522: fprintf(ficlog," + age*age ");
1.239 brouard 2523: }
2524: for(j=1;j <=ncovmodel-2;j++){
2525: if(Typevar[j]==0) {
2526: printf(" + V%d ",Tvar[j]);
2527: fprintf(ficlog," + V%d ",Tvar[j]);
2528: }else if(Typevar[j]==1) {
2529: printf(" + V%d*age ",Tvar[j]);
2530: fprintf(ficlog," + V%d*age ",Tvar[j]);
2531: }else if(Typevar[j]==2) {
2532: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2533: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2534: }
2535: }
1.126 brouard 2536: printf("\n");
1.239 brouard 2537: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2538: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2539: fprintf(ficlog,"\n");
1.239 brouard 2540: for(i=1,jk=1; i <=nlstate; i++){
2541: for(k=1; k <=(nlstate+ndeath); k++){
2542: if (k != i) {
2543: printf("%d%d ",i,k);
2544: fprintf(ficlog,"%d%d ",i,k);
2545: for(j=1; j <=ncovmodel; j++){
2546: printf("%12.7f ",p[jk]);
2547: fprintf(ficlog,"%12.7f ",p[jk]);
2548: jk++;
2549: }
2550: printf("\n");
2551: fprintf(ficlog,"\n");
2552: }
2553: }
2554: }
1.241 brouard 2555: if(*iter <=3 && *iter >1){
1.157 brouard 2556: tml = *localtime(&rcurr_time);
2557: strcpy(strcurr,asctime(&tml));
2558: rforecast_time=rcurr_time;
1.126 brouard 2559: itmp = strlen(strcurr);
2560: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2561: strcurr[itmp-1]='\0';
1.162 brouard 2562: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2563: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2564: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2565: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2566: forecast_time = *localtime(&rforecast_time);
2567: strcpy(strfor,asctime(&forecast_time));
2568: itmp = strlen(strfor);
2569: if(strfor[itmp-1]=='\n')
2570: strfor[itmp-1]='\0';
2571: 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);
2572: 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 2573: }
2574: }
1.187 brouard 2575: for (i=1;i<=n;i++) { /* For each direction i */
2576: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2577: fptt=(*fret);
2578: #ifdef DEBUG
1.203 brouard 2579: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2580: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2581: #endif
1.203 brouard 2582: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2583: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2584: #ifdef LINMINORIGINAL
1.188 brouard 2585: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2586: #else
2587: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2588: flatdir[i]=flat; /* Function is vanishing in that direction i */
2589: #endif
2590: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2591: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2592: /* because that direction will be replaced unless the gain del is small */
2593: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2594: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2595: /* with the new direction. */
2596: del=fabs(fptt-(*fret));
2597: ibig=i;
1.126 brouard 2598: }
2599: #ifdef DEBUG
2600: printf("%d %.12e",i,(*fret));
2601: fprintf(ficlog,"%d %.12e",i,(*fret));
2602: for (j=1;j<=n;j++) {
1.224 brouard 2603: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2604: printf(" x(%d)=%.12e",j,xit[j]);
2605: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2606: }
2607: for(j=1;j<=n;j++) {
1.225 brouard 2608: printf(" p(%d)=%.12e",j,p[j]);
2609: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2610: }
2611: printf("\n");
2612: fprintf(ficlog,"\n");
2613: #endif
1.187 brouard 2614: } /* end loop on each direction i */
2615: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2616: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2617: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2618: for(j=1;j<=n;j++) {
2619: if(flatdir[j] >0){
2620: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2621: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2622: }
1.319 brouard 2623: /* printf("\n"); */
2624: /* fprintf(ficlog,"\n"); */
2625: }
1.243 brouard 2626: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2627: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2628: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2629: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2630: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2631: /* decreased of more than 3.84 */
2632: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2633: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2634: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2635:
1.188 brouard 2636: /* Starting the program with initial values given by a former maximization will simply change */
2637: /* the scales of the directions and the directions, because the are reset to canonical directions */
2638: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2639: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2640: #ifdef DEBUG
2641: int k[2],l;
2642: k[0]=1;
2643: k[1]=-1;
2644: printf("Max: %.12e",(*func)(p));
2645: fprintf(ficlog,"Max: %.12e",(*func)(p));
2646: for (j=1;j<=n;j++) {
2647: printf(" %.12e",p[j]);
2648: fprintf(ficlog," %.12e",p[j]);
2649: }
2650: printf("\n");
2651: fprintf(ficlog,"\n");
2652: for(l=0;l<=1;l++) {
2653: for (j=1;j<=n;j++) {
2654: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2655: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2656: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2657: }
2658: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2659: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2660: }
2661: #endif
2662:
2663: free_vector(xit,1,n);
2664: free_vector(xits,1,n);
2665: free_vector(ptt,1,n);
2666: free_vector(pt,1,n);
2667: return;
1.192 brouard 2668: } /* enough precision */
1.240 brouard 2669: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2670: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2671: ptt[j]=2.0*p[j]-pt[j];
2672: xit[j]=p[j]-pt[j];
2673: pt[j]=p[j];
2674: }
1.181 brouard 2675: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2676: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2677: if (*iter <=4) {
1.225 brouard 2678: #else
2679: #endif
1.224 brouard 2680: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2681: #else
1.161 brouard 2682: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2683: #endif
1.162 brouard 2684: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2685: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2686: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2687: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2688: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2689: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2690: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2691: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2692: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2693: /* Even if f3 <f1, directest can be negative and t >0 */
2694: /* mu² and del² are equal when f3=f1 */
2695: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2696: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2697: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2698: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2699: #ifdef NRCORIGINAL
2700: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2701: #else
2702: 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 2703: t= t- del*SQR(fp-fptt);
1.183 brouard 2704: #endif
1.202 brouard 2705: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2706: #ifdef DEBUG
1.181 brouard 2707: 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);
2708: 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 2709: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2710: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2711: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2712: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2713: 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);
2714: 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);
2715: #endif
1.183 brouard 2716: #ifdef POWELLORIGINAL
2717: if (t < 0.0) { /* Then we use it for new direction */
2718: #else
1.182 brouard 2719: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2720: 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 2721: 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 2722: 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 2723: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2724: }
1.181 brouard 2725: if (directest < 0.0) { /* Then we use it for new direction */
2726: #endif
1.191 brouard 2727: #ifdef DEBUGLINMIN
1.234 brouard 2728: printf("Before linmin in direction P%d-P0\n",n);
2729: for (j=1;j<=n;j++) {
2730: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2731: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2732: if(j % ncovmodel == 0){
2733: printf("\n");
2734: fprintf(ficlog,"\n");
2735: }
2736: }
1.224 brouard 2737: #endif
2738: #ifdef LINMINORIGINAL
1.234 brouard 2739: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2740: #else
1.234 brouard 2741: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2742: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2743: #endif
1.234 brouard 2744:
1.191 brouard 2745: #ifdef DEBUGLINMIN
1.234 brouard 2746: for (j=1;j<=n;j++) {
2747: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2748: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2749: if(j % ncovmodel == 0){
2750: printf("\n");
2751: fprintf(ficlog,"\n");
2752: }
2753: }
1.224 brouard 2754: #endif
1.234 brouard 2755: for (j=1;j<=n;j++) {
2756: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2757: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2758: }
1.224 brouard 2759: #ifdef LINMINORIGINAL
2760: #else
1.234 brouard 2761: for (j=1, flatd=0;j<=n;j++) {
2762: if(flatdir[j]>0)
2763: flatd++;
2764: }
2765: if(flatd >0){
1.255 brouard 2766: printf("%d flat directions: ",flatd);
2767: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2768: for (j=1;j<=n;j++) {
2769: if(flatdir[j]>0){
2770: printf("%d ",j);
2771: fprintf(ficlog,"%d ",j);
2772: }
2773: }
2774: printf("\n");
2775: fprintf(ficlog,"\n");
1.319 brouard 2776: #ifdef FLATSUP
2777: free_vector(xit,1,n);
2778: free_vector(xits,1,n);
2779: free_vector(ptt,1,n);
2780: free_vector(pt,1,n);
2781: return;
2782: #endif
1.234 brouard 2783: }
1.191 brouard 2784: #endif
1.234 brouard 2785: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2786: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2787:
1.126 brouard 2788: #ifdef DEBUG
1.234 brouard 2789: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2790: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2791: for(j=1;j<=n;j++){
2792: printf(" %lf",xit[j]);
2793: fprintf(ficlog," %lf",xit[j]);
2794: }
2795: printf("\n");
2796: fprintf(ficlog,"\n");
1.126 brouard 2797: #endif
1.192 brouard 2798: } /* end of t or directest negative */
1.224 brouard 2799: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2800: #else
1.234 brouard 2801: } /* end if (fptt < fp) */
1.192 brouard 2802: #endif
1.225 brouard 2803: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2804: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2805: #else
1.224 brouard 2806: #endif
1.234 brouard 2807: } /* loop iteration */
1.126 brouard 2808: }
1.234 brouard 2809:
1.126 brouard 2810: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2811:
1.235 brouard 2812: 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 2813: {
1.279 brouard 2814: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2815: * (and selected quantitative values in nres)
2816: * by left multiplying the unit
2817: * matrix by transitions matrix until convergence is reached with precision ftolpl
2818: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2819: * Wx is row vector: population in state 1, population in state 2, population dead
2820: * or prevalence in state 1, prevalence in state 2, 0
2821: * newm is the matrix after multiplications, its rows are identical at a factor.
2822: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2823: * Output is prlim.
2824: * Initial matrix pimij
2825: */
1.206 brouard 2826: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2827: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2828: /* 0, 0 , 1} */
2829: /*
2830: * and after some iteration: */
2831: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2832: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2833: /* 0, 0 , 1} */
2834: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2835: /* {0.51571254859325999, 0.4842874514067399, */
2836: /* 0.51326036147820708, 0.48673963852179264} */
2837: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2838:
1.332 brouard 2839: int i, ii,j,k, k1;
1.209 brouard 2840: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2841: /* double **matprod2(); */ /* test */
1.218 brouard 2842: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2843: double **newm;
1.209 brouard 2844: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2845: int ncvloop=0;
1.288 brouard 2846: int first=0;
1.169 brouard 2847:
1.209 brouard 2848: min=vector(1,nlstate);
2849: max=vector(1,nlstate);
2850: meandiff=vector(1,nlstate);
2851:
1.218 brouard 2852: /* Starting with matrix unity */
1.126 brouard 2853: for (ii=1;ii<=nlstate+ndeath;ii++)
2854: for (j=1;j<=nlstate+ndeath;j++){
2855: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2856: }
1.169 brouard 2857:
2858: cov[1]=1.;
2859:
2860: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2861: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2862: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2863: ncvloop++;
1.126 brouard 2864: newm=savm;
2865: /* Covariates have to be included here again */
1.138 brouard 2866: cov[2]=agefin;
1.319 brouard 2867: if(nagesqr==1){
2868: cov[3]= agefin*agefin;
2869: }
1.332 brouard 2870: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
2871: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
2872: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
2873: if(Typevar[k1]==1){ /* A product with age */
2874: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
2875: }else{
2876: cov[2+nagesqr+k1]=precov[nres][k1];
2877: }
2878: }/* End of loop on model equation */
2879:
2880: /* Start of old code (replaced by a loop on position in the model equation */
2881: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
2882: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
2883: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
2884: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
2885: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
2886: /* * k 1 2 3 4 5 6 7 8 */
2887: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
2888: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
2889: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
2890: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
2891: /* *nsd=3 (1) (2) (3) */
2892: /* *TvarsD[nsd] [1]=2 1 3 */
2893: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
2894: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
2895: /* *Tage[] [1]=1 [2]=2 [3]=3 */
2896: /* *Tvard[] [1][1]=1 [2][1]=1 */
2897: /* * [1][2]=3 [2][2]=2 */
2898: /* *Tprod[](=k) [1]=1 [2]=8 */
2899: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
2900: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
2901: /* *TvarsDpType */
2902: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
2903: /* * nsd=1 (1) (2) */
2904: /* *TvarsD[nsd] 3 2 */
2905: /* *TnsdVar (3)=1 (2)=2 */
2906: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
2907: /* *Tage[] [1]=2 [2]= 3 */
2908: /* *\/ */
2909: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
2910: /* /\* 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)); *\/ */
2911: /* } */
2912: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
2913: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
2914: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
2915: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
2916: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
2917: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2918: /* /\* 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]); *\/ */
2919: /* } */
2920: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
2921: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
2922: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
2923: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
2924: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
2925: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
2926: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2927: /* } */
2928: /* /\* 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]); *\/ */
2929: /* } */
2930: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
2931: /* /\* 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]); *\/ */
2932: /* if(Dummy[Tvard[k][1]]==0){ */
2933: /* if(Dummy[Tvard[k][2]]==0){ */
2934: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
2935: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
2936: /* }else{ */
2937: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
2938: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
2939: /* } */
2940: /* }else{ */
2941: /* if(Dummy[Tvard[k][2]]==0){ */
2942: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
2943: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
2944: /* }else{ */
2945: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
2946: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
2947: /* } */
2948: /* } */
2949: /* } /\* End product without age *\/ */
2950: /* ENd of old code */
1.138 brouard 2951: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2952: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2953: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2954: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2955: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 2956: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2957: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2958:
1.126 brouard 2959: savm=oldm;
2960: oldm=newm;
1.209 brouard 2961:
2962: for(j=1; j<=nlstate; j++){
2963: max[j]=0.;
2964: min[j]=1.;
2965: }
2966: for(i=1;i<=nlstate;i++){
2967: sumnew=0;
2968: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2969: for(j=1; j<=nlstate; j++){
2970: prlim[i][j]= newm[i][j]/(1-sumnew);
2971: max[j]=FMAX(max[j],prlim[i][j]);
2972: min[j]=FMIN(min[j],prlim[i][j]);
2973: }
2974: }
2975:
1.126 brouard 2976: maxmax=0.;
1.209 brouard 2977: for(j=1; j<=nlstate; j++){
2978: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2979: maxmax=FMAX(maxmax,meandiff[j]);
2980: /* 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 2981: } /* j loop */
1.203 brouard 2982: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2983: /* 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 2984: if(maxmax < ftolpl){
1.209 brouard 2985: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2986: free_vector(min,1,nlstate);
2987: free_vector(max,1,nlstate);
2988: free_vector(meandiff,1,nlstate);
1.126 brouard 2989: return prlim;
2990: }
1.288 brouard 2991: } /* agefin loop */
1.208 brouard 2992: /* After some age loop it doesn't converge */
1.288 brouard 2993: if(!first){
2994: first=1;
2995: 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 2996: 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);
2997: }else if (first >=1 && 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: first++;
3000: }else if (first ==10){
3001: 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);
3002: 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");
3003: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
3004: first++;
1.288 brouard 3005: }
3006:
1.209 brouard 3007: /* 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); */
3008: free_vector(min,1,nlstate);
3009: free_vector(max,1,nlstate);
3010: free_vector(meandiff,1,nlstate);
1.208 brouard 3011:
1.169 brouard 3012: return prlim; /* should not reach here */
1.126 brouard 3013: }
3014:
1.217 brouard 3015:
3016: /**** Back Prevalence limit (stable or period prevalence) ****************/
3017:
1.218 brouard 3018: /* 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) */
3019: /* 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 3020: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 3021: {
1.264 brouard 3022: /* 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 3023: matrix by transitions matrix until convergence is reached with precision ftolpl */
3024: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
3025: /* Wx is row vector: population in state 1, population in state 2, population dead */
3026: /* or prevalence in state 1, prevalence in state 2, 0 */
3027: /* newm is the matrix after multiplications, its rows are identical at a factor */
3028: /* Initial matrix pimij */
3029: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
3030: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
3031: /* 0, 0 , 1} */
3032: /*
3033: * and after some iteration: */
3034: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
3035: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
3036: /* 0, 0 , 1} */
3037: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
3038: /* {0.51571254859325999, 0.4842874514067399, */
3039: /* 0.51326036147820708, 0.48673963852179264} */
3040: /* If we start from prlim again, prlim tends to a constant matrix */
3041:
1.332 brouard 3042: int i, ii,j,k, k1;
1.247 brouard 3043: int first=0;
1.217 brouard 3044: double *min, *max, *meandiff, maxmax,sumnew=0.;
3045: /* double **matprod2(); */ /* test */
3046: double **out, cov[NCOVMAX+1], **bmij();
3047: double **newm;
1.218 brouard 3048: double **dnewm, **doldm, **dsavm; /* for use */
3049: double **oldm, **savm; /* for use */
3050:
1.217 brouard 3051: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
3052: int ncvloop=0;
3053:
3054: min=vector(1,nlstate);
3055: max=vector(1,nlstate);
3056: meandiff=vector(1,nlstate);
3057:
1.266 brouard 3058: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
3059: oldm=oldms; savm=savms;
3060:
3061: /* Starting with matrix unity */
3062: for (ii=1;ii<=nlstate+ndeath;ii++)
3063: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 3064: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3065: }
3066:
3067: cov[1]=1.;
3068:
3069: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3070: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 3071: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 3072: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
3073: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 3074: ncvloop++;
1.218 brouard 3075: newm=savm; /* oldm should be kept from previous iteration or unity at start */
3076: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 3077: /* Covariates have to be included here again */
3078: cov[2]=agefin;
1.319 brouard 3079: if(nagesqr==1){
1.217 brouard 3080: cov[3]= agefin*agefin;;
1.319 brouard 3081: }
1.332 brouard 3082: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3083: if(Typevar[k1]==1){ /* A product with age */
3084: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 3085: }else{
1.332 brouard 3086: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 3087: }
1.332 brouard 3088: }/* End of loop on model equation */
3089:
3090: /* Old code */
3091:
3092: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
3093: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
3094: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
3095: /* /\* 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)); *\/ */
3096: /* } */
3097: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
3098: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3099: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3100: /* /\* /\\* 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])]); *\\/ *\/ */
3101: /* /\* } *\/ */
3102: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3103: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3104: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3105: /* /\* 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]); *\/ */
3106: /* } */
3107: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
3108: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
3109: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
3110: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3111: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3112: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
3113: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3114: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3115: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3116: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3117: /* } */
3118: /* /\* 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]); *\/ */
3119: /* } */
3120: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3121: /* /\* 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]); *\/ */
3122: /* if(Dummy[Tvard[k][1]]==0){ */
3123: /* if(Dummy[Tvard[k][2]]==0){ */
3124: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3125: /* }else{ */
3126: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3127: /* } */
3128: /* }else{ */
3129: /* if(Dummy[Tvard[k][2]]==0){ */
3130: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3131: /* }else{ */
3132: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3133: /* } */
3134: /* } */
3135: /* } */
1.217 brouard 3136:
3137: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3138: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3139: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3140: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3141: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3142: /* ij should be linked to the correct index of cov */
3143: /* age and covariate values ij are in 'cov', but we need to pass
3144: * ij for the observed prevalence at age and status and covariate
3145: * number: prevacurrent[(int)agefin][ii][ij]
3146: */
3147: /* 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 *\/ */
3148: /* 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 *\/ */
3149: 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 3150: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3151: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3152: /* for(i=1; i<=nlstate+ndeath; i++) { */
3153: /* printf("%d newm= ",i); */
3154: /* for(j=1;j<=nlstate+ndeath;j++) { */
3155: /* printf("%f ",newm[i][j]); */
3156: /* } */
3157: /* printf("oldm * "); */
3158: /* for(j=1;j<=nlstate+ndeath;j++) { */
3159: /* printf("%f ",oldm[i][j]); */
3160: /* } */
1.268 brouard 3161: /* printf(" bmmij "); */
1.266 brouard 3162: /* for(j=1;j<=nlstate+ndeath;j++) { */
3163: /* printf("%f ",pmmij[i][j]); */
3164: /* } */
3165: /* printf("\n"); */
3166: /* } */
3167: /* } */
1.217 brouard 3168: savm=oldm;
3169: oldm=newm;
1.266 brouard 3170:
1.217 brouard 3171: for(j=1; j<=nlstate; j++){
3172: max[j]=0.;
3173: min[j]=1.;
3174: }
3175: for(j=1; j<=nlstate; j++){
3176: for(i=1;i<=nlstate;i++){
1.234 brouard 3177: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3178: bprlim[i][j]= newm[i][j];
3179: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3180: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3181: }
3182: }
1.218 brouard 3183:
1.217 brouard 3184: maxmax=0.;
3185: for(i=1; i<=nlstate; i++){
1.318 brouard 3186: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3187: maxmax=FMAX(maxmax,meandiff[i]);
3188: /* 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 3189: } /* i loop */
1.217 brouard 3190: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3191: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3192: if(maxmax < ftolpl){
1.220 brouard 3193: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3194: free_vector(min,1,nlstate);
3195: free_vector(max,1,nlstate);
3196: free_vector(meandiff,1,nlstate);
3197: return bprlim;
3198: }
1.288 brouard 3199: } /* agefin loop */
1.217 brouard 3200: /* After some age loop it doesn't converge */
1.288 brouard 3201: if(!first){
1.247 brouard 3202: first=1;
3203: 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\
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: }
3206: 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 3207: 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);
3208: /* 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); */
3209: free_vector(min,1,nlstate);
3210: free_vector(max,1,nlstate);
3211: free_vector(meandiff,1,nlstate);
3212:
3213: return bprlim; /* should not reach here */
3214: }
3215:
1.126 brouard 3216: /*************** transition probabilities ***************/
3217:
3218: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3219: {
1.138 brouard 3220: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3221: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3222: model to the ncovmodel covariates (including constant and age).
3223: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3224: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3225: ncth covariate in the global vector x is given by the formula:
3226: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3227: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3228: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3229: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3230: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3231: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3232: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3233: */
3234: double s1, lnpijopii;
1.126 brouard 3235: /*double t34;*/
1.164 brouard 3236: int i,j, nc, ii, jj;
1.126 brouard 3237:
1.223 brouard 3238: for(i=1; i<= nlstate; i++){
3239: for(j=1; j<i;j++){
3240: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3241: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3242: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3243: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3244: }
3245: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3246: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3247: }
3248: for(j=i+1; j<=nlstate+ndeath;j++){
3249: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3250: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3251: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3252: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3253: }
3254: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3255: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3256: }
3257: }
1.218 brouard 3258:
1.223 brouard 3259: for(i=1; i<= nlstate; i++){
3260: s1=0;
3261: for(j=1; j<i; j++){
3262: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 brouard 3263: /* 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 3264: }
3265: for(j=i+1; j<=nlstate+ndeath; j++){
3266: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 brouard 3267: /* 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 3268: }
3269: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3270: ps[i][i]=1./(s1+1.);
3271: /* Computing other pijs */
3272: for(j=1; j<i; j++)
1.325 brouard 3273: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3274: for(j=i+1; j<=nlstate+ndeath; j++)
3275: ps[i][j]= exp(ps[i][j])*ps[i][i];
3276: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3277: } /* end i */
1.218 brouard 3278:
1.223 brouard 3279: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3280: for(jj=1; jj<= nlstate+ndeath; jj++){
3281: ps[ii][jj]=0;
3282: ps[ii][ii]=1;
3283: }
3284: }
1.294 brouard 3285:
3286:
1.223 brouard 3287: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3288: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3289: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3290: /* } */
3291: /* printf("\n "); */
3292: /* } */
3293: /* printf("\n ");printf("%lf ",cov[2]);*/
3294: /*
3295: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3296: goto end;*/
1.266 brouard 3297: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3298: }
3299:
1.218 brouard 3300: /*************** backward transition probabilities ***************/
3301:
3302: /* 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 ) */
3303: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3304: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3305: {
1.302 brouard 3306: /* 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 3307: * 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 3308: */
1.218 brouard 3309: int i, ii, j,k;
1.222 brouard 3310:
3311: double **out, **pmij();
3312: double sumnew=0.;
1.218 brouard 3313: double agefin;
1.292 brouard 3314: 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 3315: double **dnewm, **dsavm, **doldm;
3316: double **bbmij;
3317:
1.218 brouard 3318: doldm=ddoldms; /* global pointers */
1.222 brouard 3319: dnewm=ddnewms;
3320: dsavm=ddsavms;
1.318 brouard 3321:
3322: /* Debug */
3323: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3324: agefin=cov[2];
1.268 brouard 3325: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3326: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3327: the observed prevalence (with this covariate ij) at beginning of transition */
3328: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3329:
3330: /* P_x */
1.325 brouard 3331: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3332: /* outputs pmmij which is a stochastic matrix in row */
3333:
3334: /* Diag(w_x) */
1.292 brouard 3335: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3336: sumnew=0.;
1.269 brouard 3337: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3338: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3339: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3340: sumnew+=prevacurrent[(int)agefin][ii][ij];
3341: }
3342: if(sumnew >0.01){ /* At least some value in the prevalence */
3343: for (ii=1;ii<=nlstate+ndeath;ii++){
3344: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3345: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3346: }
3347: }else{
3348: for (ii=1;ii<=nlstate+ndeath;ii++){
3349: for (j=1;j<=nlstate+ndeath;j++)
3350: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3351: }
3352: /* if(sumnew <0.9){ */
3353: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3354: /* } */
3355: }
3356: k3=0.0; /* We put the last diagonal to 0 */
3357: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3358: doldm[ii][ii]= k3;
3359: }
3360: /* End doldm, At the end doldm is diag[(w_i)] */
3361:
1.292 brouard 3362: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3363: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3364:
1.292 brouard 3365: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3366: /* 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 3367: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3368: sumnew=0.;
1.222 brouard 3369: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3370: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3371: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3372: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3373: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3374: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3375: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3376: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3377: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3378: /* }else */
1.268 brouard 3379: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3380: } /*End ii */
3381: } /* 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 */
3382:
1.292 brouard 3383: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3384: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3385: /* end bmij */
1.266 brouard 3386: return ps; /*pointer is unchanged */
1.218 brouard 3387: }
1.217 brouard 3388: /*************** transition probabilities ***************/
3389:
1.218 brouard 3390: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3391: {
3392: /* According to parameters values stored in x and the covariate's values stored in cov,
3393: computes the probability to be observed in state j being in state i by appying the
3394: model to the ncovmodel covariates (including constant and age).
3395: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3396: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3397: ncth covariate in the global vector x is given by the formula:
3398: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3399: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3400: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3401: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3402: Outputs ps[i][j] the probability to be observed in j being in j according to
3403: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3404: */
3405: double s1, lnpijopii;
3406: /*double t34;*/
3407: int i,j, nc, ii, jj;
3408:
1.234 brouard 3409: for(i=1; i<= nlstate; i++){
3410: for(j=1; j<i;j++){
3411: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3412: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3413: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3414: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3415: }
3416: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3417: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3418: }
3419: for(j=i+1; j<=nlstate+ndeath;j++){
3420: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3421: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3422: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3423: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3424: }
3425: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3426: }
3427: }
3428:
3429: for(i=1; i<= nlstate; i++){
3430: s1=0;
3431: for(j=1; j<i; j++){
3432: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3433: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3434: }
3435: for(j=i+1; j<=nlstate+ndeath; j++){
3436: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3437: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3438: }
3439: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3440: ps[i][i]=1./(s1+1.);
3441: /* Computing other pijs */
3442: for(j=1; j<i; j++)
3443: ps[i][j]= exp(ps[i][j])*ps[i][i];
3444: for(j=i+1; j<=nlstate+ndeath; j++)
3445: ps[i][j]= exp(ps[i][j])*ps[i][i];
3446: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3447: } /* end i */
3448:
3449: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3450: for(jj=1; jj<= nlstate+ndeath; jj++){
3451: ps[ii][jj]=0;
3452: ps[ii][ii]=1;
3453: }
3454: }
1.296 brouard 3455: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3456: for(jj=1; jj<= nlstate+ndeath; jj++){
3457: s1=0.;
3458: for(ii=1; ii<= nlstate+ndeath; ii++){
3459: s1+=ps[ii][jj];
3460: }
3461: for(ii=1; ii<= nlstate; ii++){
3462: ps[ii][jj]=ps[ii][jj]/s1;
3463: }
3464: }
3465: /* Transposition */
3466: for(jj=1; jj<= nlstate+ndeath; jj++){
3467: for(ii=jj; ii<= nlstate+ndeath; ii++){
3468: s1=ps[ii][jj];
3469: ps[ii][jj]=ps[jj][ii];
3470: ps[jj][ii]=s1;
3471: }
3472: }
3473: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3474: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3475: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3476: /* } */
3477: /* printf("\n "); */
3478: /* } */
3479: /* printf("\n ");printf("%lf ",cov[2]);*/
3480: /*
3481: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3482: goto end;*/
3483: return ps;
1.217 brouard 3484: }
3485:
3486:
1.126 brouard 3487: /**************** Product of 2 matrices ******************/
3488:
1.145 brouard 3489: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3490: {
3491: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3492: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3493: /* in, b, out are matrice of pointers which should have been initialized
3494: before: only the contents of out is modified. The function returns
3495: a pointer to pointers identical to out */
1.145 brouard 3496: int i, j, k;
1.126 brouard 3497: for(i=nrl; i<= nrh; i++)
1.145 brouard 3498: for(k=ncolol; k<=ncoloh; k++){
3499: out[i][k]=0.;
3500: for(j=ncl; j<=nch; j++)
3501: out[i][k] +=in[i][j]*b[j][k];
3502: }
1.126 brouard 3503: return out;
3504: }
3505:
3506:
3507: /************* Higher Matrix Product ***************/
3508:
1.235 brouard 3509: 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 3510: {
1.336 ! brouard 3511: /* Already optimized with precov.
! 3512: 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 3513: 'nhstepm*hstepm*stepm' months (i.e. until
3514: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3515: nhstepm*hstepm matrices.
3516: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3517: (typically every 2 years instead of every month which is too big
3518: for the memory).
3519: Model is determined by parameters x and covariates have to be
3520: included manually here.
3521:
3522: */
3523:
1.330 brouard 3524: int i, j, d, h, k, k1;
1.131 brouard 3525: double **out, cov[NCOVMAX+1];
1.126 brouard 3526: double **newm;
1.187 brouard 3527: double agexact;
1.214 brouard 3528: double agebegin, ageend;
1.126 brouard 3529:
3530: /* Hstepm could be zero and should return the unit matrix */
3531: for (i=1;i<=nlstate+ndeath;i++)
3532: for (j=1;j<=nlstate+ndeath;j++){
3533: oldm[i][j]=(i==j ? 1.0 : 0.0);
3534: po[i][j][0]=(i==j ? 1.0 : 0.0);
3535: }
3536: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3537: for(h=1; h <=nhstepm; h++){
3538: for(d=1; d <=hstepm; d++){
3539: newm=savm;
3540: /* Covariates have to be included here again */
3541: cov[1]=1.;
1.214 brouard 3542: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3543: cov[2]=agexact;
1.319 brouard 3544: if(nagesqr==1){
1.227 brouard 3545: cov[3]= agexact*agexact;
1.319 brouard 3546: }
1.330 brouard 3547: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
3548: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
3549: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.332 brouard 3550: if(Typevar[k1]==1){ /* A product with age */
3551: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
3552: }else{
3553: cov[2+nagesqr+k1]=precov[nres][k1];
3554: }
3555: }/* End of loop on model equation */
3556: /* Old code */
3557: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
3558: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
3559: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
3560: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
3561: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
3562: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3563: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3564: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
3565: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
3566: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
3567: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
3568: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
3569: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3570: /* /\* 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]])); *\/ */
3571: /* 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); */
3572: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3573: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
3574: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
3575: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
3576: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
3577: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
3578: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3579: /* 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]]); */
3580: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3581: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
3582: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
3583: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
3584: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
3585: /* 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]); */
3586: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3587:
3588: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
3589: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
3590: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
3591: /* /\* *\/ */
1.330 brouard 3592: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3593: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3594: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 3595: /* /\*cptcovage=2 1 2 *\/ */
3596: /* /\*Tage[k]= 5 8 *\/ */
3597: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
3598: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3599: /* 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]]); */
3600: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3601: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
3602: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
3603: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
3604: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
3605: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
3606: /* /\* 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); *\/ */
3607: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
3608: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
3609: /* /\* } *\/ */
3610: /* /\* 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]); *\/ */
3611: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
3612: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
3613: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
3614: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
3615: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
3616: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
3617: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
3618: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
3619: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 3620:
1.332 brouard 3621: /* /\* 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])]); *\/ */
3622: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3623: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
3624: /* 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]]); */
3625: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3626:
3627: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
3628: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
3629: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3630: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
3631: /* /\* 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]])]; *\/ */
3632: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
3633: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
3634: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
3635: /* /\* } *\/ */
3636: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
3637: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
3638: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
3639: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3640: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
3641: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
3642: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3643: /* /\* } *\/ */
3644: /* /\* }/\\*end of products quantitative *\\/ *\/ */
3645: /* }/\*end of products *\/ */
3646: /* } /\* End of loop on model equation *\/ */
1.235 brouard 3647: /* for (k=1; k<=cptcovn;k++) */
3648: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3649: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3650: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3651: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3652: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3653:
3654:
1.126 brouard 3655: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3656: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3657: /* right multiplication of oldm by the current matrix */
1.126 brouard 3658: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3659: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3660: /* if((int)age == 70){ */
3661: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3662: /* for(i=1; i<=nlstate+ndeath; i++) { */
3663: /* printf("%d pmmij ",i); */
3664: /* for(j=1;j<=nlstate+ndeath;j++) { */
3665: /* printf("%f ",pmmij[i][j]); */
3666: /* } */
3667: /* printf(" oldm "); */
3668: /* for(j=1;j<=nlstate+ndeath;j++) { */
3669: /* printf("%f ",oldm[i][j]); */
3670: /* } */
3671: /* printf("\n"); */
3672: /* } */
3673: /* } */
1.126 brouard 3674: savm=oldm;
3675: oldm=newm;
3676: }
3677: for(i=1; i<=nlstate+ndeath; i++)
3678: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3679: po[i][j][h]=newm[i][j];
3680: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3681: }
1.128 brouard 3682: /*printf("h=%d ",h);*/
1.126 brouard 3683: } /* end h */
1.267 brouard 3684: /* printf("\n H=%d \n",h); */
1.126 brouard 3685: return po;
3686: }
3687:
1.217 brouard 3688: /************* Higher Back Matrix Product ***************/
1.218 brouard 3689: /* 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 3690: 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 3691: {
1.332 brouard 3692: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
3693: computes the transition matrix starting at age 'age' over
1.217 brouard 3694: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3695: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3696: nhstepm*hstepm matrices.
3697: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3698: (typically every 2 years instead of every month which is too big
1.217 brouard 3699: for the memory).
1.218 brouard 3700: Model is determined by parameters x and covariates have to be
1.266 brouard 3701: included manually here. Then we use a call to bmij(x and cov)
3702: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3703: */
1.217 brouard 3704:
1.332 brouard 3705: int i, j, d, h, k, k1;
1.266 brouard 3706: double **out, cov[NCOVMAX+1], **bmij();
3707: double **newm, ***newmm;
1.217 brouard 3708: double agexact;
3709: double agebegin, ageend;
1.222 brouard 3710: double **oldm, **savm;
1.217 brouard 3711:
1.266 brouard 3712: newmm=po; /* To be saved */
3713: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3714: /* Hstepm could be zero and should return the unit matrix */
3715: for (i=1;i<=nlstate+ndeath;i++)
3716: for (j=1;j<=nlstate+ndeath;j++){
3717: oldm[i][j]=(i==j ? 1.0 : 0.0);
3718: po[i][j][0]=(i==j ? 1.0 : 0.0);
3719: }
3720: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3721: for(h=1; h <=nhstepm; h++){
3722: for(d=1; d <=hstepm; d++){
3723: newm=savm;
3724: /* Covariates have to be included here again */
3725: cov[1]=1.;
1.271 brouard 3726: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3727: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3728: /* Debug */
3729: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3730: cov[2]=agexact;
1.332 brouard 3731: if(nagesqr==1){
1.222 brouard 3732: cov[3]= agexact*agexact;
1.332 brouard 3733: }
3734: /** New code */
3735: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3736: if(Typevar[k1]==1){ /* A product with age */
3737: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 3738: }else{
1.332 brouard 3739: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 3740: }
1.332 brouard 3741: }/* End of loop on model equation */
3742: /** End of new code */
3743: /** This was old code */
3744: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
3745: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3746: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3747: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
3748: /* /\* 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)); *\/ */
3749: /* } */
3750: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3751: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3752: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3753: /* /\* 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]); *\/ */
3754: /* } */
3755: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
3756: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
3757: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3758: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3759: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3760: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3761: /* } */
3762: /* /\* 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]); *\/ */
3763: /* } */
3764: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
3765: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3766: /* if(Dummy[Tvard[k][1]]==0){ */
3767: /* if(Dummy[Tvard[k][2]]==0){ */
3768: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
3769: /* }else{ */
3770: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3771: /* } */
3772: /* }else{ */
3773: /* if(Dummy[Tvard[k][2]]==0){ */
3774: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3775: /* }else{ */
3776: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3777: /* } */
3778: /* } */
3779: /* } */
3780: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
3781: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
3782: /** End of old code */
3783:
1.218 brouard 3784: /* Careful transposed matrix */
1.266 brouard 3785: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3786: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3787: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3788: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3789: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3790: /* if((int)age == 70){ */
3791: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3792: /* for(i=1; i<=nlstate+ndeath; i++) { */
3793: /* printf("%d pmmij ",i); */
3794: /* for(j=1;j<=nlstate+ndeath;j++) { */
3795: /* printf("%f ",pmmij[i][j]); */
3796: /* } */
3797: /* printf(" oldm "); */
3798: /* for(j=1;j<=nlstate+ndeath;j++) { */
3799: /* printf("%f ",oldm[i][j]); */
3800: /* } */
3801: /* printf("\n"); */
3802: /* } */
3803: /* } */
3804: savm=oldm;
3805: oldm=newm;
3806: }
3807: for(i=1; i<=nlstate+ndeath; i++)
3808: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3809: po[i][j][h]=newm[i][j];
1.268 brouard 3810: /* if(h==nhstepm) */
3811: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3812: }
1.268 brouard 3813: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3814: } /* end h */
1.268 brouard 3815: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3816: return po;
3817: }
3818:
3819:
1.162 brouard 3820: #ifdef NLOPT
3821: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3822: double fret;
3823: double *xt;
3824: int j;
3825: myfunc_data *d2 = (myfunc_data *) pd;
3826: /* xt = (p1-1); */
3827: xt=vector(1,n);
3828: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3829:
3830: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3831: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3832: printf("Function = %.12lf ",fret);
3833: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3834: printf("\n");
3835: free_vector(xt,1,n);
3836: return fret;
3837: }
3838: #endif
1.126 brouard 3839:
3840: /*************** log-likelihood *************/
3841: double func( double *x)
3842: {
1.336 ! brouard 3843: int i, ii, j, k, mi, d, kk, kf=0;
1.226 brouard 3844: int ioffset=0;
3845: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3846: double **out;
3847: double lli; /* Individual log likelihood */
3848: int s1, s2;
1.228 brouard 3849: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
1.336 ! brouard 3850:
1.226 brouard 3851: double bbh, survp;
3852: double agexact;
1.336 ! brouard 3853: double agebegin, ageend;
1.226 brouard 3854: /*extern weight */
3855: /* We are differentiating ll according to initial status */
3856: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3857: /*for(i=1;i<imx;i++)
3858: printf(" %d\n",s[4][i]);
3859: */
1.162 brouard 3860:
1.226 brouard 3861: ++countcallfunc;
1.162 brouard 3862:
1.226 brouard 3863: cov[1]=1.;
1.126 brouard 3864:
1.226 brouard 3865: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3866: ioffset=0;
1.226 brouard 3867: if(mle==1){
3868: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3869: /* Computes the values of the ncovmodel covariates of the model
3870: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3871: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3872: to be observed in j being in i according to the model.
3873: */
1.243 brouard 3874: ioffset=2+nagesqr ;
1.233 brouard 3875: /* Fixed */
1.336 ! brouard 3876: for (kf=1; kf<=ncovf;kf++){ /* For each fixed covariate dummu or quant or prod */
1.319 brouard 3877: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3878: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3879: /* 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 3880: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.336 ! brouard 3881: cov[ioffset+TvarFind[kf]]=covar[Tvar[TvarFind[kf]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (TvarFind[1]=6)*/
1.319 brouard 3882: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3883: }
1.226 brouard 3884: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3885: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3886: has been calculated etc */
3887: /* For an individual i, wav[i] gives the number of effective waves */
3888: /* We compute the contribution to Likelihood of each effective transition
3889: mw[mi][i] is real wave of the mi th effectve wave */
3890: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3891: s2=s[mw[mi+1][i]][i];
3892: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3893: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3894: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3895: */
1.336 ! brouard 3896: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
! 3897: /* Wave varying (but not age varying) */
1.319 brouard 3898: 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*/
3899: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? */
1.242 brouard 3900: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3901: }
3902: for (ii=1;ii<=nlstate+ndeath;ii++)
3903: for (j=1;j<=nlstate+ndeath;j++){
3904: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3905: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3906: }
1.336 ! brouard 3907:
! 3908: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
! 3909: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
1.234 brouard 3910: for(d=0; d<dh[mi][i]; d++){
3911: newm=savm;
3912: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3913: cov[2]=agexact;
3914: if(nagesqr==1)
3915: cov[3]= agexact*agexact; /* Should be changed here */
3916: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3917: if(!FixedV[Tvar[Tage[kk]]])
3918: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3919: else
3920: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3921: }
3922: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3923: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3924: savm=oldm;
3925: oldm=newm;
3926: } /* end mult */
3927:
3928: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3929: /* But now since version 0.9 we anticipate for bias at large stepm.
3930: * If stepm is larger than one month (smallest stepm) and if the exact delay
3931: * (in months) between two waves is not a multiple of stepm, we rounded to
3932: * the nearest (and in case of equal distance, to the lowest) interval but now
3933: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3934: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3935: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3936: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3937: * -stepm/2 to stepm/2 .
3938: * For stepm=1 the results are the same as for previous versions of Imach.
3939: * For stepm > 1 the results are less biased than in previous versions.
3940: */
1.234 brouard 3941: s1=s[mw[mi][i]][i];
3942: s2=s[mw[mi+1][i]][i];
3943: bbh=(double)bh[mi][i]/(double)stepm;
3944: /* bias bh is positive if real duration
3945: * is higher than the multiple of stepm and negative otherwise.
3946: */
3947: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3948: if( s2 > nlstate){
3949: /* i.e. if s2 is a death state and if the date of death is known
3950: then the contribution to the likelihood is the probability to
3951: die between last step unit time and current step unit time,
3952: which is also equal to probability to die before dh
3953: minus probability to die before dh-stepm .
3954: In version up to 0.92 likelihood was computed
3955: as if date of death was unknown. Death was treated as any other
3956: health state: the date of the interview describes the actual state
3957: and not the date of a change in health state. The former idea was
3958: to consider that at each interview the state was recorded
3959: (healthy, disable or death) and IMaCh was corrected; but when we
3960: introduced the exact date of death then we should have modified
3961: the contribution of an exact death to the likelihood. This new
3962: contribution is smaller and very dependent of the step unit
3963: stepm. It is no more the probability to die between last interview
3964: and month of death but the probability to survive from last
3965: interview up to one month before death multiplied by the
3966: probability to die within a month. Thanks to Chris
3967: Jackson for correcting this bug. Former versions increased
3968: mortality artificially. The bad side is that we add another loop
3969: which slows down the processing. The difference can be up to 10%
3970: lower mortality.
3971: */
3972: /* If, at the beginning of the maximization mostly, the
3973: cumulative probability or probability to be dead is
3974: constant (ie = 1) over time d, the difference is equal to
3975: 0. out[s1][3] = savm[s1][3]: probability, being at state
3976: s1 at precedent wave, to be dead a month before current
3977: wave is equal to probability, being at state s1 at
3978: precedent wave, to be dead at mont of the current
3979: wave. Then the observed probability (that this person died)
3980: is null according to current estimated parameter. In fact,
3981: it should be very low but not zero otherwise the log go to
3982: infinity.
3983: */
1.183 brouard 3984: /* #ifdef INFINITYORIGINAL */
3985: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3986: /* #else */
3987: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3988: /* lli=log(mytinydouble); */
3989: /* else */
3990: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3991: /* #endif */
1.226 brouard 3992: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3993:
1.226 brouard 3994: } else if ( s2==-1 ) { /* alive */
3995: for (j=1,survp=0. ; j<=nlstate; j++)
3996: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3997: /*survp += out[s1][j]; */
3998: lli= log(survp);
3999: }
1.336 ! brouard 4000: /* else if (s2==-4) { */
! 4001: /* for (j=3,survp=0. ; j<=nlstate; j++) */
! 4002: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
! 4003: /* lli= log(survp); */
! 4004: /* } */
! 4005: /* else if (s2==-5) { */
! 4006: /* for (j=1,survp=0. ; j<=2; j++) */
! 4007: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
! 4008: /* lli= log(survp); */
! 4009: /* } */
1.226 brouard 4010: else{
4011: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4012: /* 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 */
4013: }
4014: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
4015: /*if(lli ==000.0)*/
4016: /*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); */
4017: ipmx +=1;
4018: sw += weight[i];
4019: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4020: /* if (lli < log(mytinydouble)){ */
4021: /* 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); */
4022: /* 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]); */
4023: /* } */
4024: } /* end of wave */
4025: } /* end of individual */
4026: } else if(mle==2){
4027: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 4028: ioffset=2+nagesqr ;
4029: for (k=1; k<=ncovf;k++)
4030: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 4031: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 4032: for(k=1; k <= ncovv ; k++){
4033: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4034: }
1.226 brouard 4035: for (ii=1;ii<=nlstate+ndeath;ii++)
4036: for (j=1;j<=nlstate+ndeath;j++){
4037: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4038: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4039: }
4040: for(d=0; d<=dh[mi][i]; d++){
4041: newm=savm;
4042: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4043: cov[2]=agexact;
4044: if(nagesqr==1)
4045: cov[3]= agexact*agexact;
4046: for (kk=1; kk<=cptcovage;kk++) {
4047: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4048: }
4049: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4050: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4051: savm=oldm;
4052: oldm=newm;
4053: } /* end mult */
4054:
4055: s1=s[mw[mi][i]][i];
4056: s2=s[mw[mi+1][i]][i];
4057: bbh=(double)bh[mi][i]/(double)stepm;
4058: 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 */
4059: ipmx +=1;
4060: sw += weight[i];
4061: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4062: } /* end of wave */
4063: } /* end of individual */
4064: } else if(mle==3){ /* exponential inter-extrapolation */
4065: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4066: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4067: for(mi=1; mi<= wav[i]-1; mi++){
4068: for (ii=1;ii<=nlstate+ndeath;ii++)
4069: for (j=1;j<=nlstate+ndeath;j++){
4070: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4071: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4072: }
4073: for(d=0; d<dh[mi][i]; d++){
4074: newm=savm;
4075: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4076: cov[2]=agexact;
4077: if(nagesqr==1)
4078: cov[3]= agexact*agexact;
4079: for (kk=1; kk<=cptcovage;kk++) {
4080: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4081: }
4082: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4083: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4084: savm=oldm;
4085: oldm=newm;
4086: } /* end mult */
4087:
4088: s1=s[mw[mi][i]][i];
4089: s2=s[mw[mi+1][i]][i];
4090: bbh=(double)bh[mi][i]/(double)stepm;
4091: 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 */
4092: ipmx +=1;
4093: sw += weight[i];
4094: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4095: } /* end of wave */
4096: } /* end of individual */
4097: }else if (mle==4){ /* ml=4 no inter-extrapolation */
4098: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4099: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4100: for(mi=1; mi<= wav[i]-1; mi++){
4101: for (ii=1;ii<=nlstate+ndeath;ii++)
4102: for (j=1;j<=nlstate+ndeath;j++){
4103: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4104: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4105: }
4106: for(d=0; d<dh[mi][i]; d++){
4107: newm=savm;
4108: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4109: cov[2]=agexact;
4110: if(nagesqr==1)
4111: cov[3]= agexact*agexact;
4112: for (kk=1; kk<=cptcovage;kk++) {
4113: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4114: }
1.126 brouard 4115:
1.226 brouard 4116: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4117: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4118: savm=oldm;
4119: oldm=newm;
4120: } /* end mult */
4121:
4122: s1=s[mw[mi][i]][i];
4123: s2=s[mw[mi+1][i]][i];
4124: if( s2 > nlstate){
4125: lli=log(out[s1][s2] - savm[s1][s2]);
4126: } else if ( s2==-1 ) { /* alive */
4127: for (j=1,survp=0. ; j<=nlstate; j++)
4128: survp += out[s1][j];
4129: lli= log(survp);
4130: }else{
4131: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4132: }
4133: ipmx +=1;
4134: sw += weight[i];
4135: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 4136: /* 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 4137: } /* end of wave */
4138: } /* end of individual */
4139: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4140: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4141: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4142: for(mi=1; mi<= wav[i]-1; mi++){
4143: for (ii=1;ii<=nlstate+ndeath;ii++)
4144: for (j=1;j<=nlstate+ndeath;j++){
4145: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4146: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4147: }
4148: for(d=0; d<dh[mi][i]; d++){
4149: newm=savm;
4150: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4151: cov[2]=agexact;
4152: if(nagesqr==1)
4153: cov[3]= agexact*agexact;
4154: for (kk=1; kk<=cptcovage;kk++) {
4155: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4156: }
1.126 brouard 4157:
1.226 brouard 4158: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4159: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4160: savm=oldm;
4161: oldm=newm;
4162: } /* end mult */
4163:
4164: s1=s[mw[mi][i]][i];
4165: s2=s[mw[mi+1][i]][i];
4166: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4167: ipmx +=1;
4168: sw += weight[i];
4169: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4170: /*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]);*/
4171: } /* end of wave */
4172: } /* end of individual */
4173: } /* End of if */
4174: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4175: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4176: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4177: return -l;
1.126 brouard 4178: }
4179:
4180: /*************** log-likelihood *************/
4181: double funcone( double *x)
4182: {
1.228 brouard 4183: /* Same as func but slower because of a lot of printf and if */
1.335 brouard 4184: int i, ii, j, k, mi, d, kk, kf=0;
1.228 brouard 4185: int ioffset=0;
1.131 brouard 4186: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4187: double **out;
4188: double lli; /* Individual log likelihood */
4189: double llt;
4190: int s1, s2;
1.228 brouard 4191: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4192:
1.126 brouard 4193: double bbh, survp;
1.187 brouard 4194: double agexact;
1.214 brouard 4195: double agebegin, ageend;
1.126 brouard 4196: /*extern weight */
4197: /* We are differentiating ll according to initial status */
4198: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4199: /*for(i=1;i<imx;i++)
4200: printf(" %d\n",s[4][i]);
4201: */
4202: cov[1]=1.;
4203:
4204: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4205: ioffset=0;
4206: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.336 ! brouard 4207: /* Computes the values of the ncovmodel covariates of the model
! 4208: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
! 4209: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
! 4210: to be observed in j being in i according to the model.
! 4211: */
1.243 brouard 4212: /* ioffset=2+nagesqr+cptcovage; */
4213: ioffset=2+nagesqr;
1.232 brouard 4214: /* Fixed */
1.224 brouard 4215: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4216: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.335 brouard 4217: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
4218: 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 4219: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4220: /* cov[2+6]=covar[Tvar[6]][i]; */
4221: /* cov[2+6]=covar[2][i]; V2 */
4222: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4223: /* cov[2+7]=covar[Tvar[7]][i]; */
4224: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4225: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4226: /* cov[2+9]=covar[Tvar[9]][i]; */
4227: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4228: }
1.336 ! brouard 4229: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
! 4230: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
! 4231: has been calculated etc */
! 4232: /* For an individual i, wav[i] gives the number of effective waves */
! 4233: /* We compute the contribution to Likelihood of each effective transition
! 4234: mw[mi][i] is real wave of the mi th effectve wave */
! 4235: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
! 4236: s2=s[mw[mi+1][i]][i];
! 4237: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
! 4238: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
! 4239: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
! 4240: */
! 4241: /* This part may be useless now because everythin should be in covar */
1.232 brouard 4242: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4243: /* 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?)*\/ */
4244: /* } */
1.231 brouard 4245: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4246: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4247: /* } */
1.225 brouard 4248:
1.233 brouard 4249:
4250: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 4251: /* Wave varying (but not age varying) */
4252: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 4253: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
4254: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4255: }
1.232 brouard 4256: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 4257: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4258: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4259: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4260: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4261: /* 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 4262: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4263: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4264: /* /\* 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]); *\/ */
4265: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4266: /* } */
1.126 brouard 4267: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4268: for (j=1;j<=nlstate+ndeath;j++){
4269: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4270: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4271: }
1.214 brouard 4272:
4273: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4274: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4275: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4276: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4277: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4278: and mw[mi+1][i]. dh depends on stepm.*/
4279: newm=savm;
1.247 brouard 4280: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4281: cov[2]=agexact;
4282: if(nagesqr==1)
4283: cov[3]= agexact*agexact;
4284: for (kk=1; kk<=cptcovage;kk++) {
4285: if(!FixedV[Tvar[Tage[kk]]])
4286: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4287: else
4288: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
4289: }
4290: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4291: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4292: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4293: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4294: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4295: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4296: savm=oldm;
4297: oldm=newm;
1.126 brouard 4298: } /* end mult */
1.336 ! brouard 4299: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
! 4300: /* But now since version 0.9 we anticipate for bias at large stepm.
! 4301: * If stepm is larger than one month (smallest stepm) and if the exact delay
! 4302: * (in months) between two waves is not a multiple of stepm, we rounded to
! 4303: * the nearest (and in case of equal distance, to the lowest) interval but now
! 4304: * we keep into memory the bias bh[mi][i] and also the previous matrix product
! 4305: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
! 4306: * probability in order to take into account the bias as a fraction of the way
! 4307: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
! 4308: * -stepm/2 to stepm/2 .
! 4309: * For stepm=1 the results are the same as for previous versions of Imach.
! 4310: * For stepm > 1 the results are less biased than in previous versions.
! 4311: */
1.126 brouard 4312: s1=s[mw[mi][i]][i];
4313: s2=s[mw[mi+1][i]][i];
1.217 brouard 4314: /* if(s2==-1){ */
1.268 brouard 4315: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4316: /* /\* exit(1); *\/ */
4317: /* } */
1.126 brouard 4318: bbh=(double)bh[mi][i]/(double)stepm;
4319: /* bias is positive if real duration
4320: * is higher than the multiple of stepm and negative otherwise.
4321: */
4322: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4323: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4324: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4325: for (j=1,survp=0. ; j<=nlstate; j++)
4326: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4327: lli= log(survp);
1.126 brouard 4328: }else if (mle==1){
1.242 brouard 4329: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4330: } else if(mle==2){
1.242 brouard 4331: 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 4332: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4333: 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 4334: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4335: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4336: } else{ /* mle=0 back to 1 */
1.242 brouard 4337: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4338: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4339: } /* End of if */
4340: ipmx +=1;
4341: sw += weight[i];
4342: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.335 brouard 4343: /* 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 4344: if(globpr){
1.246 brouard 4345: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4346: %11.6f %11.6f %11.6f ", \
1.242 brouard 4347: 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 4348: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.335 brouard 4349: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
4350: /* %11.6f %11.6f %11.6f ", \ */
4351: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
4352: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 4353: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4354: llt +=ll[k]*gipmx/gsw;
4355: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 brouard 4356: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 4357: }
4358: fprintf(ficresilk," %10.6f\n", -llt);
1.335 brouard 4359: /* printf(" %10.6f\n", -llt); */
1.126 brouard 4360: }
1.335 brouard 4361: } /* end of wave */
4362: } /* end of individual */
4363: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 4364: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 brouard 4365: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4366: if(globpr==0){ /* First time we count the contributions and weights */
4367: gipmx=ipmx;
4368: gsw=sw;
4369: }
1.232 brouard 4370: return -l;
1.126 brouard 4371: }
4372:
4373:
4374: /*************** function likelione ***********/
1.292 brouard 4375: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4376: {
4377: /* This routine should help understanding what is done with
4378: the selection of individuals/waves and
4379: to check the exact contribution to the likelihood.
4380: Plotting could be done.
4381: */
4382: int k;
4383:
4384: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4385: strcpy(fileresilk,"ILK_");
1.202 brouard 4386: strcat(fileresilk,fileresu);
1.126 brouard 4387: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4388: printf("Problem with resultfile: %s\n", fileresilk);
4389: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4390: }
1.214 brouard 4391: 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");
4392: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4393: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4394: for(k=1; k<=nlstate; k++)
4395: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4396: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4397: }
4398:
1.292 brouard 4399: *fretone=(*func)(p);
1.126 brouard 4400: if(*globpri !=0){
4401: fclose(ficresilk);
1.205 brouard 4402: if (mle ==0)
4403: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4404: else if(mle >=1)
4405: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4406: 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 4407: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4408:
4409: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4410: 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 4411: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4412: }
1.207 brouard 4413: 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 4414: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4415: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4416: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4417: fflush(fichtm);
1.205 brouard 4418: }
1.126 brouard 4419: return;
4420: }
4421:
4422:
4423: /*********** Maximum Likelihood Estimation ***************/
4424:
4425: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4426: {
1.319 brouard 4427: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4428: double **xi;
4429: double fret;
4430: double fretone; /* Only one call to likelihood */
4431: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4432:
4433: #ifdef NLOPT
4434: int creturn;
4435: nlopt_opt opt;
4436: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4437: double *lb;
4438: double minf; /* the minimum objective value, upon return */
4439: double * p1; /* Shifted parameters from 0 instead of 1 */
4440: myfunc_data dinst, *d = &dinst;
4441: #endif
4442:
4443:
1.126 brouard 4444: xi=matrix(1,npar,1,npar);
4445: for (i=1;i<=npar;i++)
4446: for (j=1;j<=npar;j++)
4447: xi[i][j]=(i==j ? 1.0 : 0.0);
4448: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4449: strcpy(filerespow,"POW_");
1.126 brouard 4450: strcat(filerespow,fileres);
4451: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4452: printf("Problem with resultfile: %s\n", filerespow);
4453: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4454: }
4455: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4456: for (i=1;i<=nlstate;i++)
4457: for(j=1;j<=nlstate+ndeath;j++)
4458: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4459: fprintf(ficrespow,"\n");
1.162 brouard 4460: #ifdef POWELL
1.319 brouard 4461: #ifdef LINMINORIGINAL
4462: #else /* LINMINORIGINAL */
4463:
4464: flatdir=ivector(1,npar);
4465: for (j=1;j<=npar;j++) flatdir[j]=0;
4466: #endif /*LINMINORIGINAL */
4467:
4468: #ifdef FLATSUP
4469: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4470: /* reorganizing p by suppressing flat directions */
4471: for(i=1, jk=1; i <=nlstate; i++){
4472: for(k=1; k <=(nlstate+ndeath); k++){
4473: if (k != i) {
4474: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4475: if(flatdir[jk]==1){
4476: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4477: }
4478: for(j=1; j <=ncovmodel; j++){
4479: printf("%12.7f ",p[jk]);
4480: jk++;
4481: }
4482: printf("\n");
4483: }
4484: }
4485: }
4486: /* skipping */
4487: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4488: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4489: for(k=1; k <=(nlstate+ndeath); k++){
4490: if (k != i) {
4491: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4492: if(flatdir[jk]==1){
4493: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4494: for(j=1; j <=ncovmodel; jk++,j++){
4495: printf(" p[%d]=%12.7f",jk, p[jk]);
4496: /*q[jjk]=p[jk];*/
4497: }
4498: }else{
4499: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4500: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4501: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4502: /*q[jjk]=p[jk];*/
4503: }
4504: }
4505: printf("\n");
4506: }
4507: fflush(stdout);
4508: }
4509: }
4510: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4511: #else /* FLATSUP */
1.126 brouard 4512: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4513: #endif /* FLATSUP */
4514:
4515: #ifdef LINMINORIGINAL
4516: #else
4517: free_ivector(flatdir,1,npar);
4518: #endif /* LINMINORIGINAL*/
4519: #endif /* POWELL */
1.126 brouard 4520:
1.162 brouard 4521: #ifdef NLOPT
4522: #ifdef NEWUOA
4523: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4524: #else
4525: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4526: #endif
4527: lb=vector(0,npar-1);
4528: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4529: nlopt_set_lower_bounds(opt, lb);
4530: nlopt_set_initial_step1(opt, 0.1);
4531:
4532: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4533: d->function = func;
4534: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4535: nlopt_set_min_objective(opt, myfunc, d);
4536: nlopt_set_xtol_rel(opt, ftol);
4537: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4538: printf("nlopt failed! %d\n",creturn);
4539: }
4540: else {
4541: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4542: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4543: iter=1; /* not equal */
4544: }
4545: nlopt_destroy(opt);
4546: #endif
1.319 brouard 4547: #ifdef FLATSUP
4548: /* npared = npar -flatd/ncovmodel; */
4549: /* xired= matrix(1,npared,1,npared); */
4550: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4551: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4552: /* free_matrix(xire,1,npared,1,npared); */
4553: #else /* FLATSUP */
4554: #endif /* FLATSUP */
1.126 brouard 4555: free_matrix(xi,1,npar,1,npar);
4556: fclose(ficrespow);
1.203 brouard 4557: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4558: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4559: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4560:
4561: }
4562:
4563: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4564: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4565: {
4566: double **a,**y,*x,pd;
1.203 brouard 4567: /* double **hess; */
1.164 brouard 4568: int i, j;
1.126 brouard 4569: int *indx;
4570:
4571: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4572: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4573: void lubksb(double **a, int npar, int *indx, double b[]) ;
4574: void ludcmp(double **a, int npar, int *indx, double *d) ;
4575: double gompertz(double p[]);
1.203 brouard 4576: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4577:
4578: printf("\nCalculation of the hessian matrix. Wait...\n");
4579: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4580: for (i=1;i<=npar;i++){
1.203 brouard 4581: printf("%d-",i);fflush(stdout);
4582: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4583:
4584: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4585:
4586: /* printf(" %f ",p[i]);
4587: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4588: }
4589:
4590: for (i=1;i<=npar;i++) {
4591: for (j=1;j<=npar;j++) {
4592: if (j>i) {
1.203 brouard 4593: printf(".%d-%d",i,j);fflush(stdout);
4594: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4595: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4596:
4597: hess[j][i]=hess[i][j];
4598: /*printf(" %lf ",hess[i][j]);*/
4599: }
4600: }
4601: }
4602: printf("\n");
4603: fprintf(ficlog,"\n");
4604:
4605: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4606: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4607:
4608: a=matrix(1,npar,1,npar);
4609: y=matrix(1,npar,1,npar);
4610: x=vector(1,npar);
4611: indx=ivector(1,npar);
4612: for (i=1;i<=npar;i++)
4613: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4614: ludcmp(a,npar,indx,&pd);
4615:
4616: for (j=1;j<=npar;j++) {
4617: for (i=1;i<=npar;i++) x[i]=0;
4618: x[j]=1;
4619: lubksb(a,npar,indx,x);
4620: for (i=1;i<=npar;i++){
4621: matcov[i][j]=x[i];
4622: }
4623: }
4624:
4625: printf("\n#Hessian matrix#\n");
4626: fprintf(ficlog,"\n#Hessian matrix#\n");
4627: for (i=1;i<=npar;i++) {
4628: for (j=1;j<=npar;j++) {
1.203 brouard 4629: printf("%.6e ",hess[i][j]);
4630: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4631: }
4632: printf("\n");
4633: fprintf(ficlog,"\n");
4634: }
4635:
1.203 brouard 4636: /* printf("\n#Covariance matrix#\n"); */
4637: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4638: /* for (i=1;i<=npar;i++) { */
4639: /* for (j=1;j<=npar;j++) { */
4640: /* printf("%.6e ",matcov[i][j]); */
4641: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4642: /* } */
4643: /* printf("\n"); */
4644: /* fprintf(ficlog,"\n"); */
4645: /* } */
4646:
1.126 brouard 4647: /* Recompute Inverse */
1.203 brouard 4648: /* for (i=1;i<=npar;i++) */
4649: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4650: /* ludcmp(a,npar,indx,&pd); */
4651:
4652: /* printf("\n#Hessian matrix recomputed#\n"); */
4653:
4654: /* for (j=1;j<=npar;j++) { */
4655: /* for (i=1;i<=npar;i++) x[i]=0; */
4656: /* x[j]=1; */
4657: /* lubksb(a,npar,indx,x); */
4658: /* for (i=1;i<=npar;i++){ */
4659: /* y[i][j]=x[i]; */
4660: /* printf("%.3e ",y[i][j]); */
4661: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4662: /* } */
4663: /* printf("\n"); */
4664: /* fprintf(ficlog,"\n"); */
4665: /* } */
4666:
4667: /* Verifying the inverse matrix */
4668: #ifdef DEBUGHESS
4669: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4670:
1.203 brouard 4671: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4672: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4673:
4674: for (j=1;j<=npar;j++) {
4675: for (i=1;i<=npar;i++){
1.203 brouard 4676: printf("%.2f ",y[i][j]);
4677: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4678: }
4679: printf("\n");
4680: fprintf(ficlog,"\n");
4681: }
1.203 brouard 4682: #endif
1.126 brouard 4683:
4684: free_matrix(a,1,npar,1,npar);
4685: free_matrix(y,1,npar,1,npar);
4686: free_vector(x,1,npar);
4687: free_ivector(indx,1,npar);
1.203 brouard 4688: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4689:
4690:
4691: }
4692:
4693: /*************** hessian matrix ****************/
4694: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4695: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4696: int i;
4697: int l=1, lmax=20;
1.203 brouard 4698: double k1,k2, res, fx;
1.132 brouard 4699: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4700: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4701: int k=0,kmax=10;
4702: double l1;
4703:
4704: fx=func(x);
4705: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4706: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4707: l1=pow(10,l);
4708: delts=delt;
4709: for(k=1 ; k <kmax; k=k+1){
4710: delt = delta*(l1*k);
4711: p2[theta]=x[theta] +delt;
1.145 brouard 4712: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4713: p2[theta]=x[theta]-delt;
4714: k2=func(p2)-fx;
4715: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4716: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4717:
1.203 brouard 4718: #ifdef DEBUGHESSII
1.126 brouard 4719: 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);
4720: 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);
4721: #endif
4722: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4723: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4724: k=kmax;
4725: }
4726: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4727: k=kmax; l=lmax*10;
1.126 brouard 4728: }
4729: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4730: delts=delt;
4731: }
1.203 brouard 4732: } /* End loop k */
1.126 brouard 4733: }
4734: delti[theta]=delts;
4735: return res;
4736:
4737: }
4738:
1.203 brouard 4739: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4740: {
4741: int i;
1.164 brouard 4742: int l=1, lmax=20;
1.126 brouard 4743: double k1,k2,k3,k4,res,fx;
1.132 brouard 4744: double p2[MAXPARM+1];
1.203 brouard 4745: int k, kmax=1;
4746: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4747:
4748: int firstime=0;
1.203 brouard 4749:
1.126 brouard 4750: fx=func(x);
1.203 brouard 4751: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4752: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4753: p2[thetai]=x[thetai]+delti[thetai]*k;
4754: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4755: k1=func(p2)-fx;
4756:
1.203 brouard 4757: p2[thetai]=x[thetai]+delti[thetai]*k;
4758: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4759: k2=func(p2)-fx;
4760:
1.203 brouard 4761: p2[thetai]=x[thetai]-delti[thetai]*k;
4762: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4763: k3=func(p2)-fx;
4764:
1.203 brouard 4765: p2[thetai]=x[thetai]-delti[thetai]*k;
4766: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4767: k4=func(p2)-fx;
1.203 brouard 4768: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4769: if(k1*k2*k3*k4 <0.){
1.208 brouard 4770: firstime=1;
1.203 brouard 4771: kmax=kmax+10;
1.208 brouard 4772: }
4773: if(kmax >=10 || firstime ==1){
1.246 brouard 4774: 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);
4775: 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 4776: 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);
4777: 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);
4778: }
4779: #ifdef DEBUGHESSIJ
4780: v1=hess[thetai][thetai];
4781: v2=hess[thetaj][thetaj];
4782: cv12=res;
4783: /* Computing eigen value of Hessian matrix */
4784: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4785: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4786: if ((lc2 <0) || (lc1 <0) ){
4787: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4788: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4789: 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);
4790: 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);
4791: }
1.126 brouard 4792: #endif
4793: }
4794: return res;
4795: }
4796:
1.203 brouard 4797: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4798: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4799: /* { */
4800: /* int i; */
4801: /* int l=1, lmax=20; */
4802: /* double k1,k2,k3,k4,res,fx; */
4803: /* double p2[MAXPARM+1]; */
4804: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4805: /* int k=0,kmax=10; */
4806: /* double l1; */
4807:
4808: /* fx=func(x); */
4809: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4810: /* l1=pow(10,l); */
4811: /* delts=delt; */
4812: /* for(k=1 ; k <kmax; k=k+1){ */
4813: /* delt = delti*(l1*k); */
4814: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4815: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4816: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4817: /* k1=func(p2)-fx; */
4818:
4819: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4820: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4821: /* k2=func(p2)-fx; */
4822:
4823: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4824: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4825: /* k3=func(p2)-fx; */
4826:
4827: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4828: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4829: /* k4=func(p2)-fx; */
4830: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4831: /* #ifdef DEBUGHESSIJ */
4832: /* 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); */
4833: /* 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); */
4834: /* #endif */
4835: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4836: /* k=kmax; */
4837: /* } */
4838: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4839: /* k=kmax; l=lmax*10; */
4840: /* } */
4841: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4842: /* delts=delt; */
4843: /* } */
4844: /* } /\* End loop k *\/ */
4845: /* } */
4846: /* delti[theta]=delts; */
4847: /* return res; */
4848: /* } */
4849:
4850:
1.126 brouard 4851: /************** Inverse of matrix **************/
4852: void ludcmp(double **a, int n, int *indx, double *d)
4853: {
4854: int i,imax,j,k;
4855: double big,dum,sum,temp;
4856: double *vv;
4857:
4858: vv=vector(1,n);
4859: *d=1.0;
4860: for (i=1;i<=n;i++) {
4861: big=0.0;
4862: for (j=1;j<=n;j++)
4863: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4864: if (big == 0.0){
4865: printf(" Singular Hessian matrix at row %d:\n",i);
4866: for (j=1;j<=n;j++) {
4867: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4868: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4869: }
4870: fflush(ficlog);
4871: fclose(ficlog);
4872: nrerror("Singular matrix in routine ludcmp");
4873: }
1.126 brouard 4874: vv[i]=1.0/big;
4875: }
4876: for (j=1;j<=n;j++) {
4877: for (i=1;i<j;i++) {
4878: sum=a[i][j];
4879: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4880: a[i][j]=sum;
4881: }
4882: big=0.0;
4883: for (i=j;i<=n;i++) {
4884: sum=a[i][j];
4885: for (k=1;k<j;k++)
4886: sum -= a[i][k]*a[k][j];
4887: a[i][j]=sum;
4888: if ( (dum=vv[i]*fabs(sum)) >= big) {
4889: big=dum;
4890: imax=i;
4891: }
4892: }
4893: if (j != imax) {
4894: for (k=1;k<=n;k++) {
4895: dum=a[imax][k];
4896: a[imax][k]=a[j][k];
4897: a[j][k]=dum;
4898: }
4899: *d = -(*d);
4900: vv[imax]=vv[j];
4901: }
4902: indx[j]=imax;
4903: if (a[j][j] == 0.0) a[j][j]=TINY;
4904: if (j != n) {
4905: dum=1.0/(a[j][j]);
4906: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4907: }
4908: }
4909: free_vector(vv,1,n); /* Doesn't work */
4910: ;
4911: }
4912:
4913: void lubksb(double **a, int n, int *indx, double b[])
4914: {
4915: int i,ii=0,ip,j;
4916: double sum;
4917:
4918: for (i=1;i<=n;i++) {
4919: ip=indx[i];
4920: sum=b[ip];
4921: b[ip]=b[i];
4922: if (ii)
4923: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4924: else if (sum) ii=i;
4925: b[i]=sum;
4926: }
4927: for (i=n;i>=1;i--) {
4928: sum=b[i];
4929: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4930: b[i]=sum/a[i][i];
4931: }
4932: }
4933:
4934: void pstamp(FILE *fichier)
4935: {
1.196 brouard 4936: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4937: }
4938:
1.297 brouard 4939: void date2dmy(double date,double *day, double *month, double *year){
4940: double yp=0., yp1=0., yp2=0.;
4941:
4942: yp1=modf(date,&yp);/* extracts integral of date in yp and
4943: fractional in yp1 */
4944: *year=yp;
4945: yp2=modf((yp1*12),&yp);
4946: *month=yp;
4947: yp1=modf((yp2*30.5),&yp);
4948: *day=yp;
4949: if(*day==0) *day=1;
4950: if(*month==0) *month=1;
4951: }
4952:
1.253 brouard 4953:
4954:
1.126 brouard 4955: /************ Frequencies ********************/
1.251 brouard 4956: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4957: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4958: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4959: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 4960: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 4961: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4962: int iind=0, iage=0;
4963: int mi; /* Effective wave */
4964: int first;
4965: double ***freq; /* Frequencies */
1.268 brouard 4966: 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 */
4967: 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 4968: double *meanq, *stdq, *idq;
1.226 brouard 4969: double **meanqt;
4970: double *pp, **prop, *posprop, *pospropt;
4971: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4972: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4973: double agebegin, ageend;
4974:
4975: pp=vector(1,nlstate);
1.251 brouard 4976: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4977: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4978: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4979: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4980: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4981: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4982: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4983: meanqt=matrix(1,lastpass,1,nqtveff);
4984: strcpy(fileresp,"P_");
4985: strcat(fileresp,fileresu);
4986: /*strcat(fileresphtm,fileresu);*/
4987: if((ficresp=fopen(fileresp,"w"))==NULL) {
4988: printf("Problem with prevalence resultfile: %s\n", fileresp);
4989: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4990: exit(0);
4991: }
1.240 brouard 4992:
1.226 brouard 4993: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4994: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4995: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4996: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4997: fflush(ficlog);
4998: exit(70);
4999: }
5000: else{
5001: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 5002: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5003: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5004: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5005: }
1.319 brouard 5006: 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 5007:
1.226 brouard 5008: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
5009: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
5010: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5011: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5012: fflush(ficlog);
5013: exit(70);
1.240 brouard 5014: } else{
1.226 brouard 5015: 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 5016: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5017: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5018: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5019: }
1.319 brouard 5020: 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 5021:
1.253 brouard 5022: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
5023: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 5024: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5025: j1=0;
1.126 brouard 5026:
1.227 brouard 5027: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 brouard 5028: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 5029: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 5030: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 5031:
5032:
1.226 brouard 5033: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
5034: reference=low_education V1=0,V2=0
5035: med_educ V1=1 V2=0,
5036: high_educ V1=0 V2=1
1.330 brouard 5037: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 5038: */
1.249 brouard 5039: dateintsum=0;
5040: k2cpt=0;
5041:
1.253 brouard 5042: if(cptcoveff == 0 )
1.265 brouard 5043: nl=1; /* Constant and age model only */
1.253 brouard 5044: else
5045: nl=2;
1.265 brouard 5046:
5047: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
5048: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 brouard 5049: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 5050: * freq[s1][s2][iage] =0.
5051: * Loop on iind
5052: * ++freq[s1][s2][iage] weighted
5053: * end iind
5054: * if covariate and j!0
5055: * headers Variable on one line
5056: * endif cov j!=0
5057: * header of frequency table by age
5058: * Loop on age
5059: * pp[s1]+=freq[s1][s2][iage] weighted
5060: * pos+=freq[s1][s2][iage] weighted
5061: * Loop on s1 initial state
5062: * fprintf(ficresp
5063: * end s1
5064: * end age
5065: * if j!=0 computes starting values
5066: * end compute starting values
5067: * end j1
5068: * end nl
5069: */
1.253 brouard 5070: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
5071: if(nj==1)
5072: j=0; /* First pass for the constant */
1.265 brouard 5073: else{
1.335 brouard 5074: 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 5075: }
1.251 brouard 5076: first=1;
1.332 brouard 5077: 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 5078: posproptt=0.;
1.330 brouard 5079: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 5080: scanf("%d", i);*/
5081: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 5082: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 5083: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 5084: freq[i][s2][m]=0;
1.251 brouard 5085:
5086: for (i=1; i<=nlstate; i++) {
1.240 brouard 5087: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 5088: prop[i][m]=0;
5089: posprop[i]=0;
5090: pospropt[i]=0;
5091: }
1.283 brouard 5092: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 5093: idq[z1]=0.;
5094: meanq[z1]=0.;
5095: stdq[z1]=0.;
1.283 brouard 5096: }
5097: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 5098: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 5099: /* meanqt[m][z1]=0.; */
5100: /* } */
5101: /* } */
1.251 brouard 5102: /* dateintsum=0; */
5103: /* k2cpt=0; */
5104:
1.265 brouard 5105: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 5106: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
5107: bool=1;
5108: if(j !=0){
5109: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 brouard 5110: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
5111: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 5112: /* if(Tvaraff[z1] ==-20){ */
5113: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
5114: /* }else if(Tvaraff[z1] ==-10){ */
5115: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 5116: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 brouard 5117: /* 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); */
5118: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
5119: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 5120: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 5121: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 5122: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 5123: /* 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", */
5124: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
5125: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 5126: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
5127: } /* Onlyf fixed */
5128: } /* end z1 */
1.335 brouard 5129: } /* cptcoveff > 0 */
1.251 brouard 5130: } /* end any */
5131: }/* end j==0 */
1.265 brouard 5132: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 5133: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 5134: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 5135: m=mw[mi][iind];
5136: if(j!=0){
5137: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 brouard 5138: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 5139: if( Fixed[Tmodelind[z1]]==1){
5140: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.332 brouard 5141: 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 5142: value is -1, we don't select. It differs from the
5143: constant and age model which counts them. */
5144: bool=0; /* not selected */
5145: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 5146: /* i1=Tvaraff[z1]; */
5147: /* i2=TnsdVar[i1]; */
5148: /* i3=nbcode[i1][i2]; */
5149: /* i4=covar[i1][iind]; */
5150: /* if(i4 != i3){ */
5151: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 5152: bool=0;
5153: }
5154: }
5155: }
5156: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
5157: } /* end j==0 */
5158: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 5159: if(bool==1){ /*Selected */
1.251 brouard 5160: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
5161: and mw[mi+1][iind]. dh depends on stepm. */
5162: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
5163: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
5164: if(m >=firstpass && m <=lastpass){
5165: k2=anint[m][iind]+(mint[m][iind]/12.);
5166: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
5167: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
5168: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
5169: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
5170: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
5171: if (m<lastpass) {
5172: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
5173: /* 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]); */
5174: if(s[m][iind]==-1)
5175: 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.));
5176: 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 5177: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
5178: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 5179: idq[z1]=idq[z1]+weight[iind];
5180: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
5181: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
5182: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 5183: }
1.284 brouard 5184: }
1.251 brouard 5185: /* if((int)agev[m][iind] == 55) */
5186: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5187: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5188: 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 5189: }
1.251 brouard 5190: } /* end if between passes */
5191: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5192: dateintsum=dateintsum+k2; /* on all covariates ?*/
5193: k2cpt++;
5194: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5195: }
1.251 brouard 5196: }else{
5197: bool=1;
5198: }/* end bool 2 */
5199: } /* end m */
1.284 brouard 5200: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5201: /* idq[z1]=idq[z1]+weight[iind]; */
5202: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5203: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5204: /* } */
1.251 brouard 5205: } /* end bool */
5206: } /* end iind = 1 to imx */
1.319 brouard 5207: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5208: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5209:
5210:
5211: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 brouard 5212: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5213: pstamp(ficresp);
1.335 brouard 5214: if (cptcoveff>0 && j!=0){
1.265 brouard 5215: pstamp(ficresp);
1.251 brouard 5216: printf( "\n#********** Variable ");
5217: fprintf(ficresp, "\n#********** Variable ");
5218: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5219: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5220: fprintf(ficlog, "\n#********** Variable ");
1.330 brouard 5221: for (z1=1; z1<=cptcovs; z1++){
1.251 brouard 5222: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 5223: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5224: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5225: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5226: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5227: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5228: }else{
1.330 brouard 5229: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5230: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5231: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5232: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5233: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5234: }
5235: }
5236: printf( "**********\n#");
5237: fprintf(ficresp, "**********\n#");
5238: fprintf(ficresphtm, "**********</h3>\n");
5239: fprintf(ficresphtmfr, "**********</h3>\n");
5240: fprintf(ficlog, "**********\n");
5241: }
1.284 brouard 5242: /*
5243: Printing means of quantitative variables if any
5244: */
5245: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5246: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5247: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5248: if(weightopt==1){
5249: printf(" Weighted mean and standard deviation of");
5250: fprintf(ficlog," Weighted mean and standard deviation of");
5251: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5252: }
1.311 brouard 5253: /* mu = \frac{w x}{\sum w}
5254: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5255: */
5256: 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]));
5257: 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]));
5258: 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 5259: }
5260: /* for (z1=1; z1<= nqtveff; z1++) { */
5261: /* for(m=1;m<=lastpass;m++){ */
5262: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5263: /* } */
5264: /* } */
1.283 brouard 5265:
1.251 brouard 5266: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 brouard 5267: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5268: fprintf(ficresp, " Age");
1.335 brouard 5269: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
5270: 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]]);
5271: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5272: }
1.251 brouard 5273: for(i=1; i<=nlstate;i++) {
1.335 brouard 5274: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5275: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5276: }
1.335 brouard 5277: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5278: fprintf(ficresphtm, "\n");
5279:
5280: /* Header of frequency table by age */
5281: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5282: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5283: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5284: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5285: if(s2!=0 && m!=0)
5286: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5287: }
1.226 brouard 5288: }
1.251 brouard 5289: fprintf(ficresphtmfr, "\n");
5290:
5291: /* For each age */
5292: for(iage=iagemin; iage <= iagemax+3; iage++){
5293: fprintf(ficresphtm,"<tr>");
5294: if(iage==iagemax+1){
5295: fprintf(ficlog,"1");
5296: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5297: }else if(iage==iagemax+2){
5298: fprintf(ficlog,"0");
5299: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5300: }else if(iage==iagemax+3){
5301: fprintf(ficlog,"Total");
5302: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5303: }else{
1.240 brouard 5304: if(first==1){
1.251 brouard 5305: first=0;
5306: printf("See log file for details...\n");
5307: }
5308: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5309: fprintf(ficlog,"Age %d", iage);
5310: }
1.265 brouard 5311: for(s1=1; s1 <=nlstate ; s1++){
5312: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5313: pp[s1] += freq[s1][m][iage];
1.251 brouard 5314: }
1.265 brouard 5315: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5316: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5317: pos += freq[s1][m][iage];
5318: if(pp[s1]>=1.e-10){
1.251 brouard 5319: if(first==1){
1.265 brouard 5320: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5321: }
1.265 brouard 5322: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5323: }else{
5324: if(first==1)
1.265 brouard 5325: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5326: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5327: }
5328: }
5329:
1.265 brouard 5330: for(s1=1; s1 <=nlstate ; s1++){
5331: /* posprop[s1]=0; */
5332: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5333: pp[s1] += freq[s1][m][iage];
5334: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5335:
5336: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5337: pos += pp[s1]; /* pos is the total number of transitions until this age */
5338: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5339: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5340: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5341: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5342: }
5343:
5344: /* Writing ficresp */
1.335 brouard 5345: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5346: if( iage <= iagemax){
5347: fprintf(ficresp," %d",iage);
5348: }
5349: }else if( nj==2){
5350: if( iage <= iagemax){
5351: fprintf(ficresp," %d",iage);
1.335 brouard 5352: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 5353: }
1.240 brouard 5354: }
1.265 brouard 5355: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5356: if(pos>=1.e-5){
1.251 brouard 5357: if(first==1)
1.265 brouard 5358: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5359: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5360: }else{
5361: if(first==1)
1.265 brouard 5362: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5363: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5364: }
5365: if( iage <= iagemax){
5366: if(pos>=1.e-5){
1.335 brouard 5367: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5368: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5369: }else if( nj==2){
5370: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5371: }
5372: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5373: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5374: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5375: } else{
1.335 brouard 5376: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5377: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5378: }
1.240 brouard 5379: }
1.265 brouard 5380: pospropt[s1] +=posprop[s1];
5381: } /* end loop s1 */
1.251 brouard 5382: /* pospropt=0.; */
1.265 brouard 5383: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5384: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5385: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5386: if(first==1){
1.265 brouard 5387: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5388: }
1.265 brouard 5389: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5390: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5391: }
1.265 brouard 5392: if(s1!=0 && m!=0)
5393: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5394: }
1.265 brouard 5395: } /* end loop s1 */
1.251 brouard 5396: posproptt=0.;
1.265 brouard 5397: for(s1=1; s1 <=nlstate; s1++){
5398: posproptt += pospropt[s1];
1.251 brouard 5399: }
5400: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5401: fprintf(ficresphtm,"</tr>\n");
1.335 brouard 5402: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 5403: if(iage <= iagemax)
5404: fprintf(ficresp,"\n");
1.240 brouard 5405: }
1.251 brouard 5406: if(first==1)
5407: printf("Others in log...\n");
5408: fprintf(ficlog,"\n");
5409: } /* end loop age iage */
1.265 brouard 5410:
1.251 brouard 5411: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5412: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5413: if(posproptt < 1.e-5){
1.265 brouard 5414: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5415: }else{
1.265 brouard 5416: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5417: }
1.226 brouard 5418: }
1.251 brouard 5419: fprintf(ficresphtm,"</tr>\n");
5420: fprintf(ficresphtm,"</table>\n");
5421: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5422: if(posproptt < 1.e-5){
1.251 brouard 5423: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5424: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5425: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5426: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5427: invalidvarcomb[j1]=1;
1.226 brouard 5428: }else{
1.251 brouard 5429: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
5430: invalidvarcomb[j1]=0;
1.226 brouard 5431: }
1.251 brouard 5432: fprintf(ficresphtmfr,"</table>\n");
5433: fprintf(ficlog,"\n");
5434: if(j!=0){
5435: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5436: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5437: for(k=1; k <=(nlstate+ndeath); k++){
5438: if (k != i) {
1.265 brouard 5439: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5440: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5441: if(j1==1){ /* All dummy covariates to zero */
5442: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5443: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5444: printf("%d%d ",i,k);
5445: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5446: 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]));
5447: 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]));
5448: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5449: }
1.253 brouard 5450: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5451: for(iage=iagemin; iage <= iagemax+3; iage++){
5452: x[iage]= (double)iage;
5453: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5454: /* 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 5455: }
1.268 brouard 5456: /* Some are not finite, but linreg will ignore these ages */
5457: no=0;
1.253 brouard 5458: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5459: pstart[s1]=b;
5460: pstart[s1-1]=a;
1.252 brouard 5461: }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 */
5462: 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]);
5463: 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 5464: 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 5465: printf("%d%d ",i,k);
5466: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5467: 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 5468: }else{ /* Other cases, like quantitative fixed or varying covariates */
5469: ;
5470: }
5471: /* printf("%12.7f )", param[i][jj][k]); */
5472: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5473: s1++;
1.251 brouard 5474: } /* end jj */
5475: } /* end k!= i */
5476: } /* end k */
1.265 brouard 5477: } /* end i, s1 */
1.251 brouard 5478: } /* end j !=0 */
5479: } /* end selected combination of covariate j1 */
5480: if(j==0){ /* We can estimate starting values from the occurences in each case */
5481: printf("#Freqsummary: Starting values for the constants:\n");
5482: fprintf(ficlog,"\n");
1.265 brouard 5483: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5484: for(k=1; k <=(nlstate+ndeath); k++){
5485: if (k != i) {
5486: printf("%d%d ",i,k);
5487: fprintf(ficlog,"%d%d ",i,k);
5488: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5489: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5490: if(jj==1){ /* Age has to be done */
1.265 brouard 5491: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5492: 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]));
5493: 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 5494: }
5495: /* printf("%12.7f )", param[i][jj][k]); */
5496: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5497: s1++;
1.250 brouard 5498: }
1.251 brouard 5499: printf("\n");
5500: fprintf(ficlog,"\n");
1.250 brouard 5501: }
5502: }
1.284 brouard 5503: } /* end of state i */
1.251 brouard 5504: printf("#Freqsummary\n");
5505: fprintf(ficlog,"\n");
1.265 brouard 5506: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5507: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5508: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5509: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5510: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5511: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5512: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5513: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5514: /* } */
5515: }
1.265 brouard 5516: } /* end loop s1 */
1.251 brouard 5517:
5518: printf("\n");
5519: fprintf(ficlog,"\n");
5520: } /* end j=0 */
1.249 brouard 5521: } /* end j */
1.252 brouard 5522:
1.253 brouard 5523: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5524: for(i=1, jk=1; i <=nlstate; i++){
5525: for(j=1; j <=nlstate+ndeath; j++){
5526: if(j!=i){
5527: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5528: printf("%1d%1d",i,j);
5529: fprintf(ficparo,"%1d%1d",i,j);
5530: for(k=1; k<=ncovmodel;k++){
5531: /* printf(" %lf",param[i][j][k]); */
5532: /* fprintf(ficparo," %lf",param[i][j][k]); */
5533: p[jk]=pstart[jk];
5534: printf(" %f ",pstart[jk]);
5535: fprintf(ficparo," %f ",pstart[jk]);
5536: jk++;
5537: }
5538: printf("\n");
5539: fprintf(ficparo,"\n");
5540: }
5541: }
5542: }
5543: } /* end mle=-2 */
1.226 brouard 5544: dateintmean=dateintsum/k2cpt;
1.296 brouard 5545: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5546:
1.226 brouard 5547: fclose(ficresp);
5548: fclose(ficresphtm);
5549: fclose(ficresphtmfr);
1.283 brouard 5550: free_vector(idq,1,nqfveff);
1.226 brouard 5551: free_vector(meanq,1,nqfveff);
1.284 brouard 5552: free_vector(stdq,1,nqfveff);
1.226 brouard 5553: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5554: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5555: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5556: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5557: free_vector(pospropt,1,nlstate);
5558: free_vector(posprop,1,nlstate);
1.251 brouard 5559: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5560: free_vector(pp,1,nlstate);
5561: /* End of freqsummary */
5562: }
1.126 brouard 5563:
1.268 brouard 5564: /* Simple linear regression */
5565: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5566:
5567: /* y=a+bx regression */
5568: double sumx = 0.0; /* sum of x */
5569: double sumx2 = 0.0; /* sum of x**2 */
5570: double sumxy = 0.0; /* sum of x * y */
5571: double sumy = 0.0; /* sum of y */
5572: double sumy2 = 0.0; /* sum of y**2 */
5573: double sume2 = 0.0; /* sum of square or residuals */
5574: double yhat;
5575:
5576: double denom=0;
5577: int i;
5578: int ne=*no;
5579:
5580: for ( i=ifi, ne=0;i<=ila;i++) {
5581: if(!isfinite(x[i]) || !isfinite(y[i])){
5582: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5583: continue;
5584: }
5585: ne=ne+1;
5586: sumx += x[i];
5587: sumx2 += x[i]*x[i];
5588: sumxy += x[i] * y[i];
5589: sumy += y[i];
5590: sumy2 += y[i]*y[i];
5591: denom = (ne * sumx2 - sumx*sumx);
5592: /* 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); */
5593: }
5594:
5595: denom = (ne * sumx2 - sumx*sumx);
5596: if (denom == 0) {
5597: // vertical, slope m is infinity
5598: *b = INFINITY;
5599: *a = 0;
5600: if (r) *r = 0;
5601: return 1;
5602: }
5603:
5604: *b = (ne * sumxy - sumx * sumy) / denom;
5605: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5606: if (r!=NULL) {
5607: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5608: sqrt((sumx2 - sumx*sumx/ne) *
5609: (sumy2 - sumy*sumy/ne));
5610: }
5611: *no=ne;
5612: for ( i=ifi, ne=0;i<=ila;i++) {
5613: if(!isfinite(x[i]) || !isfinite(y[i])){
5614: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5615: continue;
5616: }
5617: ne=ne+1;
5618: yhat = y[i] - *a -*b* x[i];
5619: sume2 += yhat * yhat ;
5620:
5621: denom = (ne * sumx2 - sumx*sumx);
5622: /* 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); */
5623: }
5624: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5625: *sa= *sb * sqrt(sumx2/ne);
5626:
5627: return 0;
5628: }
5629:
1.126 brouard 5630: /************ Prevalence ********************/
1.227 brouard 5631: 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)
5632: {
5633: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5634: in each health status at the date of interview (if between dateprev1 and dateprev2).
5635: We still use firstpass and lastpass as another selection.
5636: */
1.126 brouard 5637:
1.227 brouard 5638: int i, m, jk, j1, bool, z1,j, iv;
5639: int mi; /* Effective wave */
5640: int iage;
5641: double agebegin, ageend;
5642:
5643: double **prop;
5644: double posprop;
5645: double y2; /* in fractional years */
5646: int iagemin, iagemax;
5647: int first; /** to stop verbosity which is redirected to log file */
5648:
5649: iagemin= (int) agemin;
5650: iagemax= (int) agemax;
5651: /*pp=vector(1,nlstate);*/
1.251 brouard 5652: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5653: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5654: j1=0;
1.222 brouard 5655:
1.227 brouard 5656: /*j=cptcoveff;*/
5657: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5658:
1.288 brouard 5659: first=0;
1.335 brouard 5660: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 5661: for (i=1; i<=nlstate; i++)
1.251 brouard 5662: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5663: prop[i][iage]=0.0;
5664: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5665: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5666: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5667:
5668: for (i=1; i<=imx; i++) { /* Each individual */
5669: bool=1;
5670: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5671: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5672: m=mw[mi][i];
5673: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5674: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5675: for (z1=1; z1<=cptcoveff; z1++){
5676: if( Fixed[Tmodelind[z1]]==1){
5677: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.332 brouard 5678: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 5679: bool=0;
5680: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 5681: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 5682: bool=0;
5683: }
5684: }
5685: if(bool==1){ /* Otherwise we skip that wave/person */
5686: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5687: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5688: if(m >=firstpass && m <=lastpass){
5689: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5690: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5691: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5692: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5693: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5694: 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);
5695: exit(1);
5696: }
5697: if (s[m][i]>0 && s[m][i]<=nlstate) {
5698: /*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]]);*/
5699: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5700: prop[s[m][i]][iagemax+3] += weight[i];
5701: } /* end valid statuses */
5702: } /* end selection of dates */
5703: } /* end selection of waves */
5704: } /* end bool */
5705: } /* end wave */
5706: } /* end individual */
5707: for(i=iagemin; i <= iagemax+3; i++){
5708: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5709: posprop += prop[jk][i];
5710: }
5711:
5712: for(jk=1; jk <=nlstate ; jk++){
5713: if( i <= iagemax){
5714: if(posprop>=1.e-5){
5715: probs[i][jk][j1]= prop[jk][i]/posprop;
5716: } else{
1.288 brouard 5717: if(!first){
5718: first=1;
1.266 brouard 5719: 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]);
5720: }else{
1.288 brouard 5721: 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 5722: }
5723: }
5724: }
5725: }/* end jk */
5726: }/* end i */
1.222 brouard 5727: /*} *//* end i1 */
1.227 brouard 5728: } /* end j1 */
1.222 brouard 5729:
1.227 brouard 5730: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5731: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5732: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5733: } /* End of prevalence */
1.126 brouard 5734:
5735: /************* Waves Concatenation ***************/
5736:
5737: 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)
5738: {
1.298 brouard 5739: /* 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 5740: Death is a valid wave (if date is known).
5741: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5742: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5743: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5744: */
1.126 brouard 5745:
1.224 brouard 5746: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5747: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5748: double sum=0., jmean=0.;*/
1.224 brouard 5749: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5750: int j, k=0,jk, ju, jl;
5751: double sum=0.;
5752: first=0;
1.214 brouard 5753: firstwo=0;
1.217 brouard 5754: firsthree=0;
1.218 brouard 5755: firstfour=0;
1.164 brouard 5756: jmin=100000;
1.126 brouard 5757: jmax=-1;
5758: jmean=0.;
1.224 brouard 5759:
5760: /* Treating live states */
1.214 brouard 5761: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5762: mi=0; /* First valid wave */
1.227 brouard 5763: mli=0; /* Last valid wave */
1.309 brouard 5764: m=firstpass; /* Loop on waves */
5765: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5766: 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 */
5767: mli=m-1;/* mw[++mi][i]=m-1; */
5768: }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 5769: 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 5770: mli=m;
1.224 brouard 5771: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5772: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5773: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5774: }
1.309 brouard 5775: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5776: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5777: break;
1.224 brouard 5778: #else
1.317 brouard 5779: 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 5780: if(firsthree == 0){
1.302 brouard 5781: 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 5782: firsthree=1;
1.317 brouard 5783: }else if(firsthree >=1 && firsthree < 10){
5784: 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);
5785: firsthree++;
5786: }else if(firsthree == 10){
5787: printf("Information, too many Information flags: no more reported to log either\n");
5788: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5789: firsthree++;
5790: }else{
5791: firsthree++;
1.227 brouard 5792: }
1.309 brouard 5793: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5794: mli=m;
5795: }
5796: if(s[m][i]==-2){ /* Vital status is really unknown */
5797: nbwarn++;
1.309 brouard 5798: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5799: 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);
5800: 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);
5801: }
5802: break;
5803: }
5804: break;
1.224 brouard 5805: #endif
1.227 brouard 5806: }/* End m >= lastpass */
1.126 brouard 5807: }/* end while */
1.224 brouard 5808:
1.227 brouard 5809: /* 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 5810: /* After last pass */
1.224 brouard 5811: /* Treating death states */
1.214 brouard 5812: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5813: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5814: /* } */
1.126 brouard 5815: mi++; /* Death is another wave */
5816: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5817: /* Only death is a correct wave */
1.126 brouard 5818: mw[mi][i]=m;
1.257 brouard 5819: } /* else not in a death state */
1.224 brouard 5820: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5821: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5822: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5823: 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 5824: nbwarn++;
5825: if(firstfiv==0){
1.309 brouard 5826: 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 5827: firstfiv=1;
5828: }else{
1.309 brouard 5829: 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 5830: }
1.309 brouard 5831: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5832: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5833: nberr++;
5834: if(firstwo==0){
1.309 brouard 5835: 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 5836: firstwo=1;
5837: }
1.309 brouard 5838: 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 5839: }
1.257 brouard 5840: }else{ /* if date of interview is unknown */
1.227 brouard 5841: /* death is known but not confirmed by death status at any wave */
5842: if(firstfour==0){
1.309 brouard 5843: 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 5844: firstfour=1;
5845: }
1.309 brouard 5846: 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 5847: }
1.224 brouard 5848: } /* end if date of death is known */
5849: #endif
1.309 brouard 5850: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5851: /* wav[i]=mw[mi][i]; */
1.126 brouard 5852: if(mi==0){
5853: nbwarn++;
5854: if(first==0){
1.227 brouard 5855: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5856: first=1;
1.126 brouard 5857: }
5858: if(first==1){
1.227 brouard 5859: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5860: }
5861: } /* end mi==0 */
5862: } /* End individuals */
1.214 brouard 5863: /* wav and mw are no more changed */
1.223 brouard 5864:
1.317 brouard 5865: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
5866: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
5867:
5868:
1.126 brouard 5869: for(i=1; i<=imx; i++){
5870: for(mi=1; mi<wav[i];mi++){
5871: if (stepm <=0)
1.227 brouard 5872: dh[mi][i]=1;
1.126 brouard 5873: else{
1.260 brouard 5874: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5875: if (agedc[i] < 2*AGESUP) {
5876: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5877: if(j==0) j=1; /* Survives at least one month after exam */
5878: else if(j<0){
5879: nberr++;
5880: 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]);
5881: j=1; /* Temporary Dangerous patch */
5882: 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);
5883: 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]);
5884: 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);
5885: }
5886: k=k+1;
5887: if (j >= jmax){
5888: jmax=j;
5889: ijmax=i;
5890: }
5891: if (j <= jmin){
5892: jmin=j;
5893: ijmin=i;
5894: }
5895: sum=sum+j;
5896: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5897: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5898: }
5899: }
5900: else{
5901: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5902: /* 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 5903:
1.227 brouard 5904: k=k+1;
5905: if (j >= jmax) {
5906: jmax=j;
5907: ijmax=i;
5908: }
5909: else if (j <= jmin){
5910: jmin=j;
5911: ijmin=i;
5912: }
5913: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5914: /*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]);*/
5915: if(j<0){
5916: nberr++;
5917: 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]);
5918: 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]);
5919: }
5920: sum=sum+j;
5921: }
5922: jk= j/stepm;
5923: jl= j -jk*stepm;
5924: ju= j -(jk+1)*stepm;
5925: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5926: if(jl==0){
5927: dh[mi][i]=jk;
5928: bh[mi][i]=0;
5929: }else{ /* We want a negative bias in order to only have interpolation ie
5930: * to avoid the price of an extra matrix product in likelihood */
5931: dh[mi][i]=jk+1;
5932: bh[mi][i]=ju;
5933: }
5934: }else{
5935: if(jl <= -ju){
5936: dh[mi][i]=jk;
5937: bh[mi][i]=jl; /* bias is positive if real duration
5938: * is higher than the multiple of stepm and negative otherwise.
5939: */
5940: }
5941: else{
5942: dh[mi][i]=jk+1;
5943: bh[mi][i]=ju;
5944: }
5945: if(dh[mi][i]==0){
5946: dh[mi][i]=1; /* At least one step */
5947: bh[mi][i]=ju; /* At least one step */
5948: /* 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);*/
5949: }
5950: } /* end if mle */
1.126 brouard 5951: }
5952: } /* end wave */
5953: }
5954: jmean=sum/k;
5955: 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 5956: 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 5957: }
1.126 brouard 5958:
5959: /*********** Tricode ****************************/
1.220 brouard 5960: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5961: {
5962: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5963: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5964: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5965: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5966: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5967: */
1.130 brouard 5968:
1.242 brouard 5969: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5970: int modmaxcovj=0; /* Modality max of covariates j */
5971: int cptcode=0; /* Modality max of covariates j */
5972: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5973:
5974:
1.242 brouard 5975: /* cptcoveff=0; */
5976: /* *cptcov=0; */
1.126 brouard 5977:
1.242 brouard 5978: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5979: for (k=1; k <= maxncov; k++)
5980: for(j=1; j<=2; j++)
5981: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5982:
1.242 brouard 5983: /* Loop on covariates without age and products and no quantitative variable */
1.335 brouard 5984: 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 5985: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5986: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5987: switch(Fixed[k]) {
5988: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5989: modmaxcovj=0;
5990: modmincovj=0;
1.242 brouard 5991: 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*/
5992: ij=(int)(covar[Tvar[k]][i]);
5993: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5994: * If product of Vn*Vm, still boolean *:
5995: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5996: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5997: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5998: modality of the nth covariate of individual i. */
5999: if (ij > modmaxcovj)
6000: modmaxcovj=ij;
6001: else if (ij < modmincovj)
6002: modmincovj=ij;
1.287 brouard 6003: if (ij <0 || ij >1 ){
1.311 brouard 6004: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6005: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6006: fflush(ficlog);
6007: exit(1);
1.287 brouard 6008: }
6009: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 6010: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
6011: exit(1);
6012: }else
6013: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
6014: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
6015: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
6016: /* getting the maximum value of the modality of the covariate
6017: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
6018: female ies 1, then modmaxcovj=1.
6019: */
6020: } /* end for loop on individuals i */
6021: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6022: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6023: cptcode=modmaxcovj;
6024: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
6025: /*for (i=0; i<=cptcode; i++) {*/
6026: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
6027: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6028: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6029: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
6030: if( j != -1){
6031: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
6032: covariate for which somebody answered excluding
6033: undefined. Usually 2: 0 and 1. */
6034: }
6035: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
6036: covariate for which somebody answered including
6037: undefined. Usually 3: -1, 0 and 1. */
6038: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
6039: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
6040: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 6041:
1.242 brouard 6042: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
6043: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
6044: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
6045: /* modmincovj=3; modmaxcovj = 7; */
6046: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
6047: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
6048: /* defining two dummy variables: variables V1_1 and V1_2.*/
6049: /* nbcode[Tvar[j]][ij]=k; */
6050: /* nbcode[Tvar[j]][1]=0; */
6051: /* nbcode[Tvar[j]][2]=1; */
6052: /* nbcode[Tvar[j]][3]=2; */
6053: /* To be continued (not working yet). */
6054: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 6055:
6056: /* 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*/
6057: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
6058: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
6059: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
6060: /*, could be restored in the future */
6061: 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 6062: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
6063: break;
6064: }
6065: ij++;
1.287 brouard 6066: 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 6067: cptcode = ij; /* New max modality for covar j */
6068: } /* end of loop on modality i=-1 to 1 or more */
6069: break;
6070: case 1: /* Testing on varying covariate, could be simple and
6071: * should look at waves or product of fixed *
6072: * varying. No time to test -1, assuming 0 and 1 only */
6073: ij=0;
6074: for(i=0; i<=1;i++){
6075: nbcode[Tvar[k]][++ij]=i;
6076: }
6077: break;
6078: default:
6079: break;
6080: } /* end switch */
6081: } /* end dummy test */
1.334 brouard 6082: if(Dummy[k]==1 && Typevar[k] !=1){ /* Quantitative covariate and not age product */
1.311 brouard 6083: 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 6084: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
6085: printf("Error k=%d \n",k);
6086: exit(1);
6087: }
1.311 brouard 6088: if(isnan(covar[Tvar[k]][i])){
6089: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6090: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6091: fflush(ficlog);
6092: exit(1);
6093: }
6094: }
1.335 brouard 6095: } /* end Quanti */
1.287 brouard 6096: } /* 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 6097:
6098: for (k=-1; k< maxncov; k++) Ndum[k]=0;
6099: /* Look at fixed dummy (single or product) covariates to check empty modalities */
6100: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
6101: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
6102: 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 */
6103: 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 */
6104: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
6105: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
6106:
6107: ij=0;
6108: /* 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 6109: 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 */
6110: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 6111: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
6112: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 brouard 6113: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
6114: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
6115: /* 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 6116: /* If product not in single variable we don't print results */
6117: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 brouard 6118: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
6119: /* k= 1 2 3 4 5 6 7 8 9 */
6120: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
6121: /* ij 1 2 3 */
6122: /* Tvaraff[ij]= 4 3 1 */
6123: /* Tmodelind[ij]=2 3 9 */
6124: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 6125: 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*/
6126: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
6127: 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 */
6128: if(Fixed[k]!=0)
6129: anyvaryingduminmodel=1;
6130: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
6131: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
6132: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
6133: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
6134: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
6135: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
6136: }
6137: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
6138: /* ij--; */
6139: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 brouard 6140: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 6141: * because they can be excluded from the model and real
6142: * if in the model but excluded because missing values, but how to get k from ij?*/
6143: for(j=ij+1; j<= cptcovt; j++){
6144: Tvaraff[j]=0;
6145: Tmodelind[j]=0;
6146: }
6147: for(j=ntveff+1; j<= cptcovt; j++){
6148: TmodelInvind[j]=0;
6149: }
6150: /* To be sorted */
6151: ;
6152: }
1.126 brouard 6153:
1.145 brouard 6154:
1.126 brouard 6155: /*********** Health Expectancies ****************/
6156:
1.235 brouard 6157: 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 6158:
6159: {
6160: /* Health expectancies, no variances */
1.329 brouard 6161: /* cij is the combination in the list of combination of dummy covariates */
6162: /* strstart is a string of time at start of computing */
1.164 brouard 6163: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 6164: int nhstepma, nstepma; /* Decreasing with age */
6165: double age, agelim, hf;
6166: double ***p3mat;
6167: double eip;
6168:
1.238 brouard 6169: /* pstamp(ficreseij); */
1.126 brouard 6170: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
6171: fprintf(ficreseij,"# Age");
6172: for(i=1; i<=nlstate;i++){
6173: for(j=1; j<=nlstate;j++){
6174: fprintf(ficreseij," e%1d%1d ",i,j);
6175: }
6176: fprintf(ficreseij," e%1d. ",i);
6177: }
6178: fprintf(ficreseij,"\n");
6179:
6180:
6181: if(estepm < stepm){
6182: printf ("Problem %d lower than %d\n",estepm, stepm);
6183: }
6184: else hstepm=estepm;
6185: /* We compute the life expectancy from trapezoids spaced every estepm months
6186: * This is mainly to measure the difference between two models: for example
6187: * if stepm=24 months pijx are given only every 2 years and by summing them
6188: * we are calculating an estimate of the Life Expectancy assuming a linear
6189: * progression in between and thus overestimating or underestimating according
6190: * to the curvature of the survival function. If, for the same date, we
6191: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6192: * to compare the new estimate of Life expectancy with the same linear
6193: * hypothesis. A more precise result, taking into account a more precise
6194: * curvature will be obtained if estepm is as small as stepm. */
6195:
6196: /* For example we decided to compute the life expectancy with the smallest unit */
6197: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6198: nhstepm is the number of hstepm from age to agelim
6199: nstepm is the number of stepm from age to agelin.
1.270 brouard 6200: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6201: and note for a fixed period like estepm months */
6202: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6203: survival function given by stepm (the optimization length). Unfortunately it
6204: means that if the survival funtion is printed only each two years of age and if
6205: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6206: results. So we changed our mind and took the option of the best precision.
6207: */
6208: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6209:
6210: agelim=AGESUP;
6211: /* If stepm=6 months */
6212: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6213: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6214:
6215: /* nhstepm age range expressed in number of stepm */
6216: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6217: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6218: /* if (stepm >= YEARM) hstepm=1;*/
6219: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6220: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6221:
6222: for (age=bage; age<=fage; age ++){
6223: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6224: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6225: /* if (stepm >= YEARM) hstepm=1;*/
6226: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6227:
6228: /* If stepm=6 months */
6229: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6230: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 6231: /* 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 6232: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6233:
6234: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6235:
6236: printf("%d|",(int)age);fflush(stdout);
6237: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6238:
6239: /* Computing expectancies */
6240: for(i=1; i<=nlstate;i++)
6241: for(j=1; j<=nlstate;j++)
6242: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6243: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6244:
6245: /* 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]);*/
6246:
6247: }
6248:
6249: fprintf(ficreseij,"%3.0f",age );
6250: for(i=1; i<=nlstate;i++){
6251: eip=0;
6252: for(j=1; j<=nlstate;j++){
6253: eip +=eij[i][j][(int)age];
6254: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6255: }
6256: fprintf(ficreseij,"%9.4f", eip );
6257: }
6258: fprintf(ficreseij,"\n");
6259:
6260: }
6261: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6262: printf("\n");
6263: fprintf(ficlog,"\n");
6264:
6265: }
6266:
1.235 brouard 6267: 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 6268:
6269: {
6270: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6271: to initial status i, ei. .
1.126 brouard 6272: */
1.336 ! brouard 6273: /* Very time consuming function, but already optimized with precov */
1.126 brouard 6274: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6275: int nhstepma, nstepma; /* Decreasing with age */
6276: double age, agelim, hf;
6277: double ***p3matp, ***p3matm, ***varhe;
6278: double **dnewm,**doldm;
6279: double *xp, *xm;
6280: double **gp, **gm;
6281: double ***gradg, ***trgradg;
6282: int theta;
6283:
6284: double eip, vip;
6285:
6286: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6287: xp=vector(1,npar);
6288: xm=vector(1,npar);
6289: dnewm=matrix(1,nlstate*nlstate,1,npar);
6290: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6291:
6292: pstamp(ficresstdeij);
6293: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6294: fprintf(ficresstdeij,"# Age");
6295: for(i=1; i<=nlstate;i++){
6296: for(j=1; j<=nlstate;j++)
6297: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6298: fprintf(ficresstdeij," e%1d. ",i);
6299: }
6300: fprintf(ficresstdeij,"\n");
6301:
6302: pstamp(ficrescveij);
6303: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6304: fprintf(ficrescveij,"# Age");
6305: for(i=1; i<=nlstate;i++)
6306: for(j=1; j<=nlstate;j++){
6307: cptj= (j-1)*nlstate+i;
6308: for(i2=1; i2<=nlstate;i2++)
6309: for(j2=1; j2<=nlstate;j2++){
6310: cptj2= (j2-1)*nlstate+i2;
6311: if(cptj2 <= cptj)
6312: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6313: }
6314: }
6315: fprintf(ficrescveij,"\n");
6316:
6317: if(estepm < stepm){
6318: printf ("Problem %d lower than %d\n",estepm, stepm);
6319: }
6320: else hstepm=estepm;
6321: /* We compute the life expectancy from trapezoids spaced every estepm months
6322: * This is mainly to measure the difference between two models: for example
6323: * if stepm=24 months pijx are given only every 2 years and by summing them
6324: * we are calculating an estimate of the Life Expectancy assuming a linear
6325: * progression in between and thus overestimating or underestimating according
6326: * to the curvature of the survival function. If, for the same date, we
6327: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6328: * to compare the new estimate of Life expectancy with the same linear
6329: * hypothesis. A more precise result, taking into account a more precise
6330: * curvature will be obtained if estepm is as small as stepm. */
6331:
6332: /* For example we decided to compute the life expectancy with the smallest unit */
6333: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6334: nhstepm is the number of hstepm from age to agelim
6335: nstepm is the number of stepm from age to agelin.
6336: Look at hpijx to understand the reason of that which relies in memory size
6337: and note for a fixed period like estepm months */
6338: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6339: survival function given by stepm (the optimization length). Unfortunately it
6340: means that if the survival funtion is printed only each two years of age and if
6341: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6342: results. So we changed our mind and took the option of the best precision.
6343: */
6344: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6345:
6346: /* If stepm=6 months */
6347: /* nhstepm age range expressed in number of stepm */
6348: agelim=AGESUP;
6349: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6350: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6351: /* if (stepm >= YEARM) hstepm=1;*/
6352: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6353:
6354: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6355: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6356: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6357: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6358: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6359: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6360:
6361: for (age=bage; age<=fage; age ++){
6362: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6363: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6364: /* if (stepm >= YEARM) hstepm=1;*/
6365: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6366:
1.126 brouard 6367: /* If stepm=6 months */
6368: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6369: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6370:
6371: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6372:
1.126 brouard 6373: /* Computing Variances of health expectancies */
6374: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6375: decrease memory allocation */
6376: for(theta=1; theta <=npar; theta++){
6377: for(i=1; i<=npar; i++){
1.222 brouard 6378: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6379: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6380: }
1.235 brouard 6381: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6382: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6383:
1.126 brouard 6384: for(j=1; j<= nlstate; j++){
1.222 brouard 6385: for(i=1; i<=nlstate; i++){
6386: for(h=0; h<=nhstepm-1; h++){
6387: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6388: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6389: }
6390: }
1.126 brouard 6391: }
1.218 brouard 6392:
1.126 brouard 6393: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6394: for(h=0; h<=nhstepm-1; h++){
6395: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6396: }
1.126 brouard 6397: }/* End theta */
6398:
6399:
6400: for(h=0; h<=nhstepm-1; h++)
6401: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6402: for(theta=1; theta <=npar; theta++)
6403: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6404:
1.218 brouard 6405:
1.222 brouard 6406: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6407: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6408: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6409:
1.222 brouard 6410: printf("%d|",(int)age);fflush(stdout);
6411: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6412: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6413: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6414: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6415: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6416: for(ij=1;ij<=nlstate*nlstate;ij++)
6417: for(ji=1;ji<=nlstate*nlstate;ji++)
6418: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6419: }
6420: }
1.320 brouard 6421: /* if((int)age ==50){ */
6422: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6423: /* } */
1.126 brouard 6424: /* Computing expectancies */
1.235 brouard 6425: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6426: for(i=1; i<=nlstate;i++)
6427: for(j=1; j<=nlstate;j++)
1.222 brouard 6428: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6429: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6430:
1.222 brouard 6431: /* 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 6432:
1.222 brouard 6433: }
1.269 brouard 6434:
6435: /* Standard deviation of expectancies ij */
1.126 brouard 6436: fprintf(ficresstdeij,"%3.0f",age );
6437: for(i=1; i<=nlstate;i++){
6438: eip=0.;
6439: vip=0.;
6440: for(j=1; j<=nlstate;j++){
1.222 brouard 6441: eip += eij[i][j][(int)age];
6442: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6443: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6444: 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 6445: }
6446: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6447: }
6448: fprintf(ficresstdeij,"\n");
1.218 brouard 6449:
1.269 brouard 6450: /* Variance of expectancies ij */
1.126 brouard 6451: fprintf(ficrescveij,"%3.0f",age );
6452: for(i=1; i<=nlstate;i++)
6453: for(j=1; j<=nlstate;j++){
1.222 brouard 6454: cptj= (j-1)*nlstate+i;
6455: for(i2=1; i2<=nlstate;i2++)
6456: for(j2=1; j2<=nlstate;j2++){
6457: cptj2= (j2-1)*nlstate+i2;
6458: if(cptj2 <= cptj)
6459: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6460: }
1.126 brouard 6461: }
6462: fprintf(ficrescveij,"\n");
1.218 brouard 6463:
1.126 brouard 6464: }
6465: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6466: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6467: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6468: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6469: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6470: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6471: printf("\n");
6472: fprintf(ficlog,"\n");
1.218 brouard 6473:
1.126 brouard 6474: free_vector(xm,1,npar);
6475: free_vector(xp,1,npar);
6476: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6477: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6478: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6479: }
1.218 brouard 6480:
1.126 brouard 6481: /************ Variance ******************/
1.235 brouard 6482: 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 6483: {
1.279 brouard 6484: /** Variance of health expectancies
6485: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6486: * double **newm;
6487: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6488: */
1.218 brouard 6489:
6490: /* int movingaverage(); */
6491: double **dnewm,**doldm;
6492: double **dnewmp,**doldmp;
6493: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6494: int first=0;
1.218 brouard 6495: int k;
6496: double *xp;
1.279 brouard 6497: double **gp, **gm; /**< for var eij */
6498: double ***gradg, ***trgradg; /**< for var eij */
6499: double **gradgp, **trgradgp; /**< for var p point j */
6500: double *gpp, *gmp; /**< for var p point j */
6501: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6502: double ***p3mat;
6503: double age,agelim, hf;
6504: /* double ***mobaverage; */
6505: int theta;
6506: char digit[4];
6507: char digitp[25];
6508:
6509: char fileresprobmorprev[FILENAMELENGTH];
6510:
6511: if(popbased==1){
6512: if(mobilav!=0)
6513: strcpy(digitp,"-POPULBASED-MOBILAV_");
6514: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6515: }
6516: else
6517: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6518:
1.218 brouard 6519: /* if (mobilav!=0) { */
6520: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6521: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6522: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6523: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6524: /* } */
6525: /* } */
6526:
6527: strcpy(fileresprobmorprev,"PRMORPREV-");
6528: sprintf(digit,"%-d",ij);
6529: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6530: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6531: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6532: strcat(fileresprobmorprev,fileresu);
6533: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6534: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6535: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6536: }
6537: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6538: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6539: pstamp(ficresprobmorprev);
6540: 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 6541: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.334 brouard 6542: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */ /* To be done*/
1.332 brouard 6543: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 6544: }
6545: for(j=1;j<=cptcoveff;j++)
1.334 brouard 6546: fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]);
1.238 brouard 6547: fprintf(ficresprobmorprev,"\n");
6548:
1.218 brouard 6549: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6550: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6551: fprintf(ficresprobmorprev," p.%-d SE",j);
6552: for(i=1; i<=nlstate;i++)
6553: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6554: }
6555: fprintf(ficresprobmorprev,"\n");
6556:
6557: fprintf(ficgp,"\n# Routine varevsij");
6558: fprintf(ficgp,"\nunset title \n");
6559: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6560: 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");
6561: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6562:
1.218 brouard 6563: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6564: pstamp(ficresvij);
6565: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6566: if(popbased==1)
6567: 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);
6568: else
6569: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6570: fprintf(ficresvij,"# Age");
6571: for(i=1; i<=nlstate;i++)
6572: for(j=1; j<=nlstate;j++)
6573: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6574: fprintf(ficresvij,"\n");
6575:
6576: xp=vector(1,npar);
6577: dnewm=matrix(1,nlstate,1,npar);
6578: doldm=matrix(1,nlstate,1,nlstate);
6579: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6580: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6581:
6582: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6583: gpp=vector(nlstate+1,nlstate+ndeath);
6584: gmp=vector(nlstate+1,nlstate+ndeath);
6585: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6586:
1.218 brouard 6587: if(estepm < stepm){
6588: printf ("Problem %d lower than %d\n",estepm, stepm);
6589: }
6590: else hstepm=estepm;
6591: /* For example we decided to compute the life expectancy with the smallest unit */
6592: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6593: nhstepm is the number of hstepm from age to agelim
6594: nstepm is the number of stepm from age to agelim.
6595: Look at function hpijx to understand why because of memory size limitations,
6596: we decided (b) to get a life expectancy respecting the most precise curvature of the
6597: survival function given by stepm (the optimization length). Unfortunately it
6598: means that if the survival funtion is printed every two years of age and if
6599: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6600: results. So we changed our mind and took the option of the best precision.
6601: */
6602: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6603: agelim = AGESUP;
6604: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6605: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6606: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6607: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6608: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6609: gp=matrix(0,nhstepm,1,nlstate);
6610: gm=matrix(0,nhstepm,1,nlstate);
6611:
6612:
6613: for(theta=1; theta <=npar; theta++){
6614: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6615: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6616: }
1.279 brouard 6617: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6618: * returns into prlim .
1.288 brouard 6619: */
1.242 brouard 6620: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6621:
6622: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6623: if (popbased==1) {
6624: if(mobilav ==0){
6625: for(i=1; i<=nlstate;i++)
6626: prlim[i][i]=probs[(int)age][i][ij];
6627: }else{ /* mobilav */
6628: for(i=1; i<=nlstate;i++)
6629: prlim[i][i]=mobaverage[(int)age][i][ij];
6630: }
6631: }
1.295 brouard 6632: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6633: */
6634: 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 6635: /**< 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 6636: * at horizon h in state j including mortality.
6637: */
1.218 brouard 6638: for(j=1; j<= nlstate; j++){
6639: for(h=0; h<=nhstepm; h++){
6640: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6641: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6642: }
6643: }
1.279 brouard 6644: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6645: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6646: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6647: */
6648: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6649: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6650: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6651: }
6652:
6653: /* Again with minus shift */
1.218 brouard 6654:
6655: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6656: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6657:
1.242 brouard 6658: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6659:
6660: if (popbased==1) {
6661: if(mobilav ==0){
6662: for(i=1; i<=nlstate;i++)
6663: prlim[i][i]=probs[(int)age][i][ij];
6664: }else{ /* mobilav */
6665: for(i=1; i<=nlstate;i++)
6666: prlim[i][i]=mobaverage[(int)age][i][ij];
6667: }
6668: }
6669:
1.235 brouard 6670: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6671:
6672: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6673: for(h=0; h<=nhstepm; h++){
6674: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6675: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6676: }
6677: }
6678: /* This for computing probability of death (h=1 means
6679: computed over hstepm matrices product = hstepm*stepm months)
6680: as a weighted average of prlim.
6681: */
6682: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6683: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6684: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6685: }
1.279 brouard 6686: /* end shifting computations */
6687:
6688: /**< Computing gradient matrix at horizon h
6689: */
1.218 brouard 6690: for(j=1; j<= nlstate; j++) /* vareij */
6691: for(h=0; h<=nhstepm; h++){
6692: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6693: }
1.279 brouard 6694: /**< Gradient of overall mortality p.3 (or p.j)
6695: */
6696: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6697: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6698: }
6699:
6700: } /* End theta */
1.279 brouard 6701:
6702: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6703: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6704:
6705: for(h=0; h<=nhstepm; h++) /* veij */
6706: for(j=1; j<=nlstate;j++)
6707: for(theta=1; theta <=npar; theta++)
6708: trgradg[h][j][theta]=gradg[h][theta][j];
6709:
6710: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6711: for(theta=1; theta <=npar; theta++)
6712: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6713: /**< as well as its transposed matrix
6714: */
1.218 brouard 6715:
6716: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6717: for(i=1;i<=nlstate;i++)
6718: for(j=1;j<=nlstate;j++)
6719: vareij[i][j][(int)age] =0.;
1.279 brouard 6720:
6721: /* Computing trgradg by matcov by gradg at age and summing over h
6722: * and k (nhstepm) formula 15 of article
6723: * Lievre-Brouard-Heathcote
6724: */
6725:
1.218 brouard 6726: for(h=0;h<=nhstepm;h++){
6727: for(k=0;k<=nhstepm;k++){
6728: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6729: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6730: for(i=1;i<=nlstate;i++)
6731: for(j=1;j<=nlstate;j++)
6732: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6733: }
6734: }
6735:
1.279 brouard 6736: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6737: * p.j overall mortality formula 49 but computed directly because
6738: * we compute the grad (wix pijx) instead of grad (pijx),even if
6739: * wix is independent of theta.
6740: */
1.218 brouard 6741: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6742: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6743: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6744: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6745: varppt[j][i]=doldmp[j][i];
6746: /* end ppptj */
6747: /* x centered again */
6748:
1.242 brouard 6749: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6750:
6751: if (popbased==1) {
6752: if(mobilav ==0){
6753: for(i=1; i<=nlstate;i++)
6754: prlim[i][i]=probs[(int)age][i][ij];
6755: }else{ /* mobilav */
6756: for(i=1; i<=nlstate;i++)
6757: prlim[i][i]=mobaverage[(int)age][i][ij];
6758: }
6759: }
6760:
6761: /* This for computing probability of death (h=1 means
6762: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6763: as a weighted average of prlim.
6764: */
1.235 brouard 6765: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6766: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6767: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6768: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6769: }
6770: /* end probability of death */
6771:
6772: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6773: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6774: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6775: for(i=1; i<=nlstate;i++){
6776: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6777: }
6778: }
6779: fprintf(ficresprobmorprev,"\n");
6780:
6781: fprintf(ficresvij,"%.0f ",age );
6782: for(i=1; i<=nlstate;i++)
6783: for(j=1; j<=nlstate;j++){
6784: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6785: }
6786: fprintf(ficresvij,"\n");
6787: free_matrix(gp,0,nhstepm,1,nlstate);
6788: free_matrix(gm,0,nhstepm,1,nlstate);
6789: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6790: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6791: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6792: } /* End age */
6793: free_vector(gpp,nlstate+1,nlstate+ndeath);
6794: free_vector(gmp,nlstate+1,nlstate+ndeath);
6795: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6796: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6797: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6798: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6799: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6800: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6801: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6802: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6803: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6804: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6805: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6806: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6807: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6808: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6809: 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);
6810: /* 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 6811: */
1.218 brouard 6812: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6813: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6814:
1.218 brouard 6815: free_vector(xp,1,npar);
6816: free_matrix(doldm,1,nlstate,1,nlstate);
6817: free_matrix(dnewm,1,nlstate,1,npar);
6818: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6819: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6820: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6821: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6822: fclose(ficresprobmorprev);
6823: fflush(ficgp);
6824: fflush(fichtm);
6825: } /* end varevsij */
1.126 brouard 6826:
6827: /************ Variance of prevlim ******************/
1.269 brouard 6828: 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 6829: {
1.205 brouard 6830: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6831: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6832:
1.268 brouard 6833: double **dnewmpar,**doldm;
1.126 brouard 6834: int i, j, nhstepm, hstepm;
6835: double *xp;
6836: double *gp, *gm;
6837: double **gradg, **trgradg;
1.208 brouard 6838: double **mgm, **mgp;
1.126 brouard 6839: double age,agelim;
6840: int theta;
6841:
6842: pstamp(ficresvpl);
1.288 brouard 6843: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6844: fprintf(ficresvpl,"# Age ");
6845: if(nresult >=1)
6846: fprintf(ficresvpl," Result# ");
1.126 brouard 6847: for(i=1; i<=nlstate;i++)
6848: fprintf(ficresvpl," %1d-%1d",i,i);
6849: fprintf(ficresvpl,"\n");
6850:
6851: xp=vector(1,npar);
1.268 brouard 6852: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6853: doldm=matrix(1,nlstate,1,nlstate);
6854:
6855: hstepm=1*YEARM; /* Every year of age */
6856: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6857: agelim = AGESUP;
6858: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6859: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6860: if (stepm >= YEARM) hstepm=1;
6861: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6862: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6863: mgp=matrix(1,npar,1,nlstate);
6864: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6865: gp=vector(1,nlstate);
6866: gm=vector(1,nlstate);
6867:
6868: for(theta=1; theta <=npar; theta++){
6869: for(i=1; i<=npar; i++){ /* Computes gradient */
6870: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6871: }
1.288 brouard 6872: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6873: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6874: /* else */
6875: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6876: for(i=1;i<=nlstate;i++){
1.126 brouard 6877: gp[i] = prlim[i][i];
1.208 brouard 6878: mgp[theta][i] = prlim[i][i];
6879: }
1.126 brouard 6880: for(i=1; i<=npar; i++) /* Computes gradient */
6881: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6882: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6883: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6884: /* else */
6885: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6886: for(i=1;i<=nlstate;i++){
1.126 brouard 6887: gm[i] = prlim[i][i];
1.208 brouard 6888: mgm[theta][i] = prlim[i][i];
6889: }
1.126 brouard 6890: for(i=1;i<=nlstate;i++)
6891: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6892: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6893: } /* End theta */
6894:
6895: trgradg =matrix(1,nlstate,1,npar);
6896:
6897: for(j=1; j<=nlstate;j++)
6898: for(theta=1; theta <=npar; theta++)
6899: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6900: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6901: /* printf("\nmgm mgp %d ",(int)age); */
6902: /* for(j=1; j<=nlstate;j++){ */
6903: /* printf(" %d ",j); */
6904: /* for(theta=1; theta <=npar; theta++) */
6905: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6906: /* printf("\n "); */
6907: /* } */
6908: /* } */
6909: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6910: /* printf("\n gradg %d ",(int)age); */
6911: /* for(j=1; j<=nlstate;j++){ */
6912: /* printf("%d ",j); */
6913: /* for(theta=1; theta <=npar; theta++) */
6914: /* printf("%d %lf ",theta,gradg[theta][j]); */
6915: /* printf("\n "); */
6916: /* } */
6917: /* } */
1.126 brouard 6918:
6919: for(i=1;i<=nlstate;i++)
6920: varpl[i][(int)age] =0.;
1.209 brouard 6921: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6922: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6923: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6924: }else{
1.268 brouard 6925: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6926: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6927: }
1.126 brouard 6928: for(i=1;i<=nlstate;i++)
6929: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6930:
6931: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6932: if(nresult >=1)
6933: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6934: for(i=1; i<=nlstate;i++){
1.126 brouard 6935: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6936: /* for(j=1;j<=nlstate;j++) */
6937: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6938: }
1.126 brouard 6939: fprintf(ficresvpl,"\n");
6940: free_vector(gp,1,nlstate);
6941: free_vector(gm,1,nlstate);
1.208 brouard 6942: free_matrix(mgm,1,npar,1,nlstate);
6943: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6944: free_matrix(gradg,1,npar,1,nlstate);
6945: free_matrix(trgradg,1,nlstate,1,npar);
6946: } /* End age */
6947:
6948: free_vector(xp,1,npar);
6949: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6950: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6951:
6952: }
6953:
6954:
6955: /************ Variance of backprevalence limit ******************/
1.269 brouard 6956: 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 6957: {
6958: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6959: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6960:
6961: double **dnewmpar,**doldm;
6962: int i, j, nhstepm, hstepm;
6963: double *xp;
6964: double *gp, *gm;
6965: double **gradg, **trgradg;
6966: double **mgm, **mgp;
6967: double age,agelim;
6968: int theta;
6969:
6970: pstamp(ficresvbl);
6971: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6972: fprintf(ficresvbl,"# Age ");
6973: if(nresult >=1)
6974: fprintf(ficresvbl," Result# ");
6975: for(i=1; i<=nlstate;i++)
6976: fprintf(ficresvbl," %1d-%1d",i,i);
6977: fprintf(ficresvbl,"\n");
6978:
6979: xp=vector(1,npar);
6980: dnewmpar=matrix(1,nlstate,1,npar);
6981: doldm=matrix(1,nlstate,1,nlstate);
6982:
6983: hstepm=1*YEARM; /* Every year of age */
6984: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6985: agelim = AGEINF;
6986: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6987: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6988: if (stepm >= YEARM) hstepm=1;
6989: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6990: gradg=matrix(1,npar,1,nlstate);
6991: mgp=matrix(1,npar,1,nlstate);
6992: mgm=matrix(1,npar,1,nlstate);
6993: gp=vector(1,nlstate);
6994: gm=vector(1,nlstate);
6995:
6996: for(theta=1; theta <=npar; theta++){
6997: for(i=1; i<=npar; i++){ /* Computes gradient */
6998: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6999: }
7000: if(mobilavproj > 0 )
7001: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7002: else
7003: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7004: for(i=1;i<=nlstate;i++){
7005: gp[i] = bprlim[i][i];
7006: mgp[theta][i] = bprlim[i][i];
7007: }
7008: for(i=1; i<=npar; i++) /* Computes gradient */
7009: xp[i] = x[i] - (i==theta ?delti[theta]:0);
7010: if(mobilavproj > 0 )
7011: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7012: else
7013: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7014: for(i=1;i<=nlstate;i++){
7015: gm[i] = bprlim[i][i];
7016: mgm[theta][i] = bprlim[i][i];
7017: }
7018: for(i=1;i<=nlstate;i++)
7019: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
7020: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
7021: } /* End theta */
7022:
7023: trgradg =matrix(1,nlstate,1,npar);
7024:
7025: for(j=1; j<=nlstate;j++)
7026: for(theta=1; theta <=npar; theta++)
7027: trgradg[j][theta]=gradg[theta][j];
7028: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7029: /* printf("\nmgm mgp %d ",(int)age); */
7030: /* for(j=1; j<=nlstate;j++){ */
7031: /* printf(" %d ",j); */
7032: /* for(theta=1; theta <=npar; theta++) */
7033: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7034: /* printf("\n "); */
7035: /* } */
7036: /* } */
7037: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7038: /* printf("\n gradg %d ",(int)age); */
7039: /* for(j=1; j<=nlstate;j++){ */
7040: /* printf("%d ",j); */
7041: /* for(theta=1; theta <=npar; theta++) */
7042: /* printf("%d %lf ",theta,gradg[theta][j]); */
7043: /* printf("\n "); */
7044: /* } */
7045: /* } */
7046:
7047: for(i=1;i<=nlstate;i++)
7048: varbpl[i][(int)age] =0.;
7049: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
7050: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7051: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7052: }else{
7053: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7054: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7055: }
7056: for(i=1;i<=nlstate;i++)
7057: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7058:
7059: fprintf(ficresvbl,"%.0f ",age );
7060: if(nresult >=1)
7061: fprintf(ficresvbl,"%d ",nres );
7062: for(i=1; i<=nlstate;i++)
7063: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
7064: fprintf(ficresvbl,"\n");
7065: free_vector(gp,1,nlstate);
7066: free_vector(gm,1,nlstate);
7067: free_matrix(mgm,1,npar,1,nlstate);
7068: free_matrix(mgp,1,npar,1,nlstate);
7069: free_matrix(gradg,1,npar,1,nlstate);
7070: free_matrix(trgradg,1,nlstate,1,npar);
7071: } /* End age */
7072:
7073: free_vector(xp,1,npar);
7074: free_matrix(doldm,1,nlstate,1,npar);
7075: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 7076:
7077: }
7078:
7079: /************ Variance of one-step probabilities ******************/
7080: 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 7081: {
7082: int i, j=0, k1, l1, tj;
7083: int k2, l2, j1, z1;
7084: int k=0, l;
7085: int first=1, first1, first2;
1.326 brouard 7086: int nres=0; /* New */
1.222 brouard 7087: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
7088: double **dnewm,**doldm;
7089: double *xp;
7090: double *gp, *gm;
7091: double **gradg, **trgradg;
7092: double **mu;
7093: double age, cov[NCOVMAX+1];
7094: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
7095: int theta;
7096: char fileresprob[FILENAMELENGTH];
7097: char fileresprobcov[FILENAMELENGTH];
7098: char fileresprobcor[FILENAMELENGTH];
7099: double ***varpij;
7100:
7101: strcpy(fileresprob,"PROB_");
7102: strcat(fileresprob,fileres);
7103: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
7104: printf("Problem with resultfile: %s\n", fileresprob);
7105: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
7106: }
7107: strcpy(fileresprobcov,"PROBCOV_");
7108: strcat(fileresprobcov,fileresu);
7109: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
7110: printf("Problem with resultfile: %s\n", fileresprobcov);
7111: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
7112: }
7113: strcpy(fileresprobcor,"PROBCOR_");
7114: strcat(fileresprobcor,fileresu);
7115: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
7116: printf("Problem with resultfile: %s\n", fileresprobcor);
7117: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
7118: }
7119: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7120: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7121: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7122: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7123: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7124: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7125: pstamp(ficresprob);
7126: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
7127: fprintf(ficresprob,"# Age");
7128: pstamp(ficresprobcov);
7129: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
7130: fprintf(ficresprobcov,"# Age");
7131: pstamp(ficresprobcor);
7132: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
7133: fprintf(ficresprobcor,"# Age");
1.126 brouard 7134:
7135:
1.222 brouard 7136: for(i=1; i<=nlstate;i++)
7137: for(j=1; j<=(nlstate+ndeath);j++){
7138: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
7139: fprintf(ficresprobcov," p%1d-%1d ",i,j);
7140: fprintf(ficresprobcor," p%1d-%1d ",i,j);
7141: }
7142: /* fprintf(ficresprob,"\n");
7143: fprintf(ficresprobcov,"\n");
7144: fprintf(ficresprobcor,"\n");
7145: */
7146: xp=vector(1,npar);
7147: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7148: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7149: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
7150: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
7151: first=1;
7152: fprintf(ficgp,"\n# Routine varprob");
7153: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
7154: fprintf(fichtm,"\n");
7155:
1.288 brouard 7156: 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 7157: 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);
7158: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 7159: and drawn. It helps understanding how is the covariance between two incidences.\
7160: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 7161: 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 7162: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
7163: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
7164: standard deviations wide on each axis. <br>\
7165: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
7166: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
7167: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
7168:
1.222 brouard 7169: cov[1]=1;
7170: /* tj=cptcoveff; */
1.225 brouard 7171: tj = (int) pow(2,cptcoveff);
1.222 brouard 7172: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
7173: j1=0;
1.332 brouard 7174:
7175: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
7176: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.334 brouard 7177: 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 7178: if(tj != 1 && TKresult[nres]!= j1)
7179: continue;
7180:
7181: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
7182: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
7183: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 7184: if (cptcovn>0) {
1.334 brouard 7185: fprintf(ficresprob, "\n#********** Variable ");
7186: fprintf(ficresprobcov, "\n#********** Variable ");
7187: fprintf(ficgp, "\n#********** Variable ");
7188: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
7189: fprintf(ficresprobcor, "\n#********** Variable ");
7190:
7191: /* Including quantitative variables of the resultline to be done */
7192: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
7193: printf("Varprob modelresult[%d][%d]=%d model=%s \n",nres, z1, modelresult[nres][z1], model);
7194: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=%s \n",nres, z1, modelresult[nres][z1], model);
7195: /* fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=%s resultline[%d]=%s \n",nres, z1, modelresult[nres][z1], model, nres, resultline[nres]); */
7196: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
7197: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
7198: 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 */
7199: 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 */
7200: 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 */
7201: 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 */
7202: 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 */
7203: fprintf(ficresprob,"fixed ");
7204: fprintf(ficresprobcov,"fixed ");
7205: fprintf(ficgp,"fixed ");
7206: fprintf(fichtmcov,"fixed ");
7207: fprintf(ficresprobcor,"fixed ");
7208: }else{
7209: fprintf(ficresprob,"varyi ");
7210: fprintf(ficresprobcov,"varyi ");
7211: fprintf(ficgp,"varyi ");
7212: fprintf(fichtmcov,"varyi ");
7213: fprintf(ficresprobcor,"varyi ");
7214: }
7215: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
7216: /* For each selected (single) quantitative value */
7217: fprintf(ficresprob," V%d=%f ",Tvqresult[nres][z1],Tqresult[nres][z1]);
7218: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
7219: fprintf(ficresprob,"fixed ");
7220: fprintf(ficresprobcov,"fixed ");
7221: fprintf(ficgp,"fixed ");
7222: fprintf(fichtmcov,"fixed ");
7223: fprintf(ficresprobcor,"fixed ");
7224: }else{
7225: fprintf(ficresprob,"varyi ");
7226: fprintf(ficresprobcov,"varyi ");
7227: fprintf(ficgp,"varyi ");
7228: fprintf(fichtmcov,"varyi ");
7229: fprintf(ficresprobcor,"varyi ");
7230: }
7231: }else{
7232: 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 */
7233: 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 */
7234: exit(1);
7235: }
7236: } /* End loop on variable of this resultline */
7237: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 7238: fprintf(ficresprob, "**********\n#\n");
7239: fprintf(ficresprobcov, "**********\n#\n");
7240: fprintf(ficgp, "**********\n#\n");
7241: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
7242: fprintf(ficresprobcor, "**********\n#");
7243: if(invalidvarcomb[j1]){
7244: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7245: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7246: continue;
7247: }
7248: }
7249: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7250: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7251: gp=vector(1,(nlstate)*(nlstate+ndeath));
7252: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 7253: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 7254: cov[2]=age;
7255: if(nagesqr==1)
7256: cov[3]= age*age;
1.334 brouard 7257: /* New code end of combination but for each resultline */
7258: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
7259: if(Typevar[k1]==1){ /* A product with age */
7260: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 7261: }else{
1.334 brouard 7262: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 7263: }
1.334 brouard 7264: }/* End of loop on model equation */
7265: /* Old code */
7266: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
7267: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
7268: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
7269: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
7270: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
7271: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
7272: /* * 1 1 1 1 1 */
7273: /* * 2 2 1 1 1 */
7274: /* * 3 1 2 1 1 */
7275: /* *\/ */
7276: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
7277: /* } */
7278: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
7279: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
7280: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
7281: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
7282: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
7283: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
7284: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7285: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
7286: /* 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]); */
7287: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
7288: /* /\* exit(1); *\/ */
7289: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
7290: /* } */
7291: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7292: /* } */
7293: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
7294: /* if(Dummy[Tvard[k][1]]==0){ */
7295: /* if(Dummy[Tvard[k][2]]==0){ */
7296: /* 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]])]; */
7297: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7298: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
7299: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
7300: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
7301: /* } */
7302: /* }else{ */
7303: /* if(Dummy[Tvard[k][2]]==0){ */
7304: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
7305: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
7306: /* }else{ */
7307: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
7308: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
7309: /* } */
7310: /* } */
7311: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7312: /* } */
1.326 brouard 7313: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7314: for(theta=1; theta <=npar; theta++){
7315: for(i=1; i<=npar; i++)
7316: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7317:
1.222 brouard 7318: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7319:
1.222 brouard 7320: k=0;
7321: for(i=1; i<= (nlstate); i++){
7322: for(j=1; j<=(nlstate+ndeath);j++){
7323: k=k+1;
7324: gp[k]=pmmij[i][j];
7325: }
7326: }
1.220 brouard 7327:
1.222 brouard 7328: for(i=1; i<=npar; i++)
7329: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7330:
1.222 brouard 7331: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7332: k=0;
7333: for(i=1; i<=(nlstate); i++){
7334: for(j=1; j<=(nlstate+ndeath);j++){
7335: k=k+1;
7336: gm[k]=pmmij[i][j];
7337: }
7338: }
1.220 brouard 7339:
1.222 brouard 7340: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7341: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7342: }
1.126 brouard 7343:
1.222 brouard 7344: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7345: for(theta=1; theta <=npar; theta++)
7346: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7347:
1.222 brouard 7348: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7349: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7350:
1.222 brouard 7351: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7352:
1.222 brouard 7353: k=0;
7354: for(i=1; i<=(nlstate); i++){
7355: for(j=1; j<=(nlstate+ndeath);j++){
7356: k=k+1;
7357: mu[k][(int) age]=pmmij[i][j];
7358: }
7359: }
7360: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7361: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7362: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7363:
1.222 brouard 7364: /*printf("\n%d ",(int)age);
7365: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7366: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7367: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7368: }*/
1.220 brouard 7369:
1.222 brouard 7370: fprintf(ficresprob,"\n%d ",(int)age);
7371: fprintf(ficresprobcov,"\n%d ",(int)age);
7372: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7373:
1.222 brouard 7374: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7375: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7376: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7377: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7378: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7379: }
7380: i=0;
7381: for (k=1; k<=(nlstate);k++){
7382: for (l=1; l<=(nlstate+ndeath);l++){
7383: i++;
7384: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7385: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7386: for (j=1; j<=i;j++){
7387: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7388: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7389: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7390: }
7391: }
7392: }/* end of loop for state */
7393: } /* end of loop for age */
7394: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7395: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7396: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7397: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7398:
7399: /* Confidence intervalle of pij */
7400: /*
7401: fprintf(ficgp,"\nunset parametric;unset label");
7402: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7403: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7404: 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);
7405: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7406: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7407: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7408: */
7409:
7410: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7411: first1=1;first2=2;
7412: for (k2=1; k2<=(nlstate);k2++){
7413: for (l2=1; l2<=(nlstate+ndeath);l2++){
7414: if(l2==k2) continue;
7415: j=(k2-1)*(nlstate+ndeath)+l2;
7416: for (k1=1; k1<=(nlstate);k1++){
7417: for (l1=1; l1<=(nlstate+ndeath);l1++){
7418: if(l1==k1) continue;
7419: i=(k1-1)*(nlstate+ndeath)+l1;
7420: if(i<=j) continue;
7421: for (age=bage; age<=fage; age ++){
7422: if ((int)age %5==0){
7423: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7424: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7425: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7426: mu1=mu[i][(int) age]/stepm*YEARM ;
7427: mu2=mu[j][(int) age]/stepm*YEARM;
7428: c12=cv12/sqrt(v1*v2);
7429: /* Computing eigen value of matrix of covariance */
7430: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7431: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7432: if ((lc2 <0) || (lc1 <0) ){
7433: if(first2==1){
7434: first1=0;
7435: 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);
7436: }
7437: 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);
7438: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7439: /* lc2=fabs(lc2); */
7440: }
1.220 brouard 7441:
1.222 brouard 7442: /* Eigen vectors */
1.280 brouard 7443: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7444: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7445: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7446: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7447: }else
7448: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7449: /*v21=sqrt(1.-v11*v11); *//* error */
7450: v21=(lc1-v1)/cv12*v11;
7451: v12=-v21;
7452: v22=v11;
7453: tnalp=v21/v11;
7454: if(first1==1){
7455: first1=0;
7456: 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);
7457: }
7458: 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);
7459: /*printf(fignu*/
7460: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7461: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7462: if(first==1){
7463: first=0;
7464: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7465: fprintf(ficgp,"\nset parametric;unset label");
7466: 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);
7467: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7468: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7469: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7470: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7471: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7472: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7473: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7474: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7475: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7476: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7477: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7478: 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 7479: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7480: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7481: }else{
7482: first=0;
7483: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7484: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7485: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7486: 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 7487: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7488: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7489: }/* if first */
7490: } /* age mod 5 */
7491: } /* end loop age */
7492: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7493: first=1;
7494: } /*l12 */
7495: } /* k12 */
7496: } /*l1 */
7497: }/* k1 */
1.332 brouard 7498: } /* loop on combination of covariates j1 */
1.326 brouard 7499: } /* loop on nres */
1.222 brouard 7500: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7501: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7502: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7503: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7504: free_vector(xp,1,npar);
7505: fclose(ficresprob);
7506: fclose(ficresprobcov);
7507: fclose(ficresprobcor);
7508: fflush(ficgp);
7509: fflush(fichtmcov);
7510: }
1.126 brouard 7511:
7512:
7513: /******************* Printing html file ***********/
1.201 brouard 7514: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7515: int lastpass, int stepm, int weightopt, char model[],\
7516: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7517: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7518: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7519: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7520: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7521: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7522: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7523: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7524: </ul>");
1.319 brouard 7525: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7526: /* </ul>", model); */
1.214 brouard 7527: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7528: 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",
7529: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 7530: 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 7531: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7532: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7533: fprintf(fichtm,"\
7534: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7535: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7536: fprintf(fichtm,"\
1.217 brouard 7537: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7538: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7539: fprintf(fichtm,"\
1.288 brouard 7540: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7541: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7542: fprintf(fichtm,"\
1.288 brouard 7543: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7544: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7545: fprintf(fichtm,"\
1.211 brouard 7546: - (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 7547: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7548: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7549: if(prevfcast==1){
7550: fprintf(fichtm,"\
7551: - Prevalence projections by age and states: \
1.201 brouard 7552: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7553: }
1.126 brouard 7554:
7555:
1.225 brouard 7556: m=pow(2,cptcoveff);
1.222 brouard 7557: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7558:
1.317 brouard 7559: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7560:
7561: jj1=0;
7562:
7563: fprintf(fichtm," \n<ul>");
7564: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7565: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7566: if(m != 1 && TKresult[nres]!= k1)
7567: continue;
7568: jj1++;
7569: if (cptcovn > 0) {
7570: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
7571: for (cpt=1; cpt<=cptcoveff;cpt++){
7572: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7573: }
7574: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7575: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7576: }
7577: fprintf(fichtm,"\">");
7578:
7579: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7580: fprintf(fichtm,"************ Results for covariates");
7581: for (cpt=1; cpt<=cptcoveff;cpt++){
7582: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7583: }
7584: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7585: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7586: }
7587: if(invalidvarcomb[k1]){
7588: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7589: continue;
7590: }
7591: fprintf(fichtm,"</a></li>");
7592: } /* cptcovn >0 */
7593: }
1.317 brouard 7594: fprintf(fichtm," \n</ul>");
1.264 brouard 7595:
1.222 brouard 7596: jj1=0;
1.237 brouard 7597:
7598: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7599: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7600: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7601: continue;
1.220 brouard 7602:
1.222 brouard 7603: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7604: jj1++;
7605: if (cptcovn > 0) {
1.264 brouard 7606: fprintf(fichtm,"\n<p><a name=\"rescov");
7607: for (cpt=1; cpt<=cptcoveff;cpt++){
7608: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7609: }
7610: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7611: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7612: }
7613: fprintf(fichtm,"\"</a>");
7614:
1.222 brouard 7615: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7616: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7617: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7618: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7619: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7620: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7621: }
1.237 brouard 7622: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7623: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7624: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7625: }
7626:
1.230 brouard 7627: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.321 brouard 7628: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7629: if(invalidvarcomb[k1]){
7630: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7631: printf("\nCombination (%d) ignored because no cases \n",k1);
7632: continue;
7633: }
7634: }
7635: /* aij, bij */
1.259 brouard 7636: 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 7637: <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 7638: /* Pij */
1.241 brouard 7639: 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> \
7640: <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 7641: /* Quasi-incidences */
7642: 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 7643: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7644: 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 7645: 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> \
7646: <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 7647: /* Survival functions (period) in state j */
7648: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7649: 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);
7650: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7651: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7652: }
7653: /* State specific survival functions (period) */
7654: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7655: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7656: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7657: <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);
7658: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7659: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7660: }
1.288 brouard 7661: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7662: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7663: 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);
7664: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"P_"),subdirf2(optionfilefiname,"P_"));
7665: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7666: }
1.296 brouard 7667: if(prevbcast==1){
1.288 brouard 7668: /* Backward prevalence in each health state */
1.222 brouard 7669: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7670: 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 7671: <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 7672: }
1.217 brouard 7673: }
1.222 brouard 7674: if(prevfcast==1){
1.288 brouard 7675: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7676: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7677: 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);
7678: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7679: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7680: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7681: }
7682: }
1.296 brouard 7683: if(prevbcast==1){
1.268 brouard 7684: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7685: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7686: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7687: 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 \
7688: 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 7689: 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);
7690: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7691: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7692: }
7693: }
1.220 brouard 7694:
1.222 brouard 7695: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7696: 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);
7697: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7698: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7699: }
7700: /* } /\* end i1 *\/ */
7701: }/* End k1 */
7702: fprintf(fichtm,"</ul>");
1.126 brouard 7703:
1.222 brouard 7704: fprintf(fichtm,"\
1.126 brouard 7705: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7706: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7707: - 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 7708: But because parameters are usually highly correlated (a higher incidence of disability \
7709: and a higher incidence of recovery can give very close observed transition) it might \
7710: be very useful to look not only at linear confidence intervals estimated from the \
7711: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7712: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7713: covariance matrix of the one-step probabilities. \
7714: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7715:
1.222 brouard 7716: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7717: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7718: fprintf(fichtm,"\
1.126 brouard 7719: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7720: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7721:
1.222 brouard 7722: fprintf(fichtm,"\
1.126 brouard 7723: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7724: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7725: fprintf(fichtm,"\
1.126 brouard 7726: - 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): \
7727: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7728: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7729: fprintf(fichtm,"\
1.126 brouard 7730: - (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): \
7731: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7732: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7733: fprintf(fichtm,"\
1.288 brouard 7734: - 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 7735: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7736: fprintf(fichtm,"\
1.128 brouard 7737: - 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 7738: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7739: fprintf(fichtm,"\
1.288 brouard 7740: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7741: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7742:
7743: /* if(popforecast==1) fprintf(fichtm,"\n */
7744: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7745: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7746: /* <br>",fileres,fileres,fileres,fileres); */
7747: /* else */
7748: /* 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 7749: fflush(fichtm);
1.126 brouard 7750:
1.225 brouard 7751: m=pow(2,cptcoveff);
1.222 brouard 7752: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7753:
1.317 brouard 7754: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7755:
7756: jj1=0;
7757:
7758: fprintf(fichtm," \n<ul>");
7759: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7760: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7761: if(m != 1 && TKresult[nres]!= k1)
7762: continue;
7763: jj1++;
7764: if (cptcovn > 0) {
7765: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
7766: for (cpt=1; cpt<=cptcoveff;cpt++){
7767: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7768: }
7769: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7770: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7771: }
7772: fprintf(fichtm,"\">");
7773:
7774: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7775: fprintf(fichtm,"************ Results for covariates");
7776: for (cpt=1; cpt<=cptcoveff;cpt++){
7777: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7778: }
7779: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7780: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7781: }
7782: if(invalidvarcomb[k1]){
7783: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7784: continue;
7785: }
7786: fprintf(fichtm,"</a></li>");
7787: } /* cptcovn >0 */
7788: }
7789: fprintf(fichtm," \n</ul>");
7790:
1.222 brouard 7791: jj1=0;
1.237 brouard 7792:
1.241 brouard 7793: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7794: for(k1=1; k1<=m;k1++){
1.253 brouard 7795: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7796: continue;
1.222 brouard 7797: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7798: jj1++;
1.126 brouard 7799: if (cptcovn > 0) {
1.317 brouard 7800: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
7801: for (cpt=1; cpt<=cptcoveff;cpt++){
7802: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7803: }
7804: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7805: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7806: }
7807: fprintf(fichtm,"\"</a>");
7808:
1.126 brouard 7809: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.317 brouard 7810: for (cpt=1; cpt<=cptcoveff;cpt++){ /**< cptcoveff number of variables */
1.237 brouard 7811: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
1.317 brouard 7812: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
1.237 brouard 7813: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 7814: }
1.237 brouard 7815: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7816: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7817: }
7818:
1.321 brouard 7819: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 7820:
1.222 brouard 7821: if(invalidvarcomb[k1]){
7822: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7823: continue;
7824: }
1.126 brouard 7825: }
7826: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7827: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7828: 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);
7829: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7830: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7831: }
7832: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7833: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7834: true period expectancies (those weighted with period prevalences are also\
7835: drawn in addition to the population based expectancies computed using\
1.314 brouard 7836: 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);
7837: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7838: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7839: /* } /\* end i1 *\/ */
7840: }/* End k1 */
1.241 brouard 7841: }/* End nres */
1.222 brouard 7842: fprintf(fichtm,"</ul>");
7843: fflush(fichtm);
1.126 brouard 7844: }
7845:
7846: /******************* Gnuplot file **************/
1.296 brouard 7847: 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 7848:
7849: char dirfileres[132],optfileres[132];
1.264 brouard 7850: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7851: 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 7852: int lv=0, vlv=0, kl=0;
1.130 brouard 7853: int ng=0;
1.201 brouard 7854: int vpopbased;
1.223 brouard 7855: int ioffset; /* variable offset for columns */
1.270 brouard 7856: int iyearc=1; /* variable column for year of projection */
7857: int iagec=1; /* variable column for age of projection */
1.235 brouard 7858: int nres=0; /* Index of resultline */
1.266 brouard 7859: int istart=1; /* For starting graphs in projections */
1.219 brouard 7860:
1.126 brouard 7861: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7862: /* printf("Problem with file %s",optionfilegnuplot); */
7863: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7864: /* } */
7865:
7866: /*#ifdef windows */
7867: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7868: /*#endif */
1.225 brouard 7869: m=pow(2,cptcoveff);
1.126 brouard 7870:
1.274 brouard 7871: /* diagram of the model */
7872: fprintf(ficgp,"\n#Diagram of the model \n");
7873: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7874: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7875: 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);
7876:
7877: 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);
7878: fprintf(ficgp,"\n#show arrow\nunset label\n");
7879: 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);
7880: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7881: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7882: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7883: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7884:
1.202 brouard 7885: /* Contribution to likelihood */
7886: /* Plot the probability implied in the likelihood */
1.223 brouard 7887: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7888: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7889: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7890: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7891: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7892: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7893: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7894: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7895: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7896: 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));
7897: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7898: 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));
7899: for (i=1; i<= nlstate ; i ++) {
7900: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7901: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7902: 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);
7903: for (j=2; j<= nlstate+ndeath ; j ++) {
7904: 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);
7905: }
7906: fprintf(ficgp,";\nset out; unset ylabel;\n");
7907: }
7908: /* 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 */
7909: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7910: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7911: fprintf(ficgp,"\nset out;unset log\n");
7912: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7913:
1.126 brouard 7914: strcpy(dirfileres,optionfilefiname);
7915: strcpy(optfileres,"vpl");
1.223 brouard 7916: /* 1eme*/
1.238 brouard 7917: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7918: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7919: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7920: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7921: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7922: continue;
7923: /* We are interested in selected combination by the resultline */
1.246 brouard 7924: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7925: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7926: strcpy(gplotlabel,"(");
1.238 brouard 7927: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
1.332 brouard 7928: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the value of the covariate corresponding to k1 combination *\/ */
7929: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 7930: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7931: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7932: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7933: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7934: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7935: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7936: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7937: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7938: }
7939: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7940: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7941: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7942: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7943: }
7944: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7945: /* printf("\n#\n"); */
1.238 brouard 7946: fprintf(ficgp,"\n#\n");
7947: if(invalidvarcomb[k1]){
1.260 brouard 7948: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7949: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7950: continue;
7951: }
1.235 brouard 7952:
1.241 brouard 7953: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7954: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7955: /* 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 7956: fprintf(ficgp,"set title \"Alive state %d %s model=%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 7957: 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);
7958: /* 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); */
7959: /* k1-1 error should be nres-1*/
1.238 brouard 7960: for (i=1; i<= nlstate ; i ++) {
7961: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7962: else fprintf(ficgp," %%*lf (%%*lf)");
7963: }
1.288 brouard 7964: 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 7965: for (i=1; i<= nlstate ; i ++) {
7966: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7967: else fprintf(ficgp," %%*lf (%%*lf)");
7968: }
1.260 brouard 7969: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7970: for (i=1; i<= nlstate ; i ++) {
7971: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7972: else fprintf(ficgp," %%*lf (%%*lf)");
7973: }
1.265 brouard 7974: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7975:
7976: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7977: if(cptcoveff ==0){
1.271 brouard 7978: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7979: }else{
7980: kl=0;
7981: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 7982: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
7983: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 7984: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7985: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7986: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7987: vlv= nbcode[Tvaraff[k]][lv];
7988: kl++;
7989: /* 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 *\/ */
7990: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7991: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7992: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7993: if(k==cptcoveff){
7994: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7995: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7996: }else{
7997: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7998: kl++;
7999: }
8000: } /* end covariate */
8001: } /* end if no covariate */
8002:
1.296 brouard 8003: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 8004: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 8005: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 8006: if(cptcoveff ==0){
1.245 brouard 8007: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 8008: }else{
8009: kl=0;
8010: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8011: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8012: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 8013: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8014: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8015: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8016: /* vlv= nbcode[Tvaraff[k]][lv]; */
8017: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 8018: kl++;
1.238 brouard 8019: /* 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 *\/ */
8020: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8021: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8022: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8023: if(k==cptcoveff){
1.245 brouard 8024: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 8025: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 8026: }else{
1.332 brouard 8027: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 8028: kl++;
8029: }
8030: } /* end covariate */
8031: } /* end if no covariate */
1.296 brouard 8032: if(prevbcast == 1){
1.268 brouard 8033: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
8034: /* k1-1 error should be nres-1*/
8035: for (i=1; i<= nlstate ; i ++) {
8036: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8037: else fprintf(ficgp," %%*lf (%%*lf)");
8038: }
1.271 brouard 8039: 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 8040: for (i=1; i<= nlstate ; i ++) {
8041: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8042: else fprintf(ficgp," %%*lf (%%*lf)");
8043: }
1.276 brouard 8044: 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 8045: for (i=1; i<= nlstate ; i ++) {
8046: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8047: else fprintf(ficgp," %%*lf (%%*lf)");
8048: }
1.274 brouard 8049: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 8050: } /* end if backprojcast */
1.296 brouard 8051: } /* end if prevbcast */
1.276 brouard 8052: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
8053: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 8054: } /* nres */
1.201 brouard 8055: } /* k1 */
8056: } /* cpt */
1.235 brouard 8057:
8058:
1.126 brouard 8059: /*2 eme*/
1.238 brouard 8060: for (k1=1; k1<= m ; k1 ++){
8061: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8062: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 8063: continue;
8064: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 8065: strcpy(gplotlabel,"(");
1.238 brouard 8066: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8067: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8068: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.223 brouard 8069: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8070: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8071: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8072: /* vlv= nbcode[Tvaraff[k]][lv]; */
8073: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 8074: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8075: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 8076: }
1.237 brouard 8077: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 8078: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8079: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 8080: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8081: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 8082: }
1.264 brouard 8083: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8084: fprintf(ficgp,"\n#\n");
1.223 brouard 8085: if(invalidvarcomb[k1]){
8086: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8087: continue;
8088: }
1.219 brouard 8089:
1.241 brouard 8090: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 8091: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 8092: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
8093: if(vpopbased==0){
1.238 brouard 8094: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 8095: }else
1.238 brouard 8096: fprintf(ficgp,"\nreplot ");
8097: for (i=1; i<= nlstate+1 ; i ++) {
8098: k=2*i;
1.261 brouard 8099: 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 8100: for (j=1; j<= nlstate+1 ; j ++) {
8101: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8102: else fprintf(ficgp," %%*lf (%%*lf)");
8103: }
8104: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
8105: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 8106: 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 8107: for (j=1; j<= nlstate+1 ; j ++) {
8108: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8109: else fprintf(ficgp," %%*lf (%%*lf)");
8110: }
8111: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 8112: 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 8113: for (j=1; j<= nlstate+1 ; j ++) {
8114: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8115: else fprintf(ficgp," %%*lf (%%*lf)");
8116: }
8117: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
8118: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
8119: } /* state */
8120: } /* vpopbased */
1.264 brouard 8121: 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 8122: } /* end nres */
8123: } /* k1 end 2 eme*/
8124:
8125:
8126: /*3eme*/
8127: for (k1=1; k1<= m ; k1 ++){
8128: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8129: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 8130: continue;
8131:
1.332 brouard 8132: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 8133: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 8134: strcpy(gplotlabel,"(");
1.238 brouard 8135: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8136: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8137: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.238 brouard 8138: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8139: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8140: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8141: /* vlv= nbcode[Tvaraff[k]][lv]; */
8142: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.238 brouard 8143: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8144: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 8145: }
8146: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8147: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]);
8148: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]);
1.238 brouard 8149: }
1.264 brouard 8150: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8151: fprintf(ficgp,"\n#\n");
8152: if(invalidvarcomb[k1]){
8153: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8154: continue;
8155: }
8156:
8157: /* k=2+nlstate*(2*cpt-2); */
8158: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 8159: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 8160: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 8161: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 8162: 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 8163: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8164: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8165: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
8166: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8167: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8168: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 8169:
1.238 brouard 8170: */
8171: for (i=1; i< nlstate ; i ++) {
1.261 brouard 8172: 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 8173: /* 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 8174:
1.238 brouard 8175: }
1.261 brouard 8176: 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 8177: }
1.264 brouard 8178: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 8179: } /* end nres */
8180: } /* end kl 3eme */
1.126 brouard 8181:
1.223 brouard 8182: /* 4eme */
1.201 brouard 8183: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 8184: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
8185: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8186: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 8187: continue;
1.238 brouard 8188: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 8189: strcpy(gplotlabel,"(");
1.238 brouard 8190: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
8191: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8192: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
8193: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
1.238 brouard 8194: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8195: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8196: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8197: /* vlv= nbcode[Tvaraff[k]][lv]; */
8198: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.238 brouard 8199: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8200: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 8201: }
8202: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8203: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8204: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.238 brouard 8205: }
1.264 brouard 8206: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8207: fprintf(ficgp,"\n#\n");
8208: if(invalidvarcomb[k1]){
8209: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8210: continue;
1.223 brouard 8211: }
1.238 brouard 8212:
1.241 brouard 8213: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 8214: 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 8215: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8216: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8217: k=3;
8218: for (i=1; i<= nlstate ; i ++){
8219: if(i==1){
8220: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8221: }else{
8222: fprintf(ficgp,", '' ");
8223: }
8224: l=(nlstate+ndeath)*(i-1)+1;
8225: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8226: for (j=2; j<= nlstate+ndeath ; j ++)
8227: fprintf(ficgp,"+$%d",k+l+j-1);
8228: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
8229: } /* nlstate */
1.264 brouard 8230: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8231: } /* end cpt state*/
8232: } /* end nres */
8233: } /* end covariate k1 */
8234:
1.220 brouard 8235: /* 5eme */
1.201 brouard 8236: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 8237: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
8238: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8239: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 8240: continue;
1.238 brouard 8241: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 8242: strcpy(gplotlabel,"(");
1.238 brouard 8243: 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);
8244: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8245: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
8246: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
1.238 brouard 8247: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8248: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8249: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8250: /* vlv= nbcode[Tvaraff[k]][lv]; */
8251: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.238 brouard 8252: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8253: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 8254: }
8255: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8256: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8257: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.238 brouard 8258: }
1.264 brouard 8259: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8260: fprintf(ficgp,"\n#\n");
8261: if(invalidvarcomb[k1]){
8262: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8263: continue;
8264: }
1.227 brouard 8265:
1.241 brouard 8266: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8267: 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 8268: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8269: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8270: k=3;
8271: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8272: if(j==1)
8273: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8274: else
8275: fprintf(ficgp,", '' ");
8276: l=(nlstate+ndeath)*(cpt-1) +j;
8277: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8278: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8279: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8280: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8281: } /* nlstate */
8282: fprintf(ficgp,", '' ");
8283: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8284: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8285: l=(nlstate+ndeath)*(cpt-1) +j;
8286: if(j < nlstate)
8287: fprintf(ficgp,"$%d +",k+l);
8288: else
8289: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8290: }
1.264 brouard 8291: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8292: } /* end cpt state*/
8293: } /* end covariate */
8294: } /* end nres */
1.227 brouard 8295:
1.220 brouard 8296: /* 6eme */
1.202 brouard 8297: /* CV preval stable (period) for each covariate */
1.237 brouard 8298: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8299: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8300: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8301: continue;
1.255 brouard 8302: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8303: strcpy(gplotlabel,"(");
1.288 brouard 8304: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 8305: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8306: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8307: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8308: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8309: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8310: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8311: /* vlv= nbcode[Tvaraff[k]][lv]; */
8312: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8313: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8314: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 8315: }
1.237 brouard 8316: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8317: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8318: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8319: }
1.264 brouard 8320: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8321: fprintf(ficgp,"\n#\n");
1.223 brouard 8322: if(invalidvarcomb[k1]){
1.227 brouard 8323: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8324: continue;
1.223 brouard 8325: }
1.227 brouard 8326:
1.241 brouard 8327: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8328: 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 8329: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8330: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8331: k=3; /* Offset */
1.255 brouard 8332: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8333: if(i==1)
8334: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8335: else
8336: fprintf(ficgp,", '' ");
1.255 brouard 8337: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8338: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8339: for (j=2; j<= nlstate ; j ++)
8340: fprintf(ficgp,"+$%d",k+l+j-1);
8341: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8342: } /* nlstate */
1.264 brouard 8343: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8344: } /* end cpt state*/
8345: } /* end covariate */
1.227 brouard 8346:
8347:
1.220 brouard 8348: /* 7eme */
1.296 brouard 8349: if(prevbcast == 1){
1.288 brouard 8350: /* CV backward prevalence for each covariate */
1.237 brouard 8351: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8352: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8353: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8354: continue;
1.268 brouard 8355: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8356: strcpy(gplotlabel,"(");
1.288 brouard 8357: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8358: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8359: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8360: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8361: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8362: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 8363: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8364: /* vlv= nbcode[Tvaraff[k]][lv]; */
8365: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8366: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8367: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8368: }
1.237 brouard 8369: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8370: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8371: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8372: }
1.264 brouard 8373: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8374: fprintf(ficgp,"\n#\n");
8375: if(invalidvarcomb[k1]){
8376: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8377: continue;
8378: }
8379:
1.241 brouard 8380: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8381: 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 8382: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8383: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8384: k=3; /* Offset */
1.268 brouard 8385: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8386: if(i==1)
8387: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8388: else
8389: fprintf(ficgp,", '' ");
8390: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8391: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8392: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8393: /* 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 8394: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8395: /* for (j=2; j<= nlstate ; j ++) */
8396: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8397: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8398: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8399: } /* nlstate */
1.264 brouard 8400: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8401: } /* end cpt state*/
8402: } /* end covariate */
1.296 brouard 8403: } /* End if prevbcast */
1.218 brouard 8404:
1.223 brouard 8405: /* 8eme */
1.218 brouard 8406: if(prevfcast==1){
1.288 brouard 8407: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8408:
1.237 brouard 8409: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8410: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8411: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8412: continue;
1.211 brouard 8413: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8414: strcpy(gplotlabel,"(");
1.288 brouard 8415: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8416: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
1.332 brouard 8417: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8418: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8419: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8420: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8421: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8422: /* vlv= nbcode[Tvaraff[k]][lv]; */
8423: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8424: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8425: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8426: }
1.237 brouard 8427: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8428: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8429: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8430: }
1.264 brouard 8431: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8432: fprintf(ficgp,"\n#\n");
8433: if(invalidvarcomb[k1]){
8434: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8435: continue;
8436: }
8437:
8438: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8439: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8440: 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 8441: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8442: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8443:
8444: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8445: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8446: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8447: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8448: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8449: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8450: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8451: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8452: if(i==istart){
1.227 brouard 8453: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8454: }else{
8455: fprintf(ficgp,",\\\n '' ");
8456: }
8457: if(cptcoveff ==0){ /* No covariate */
8458: ioffset=2; /* Age is in 2 */
8459: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8460: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8461: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8462: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8463: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8464: if(i==nlstate+1){
1.270 brouard 8465: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8466: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8467: fprintf(ficgp,",\\\n '' ");
8468: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8469: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8470: offyear, \
1.268 brouard 8471: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8472: }else
1.227 brouard 8473: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8474: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8475: }else{ /* more than 2 covariates */
1.270 brouard 8476: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8477: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8478: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8479: iyearc=ioffset-1;
8480: iagec=ioffset;
1.227 brouard 8481: fprintf(ficgp," u %d:(",ioffset);
8482: kl=0;
8483: strcpy(gplotcondition,"(");
8484: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8485: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8486: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8487: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8488: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8489: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8490: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8491: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8492: kl++;
8493: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8494: kl++;
8495: if(k <cptcoveff && cptcoveff>1)
8496: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8497: }
8498: strcpy(gplotcondition+strlen(gplotcondition),")");
8499: /* 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 *\/ */
8500: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8501: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8502: /* '' 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*/
8503: if(i==nlstate+1){
1.270 brouard 8504: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8505: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8506: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8507: fprintf(ficgp," u %d:(",iagec);
8508: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8509: iyearc, iagec, offyear, \
8510: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8511: /* '' 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 8512: }else{
8513: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8514: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8515: }
8516: } /* end if covariate */
8517: } /* nlstate */
1.264 brouard 8518: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8519: } /* end cpt state*/
8520: } /* end covariate */
8521: } /* End if prevfcast */
1.227 brouard 8522:
1.296 brouard 8523: if(prevbcast==1){
1.268 brouard 8524: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8525:
8526: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8527: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8528: if(m != 1 && TKresult[nres]!= k1)
8529: continue;
8530: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8531: strcpy(gplotlabel,"(");
8532: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
8533: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
1.332 brouard 8534: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8535: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.268 brouard 8536: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8537: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8538: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8539: /* vlv= nbcode[Tvaraff[k]][lv]; */
8540: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.268 brouard 8541: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8542: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8543: }
8544: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8545: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8546: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.268 brouard 8547: }
8548: strcpy(gplotlabel+strlen(gplotlabel),")");
8549: fprintf(ficgp,"\n#\n");
8550: if(invalidvarcomb[k1]){
8551: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8552: continue;
8553: }
8554:
8555: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8556: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8557: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8558: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8559: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8560:
8561: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8562: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8563: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8564: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8565: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8566: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8567: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8568: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8569: if(i==istart){
8570: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8571: }else{
8572: fprintf(ficgp,",\\\n '' ");
8573: }
8574: if(cptcoveff ==0){ /* No covariate */
8575: ioffset=2; /* Age is in 2 */
8576: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8577: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8578: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8579: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8580: fprintf(ficgp," u %d:(", ioffset);
8581: if(i==nlstate+1){
1.270 brouard 8582: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8583: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8584: fprintf(ficgp,",\\\n '' ");
8585: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8586: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8587: offbyear, \
8588: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8589: }else
8590: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8591: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8592: }else{ /* more than 2 covariates */
1.270 brouard 8593: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8594: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8595: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8596: iyearc=ioffset-1;
8597: iagec=ioffset;
1.268 brouard 8598: fprintf(ficgp," u %d:(",ioffset);
8599: kl=0;
8600: strcpy(gplotcondition,"(");
8601: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8602: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8603: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.268 brouard 8604: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8605: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8606: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8607: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8608: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.268 brouard 8609: kl++;
8610: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8611: kl++;
8612: if(k <cptcoveff && cptcoveff>1)
8613: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8614: }
8615: strcpy(gplotcondition+strlen(gplotcondition),")");
8616: /* 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 *\/ */
8617: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8618: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8619: /* '' 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*/
8620: if(i==nlstate+1){
1.270 brouard 8621: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8622: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8623: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8624: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8625: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8626: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8627: iyearc,iagec,offbyear, \
8628: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8629: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8630: }else{
8631: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8632: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8633: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8634: }
8635: } /* end if covariate */
8636: } /* nlstate */
8637: fprintf(ficgp,"\nset out; unset label;\n");
8638: } /* end cpt state*/
8639: } /* end covariate */
1.296 brouard 8640: } /* End if prevbcast */
1.268 brouard 8641:
1.227 brouard 8642:
1.238 brouard 8643: /* 9eme writing MLE parameters */
8644: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8645: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8646: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8647: for(k=1; k <=(nlstate+ndeath); k++){
8648: if (k != i) {
1.227 brouard 8649: fprintf(ficgp,"# current state %d\n",k);
8650: for(j=1; j <=ncovmodel; j++){
8651: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8652: jk++;
8653: }
8654: fprintf(ficgp,"\n");
1.126 brouard 8655: }
8656: }
1.223 brouard 8657: }
1.187 brouard 8658: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8659:
1.145 brouard 8660: /*goto avoid;*/
1.238 brouard 8661: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8662: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8663: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8664: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8665: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8666: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8667: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8668: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8669: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8670: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8671: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8672: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8673: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8674: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8675: fprintf(ficgp,"#\n");
1.223 brouard 8676: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8677: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8678: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8679: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8680: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8681: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8682: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8683: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8684: continue;
1.264 brouard 8685: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8686: strcpy(gplotlabel,"(");
1.276 brouard 8687: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8688: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
1.332 brouard 8689: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8690: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.264 brouard 8691: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8692: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8693: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8694: /* vlv= nbcode[Tvaraff[k]][lv]; */
8695: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.264 brouard 8696: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8697: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8698: }
1.237 brouard 8699: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8700: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8701: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8702: }
1.264 brouard 8703: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8704: fprintf(ficgp,"\n#\n");
1.264 brouard 8705: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8706: fprintf(ficgp,"\nset key outside ");
8707: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8708: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8709: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8710: if (ng==1){
8711: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8712: fprintf(ficgp,"\nunset log y");
8713: }else if (ng==2){
8714: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8715: fprintf(ficgp,"\nset log y");
8716: }else if (ng==3){
8717: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8718: fprintf(ficgp,"\nset log y");
8719: }else
8720: fprintf(ficgp,"\nunset title ");
8721: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8722: i=1;
8723: for(k2=1; k2<=nlstate; k2++) {
8724: k3=i;
8725: for(k=1; k<=(nlstate+ndeath); k++) {
8726: if (k != k2){
8727: switch( ng) {
8728: case 1:
8729: if(nagesqr==0)
8730: fprintf(ficgp," p%d+p%d*x",i,i+1);
8731: else /* nagesqr =1 */
8732: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8733: break;
8734: case 2: /* ng=2 */
8735: if(nagesqr==0)
8736: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8737: else /* nagesqr =1 */
8738: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8739: break;
8740: case 3:
8741: if(nagesqr==0)
8742: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8743: else /* nagesqr =1 */
8744: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8745: break;
8746: }
8747: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8748: ijp=1; /* product no age */
8749: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8750: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8751: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 8752: switch(Typevar[j]){
8753: case 1:
8754: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8755: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
8756: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8757: if(DummyV[j]==0){/* Bug valgrind */
8758: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8759: }else{ /* quantitative */
8760: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8761: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8762: }
8763: ij++;
1.268 brouard 8764: }
1.237 brouard 8765: }
1.329 brouard 8766: }
8767: break;
8768: case 2:
8769: if(cptcovprod >0){
8770: if(j==Tprod[ijp]) { /* */
8771: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8772: if(ijp <=cptcovprod) { /* Product */
8773: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8774: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8775: /* 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)]); */
8776: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8777: }else{ /* Vn is dummy and Vm is quanti */
8778: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8779: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8780: }
8781: }else{ /* Vn*Vm Vn is quanti */
8782: if(DummyV[Tvard[ijp][2]]==0){
8783: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8784: }else{ /* Both quanti */
8785: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8786: }
1.268 brouard 8787: }
1.329 brouard 8788: ijp++;
1.237 brouard 8789: }
1.329 brouard 8790: } /* end Tprod */
8791: }
8792: break;
8793: case 0:
8794: /* simple covariate */
1.264 brouard 8795: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8796: if(Dummy[j]==0){
8797: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8798: }else{ /* quantitative */
8799: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8800: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8801: }
1.329 brouard 8802: /* end simple */
8803: break;
8804: default:
8805: break;
8806: } /* end switch */
1.237 brouard 8807: } /* end j */
1.329 brouard 8808: }else{ /* k=k2 */
8809: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
8810: fprintf(ficgp," (1.");i=i-ncovmodel;
8811: }else
8812: i=i-ncovmodel;
1.223 brouard 8813: }
1.227 brouard 8814:
1.223 brouard 8815: if(ng != 1){
8816: fprintf(ficgp,")/(1");
1.227 brouard 8817:
1.264 brouard 8818: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8819: if(nagesqr==0)
1.264 brouard 8820: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8821: else /* nagesqr =1 */
1.264 brouard 8822: 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 8823:
1.223 brouard 8824: ij=1;
1.329 brouard 8825: ijp=1;
8826: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
8827: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
8828: switch(Typevar[j]){
8829: case 1:
8830: if(cptcovage >0){
8831: if(j==Tage[ij]) { /* Bug valgrind */
8832: if(ij <=cptcovage) { /* Bug valgrind */
8833: if(DummyV[j]==0){/* Bug valgrind */
8834: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
8835: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
8836: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
8837: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
8838: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8839: }else{ /* quantitative */
8840: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8841: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8842: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8843: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8844: }
8845: ij++;
8846: }
8847: }
8848: }
8849: break;
8850: case 2:
8851: if(cptcovprod >0){
8852: if(j==Tprod[ijp]) { /* */
8853: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8854: if(ijp <=cptcovprod) { /* Product */
8855: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8856: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8857: /* 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)]); */
8858: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8859: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
8860: }else{ /* Vn is dummy and Vm is quanti */
8861: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8862: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8863: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8864: }
8865: }else{ /* Vn*Vm Vn is quanti */
8866: if(DummyV[Tvard[ijp][2]]==0){
8867: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8868: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
8869: }else{ /* Both quanti */
8870: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8871: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8872: }
8873: }
8874: ijp++;
8875: }
8876: } /* end Tprod */
8877: } /* end if */
8878: break;
8879: case 0:
8880: /* simple covariate */
8881: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
8882: if(Dummy[j]==0){
8883: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8884: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
8885: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8886: }else{ /* quantitative */
8887: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
8888: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
8889: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8890: }
8891: /* end simple */
8892: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
8893: break;
8894: default:
8895: break;
8896: } /* end switch */
1.223 brouard 8897: }
8898: fprintf(ficgp,")");
8899: }
8900: fprintf(ficgp,")");
8901: if(ng ==2)
1.276 brouard 8902: 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 8903: else /* ng= 3 */
1.276 brouard 8904: 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 8905: }else{ /* end ng <> 1 */
1.223 brouard 8906: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8907: 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 8908: }
8909: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8910: fprintf(ficgp,",");
8911: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8912: fprintf(ficgp,",");
8913: i=i+ncovmodel;
8914: } /* end k */
8915: } /* end k2 */
1.276 brouard 8916: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8917: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8918: } /* end k1 */
1.223 brouard 8919: } /* end ng */
8920: /* avoid: */
8921: fflush(ficgp);
1.126 brouard 8922: } /* end gnuplot */
8923:
8924:
8925: /*************** Moving average **************/
1.219 brouard 8926: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8927: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8928:
1.222 brouard 8929: int i, cpt, cptcod;
8930: int modcovmax =1;
8931: int mobilavrange, mob;
8932: int iage=0;
1.288 brouard 8933: int firstA1=0, firstA2=0;
1.222 brouard 8934:
1.266 brouard 8935: double sum=0., sumr=0.;
1.222 brouard 8936: double age;
1.266 brouard 8937: double *sumnewp, *sumnewm, *sumnewmr;
8938: double *agemingood, *agemaxgood;
8939: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8940:
8941:
1.278 brouard 8942: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8943: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8944:
8945: sumnewp = vector(1,ncovcombmax);
8946: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8947: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8948: agemingood = vector(1,ncovcombmax);
1.266 brouard 8949: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8950: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8951: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8952:
8953: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8954: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8955: sumnewp[cptcod]=0.;
1.266 brouard 8956: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8957: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8958: }
8959: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8960:
1.266 brouard 8961: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8962: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8963: else mobilavrange=mobilav;
8964: for (age=bage; age<=fage; age++)
8965: for (i=1; i<=nlstate;i++)
8966: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8967: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8968: /* We keep the original values on the extreme ages bage, fage and for
8969: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8970: we use a 5 terms etc. until the borders are no more concerned.
8971: */
8972: for (mob=3;mob <=mobilavrange;mob=mob+2){
8973: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8974: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8975: sumnewm[cptcod]=0.;
8976: for (i=1; i<=nlstate;i++){
1.222 brouard 8977: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8978: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8979: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8980: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8981: }
8982: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8983: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8984: } /* end i */
8985: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8986: } /* end cptcod */
1.222 brouard 8987: }/* end age */
8988: }/* end mob */
1.266 brouard 8989: }else{
8990: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8991: return -1;
1.266 brouard 8992: }
8993:
8994: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8995: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8996: if(invalidvarcomb[cptcod]){
8997: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8998: continue;
8999: }
1.219 brouard 9000:
1.266 brouard 9001: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
9002: sumnewm[cptcod]=0.;
9003: sumnewmr[cptcod]=0.;
9004: for (i=1; i<=nlstate;i++){
9005: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9006: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9007: }
9008: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9009: agemingoodr[cptcod]=age;
9010: }
9011: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9012: agemingood[cptcod]=age;
9013: }
9014: } /* age */
9015: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 9016: sumnewm[cptcod]=0.;
1.266 brouard 9017: sumnewmr[cptcod]=0.;
1.222 brouard 9018: for (i=1; i<=nlstate;i++){
9019: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9020: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9021: }
9022: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9023: agemaxgoodr[cptcod]=age;
1.222 brouard 9024: }
9025: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 9026: agemaxgood[cptcod]=age;
9027: }
9028: } /* age */
9029: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
9030: /* but they will change */
1.288 brouard 9031: firstA1=0;firstA2=0;
1.266 brouard 9032: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
9033: sumnewm[cptcod]=0.;
9034: sumnewmr[cptcod]=0.;
9035: for (i=1; i<=nlstate;i++){
9036: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9037: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9038: }
9039: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9040: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9041: agemaxgoodr[cptcod]=age; /* age min */
9042: for (i=1; i<=nlstate;i++)
9043: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9044: }else{ /* bad we change the value with the values of good ages */
9045: for (i=1; i<=nlstate;i++){
9046: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
9047: } /* i */
9048: } /* end bad */
9049: }else{
9050: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9051: agemaxgood[cptcod]=age;
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)agemaxgood[cptcod]][i][cptcod];
9055: } /* i */
9056: } /* end bad */
9057: }/* end else */
9058: sum=0.;sumr=0.;
9059: for (i=1; i<=nlstate;i++){
9060: sum+=mobaverage[(int)age][i][cptcod];
9061: sumr+=probs[(int)age][i][cptcod];
9062: }
9063: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 9064: if(!firstA1){
9065: firstA1=1;
9066: 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);
9067: }
9068: 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 9069: } /* end bad */
9070: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9071: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 9072: if(!firstA2){
9073: firstA2=1;
9074: 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);
9075: }
9076: 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 9077: } /* end bad */
9078: }/* age */
1.266 brouard 9079:
9080: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 9081: sumnewm[cptcod]=0.;
1.266 brouard 9082: sumnewmr[cptcod]=0.;
1.222 brouard 9083: for (i=1; i<=nlstate;i++){
9084: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9085: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9086: }
9087: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9088: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
9089: agemingoodr[cptcod]=age;
9090: for (i=1; i<=nlstate;i++)
9091: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9092: }else{ /* bad we change the value with the values of good ages */
9093: for (i=1; i<=nlstate;i++){
9094: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
9095: } /* i */
9096: } /* end bad */
9097: }else{
9098: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9099: agemingood[cptcod]=age;
9100: }else{ /* bad */
9101: for (i=1; i<=nlstate;i++){
9102: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
9103: } /* i */
9104: } /* end bad */
9105: }/* end else */
9106: sum=0.;sumr=0.;
9107: for (i=1; i<=nlstate;i++){
9108: sum+=mobaverage[(int)age][i][cptcod];
9109: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 9110: }
1.266 brouard 9111: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 9112: 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 9113: } /* end bad */
9114: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9115: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 9116: 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 9117: } /* end bad */
9118: }/* age */
1.266 brouard 9119:
1.222 brouard 9120:
9121: for (age=bage; age<=fage; age++){
1.235 brouard 9122: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 9123: sumnewp[cptcod]=0.;
9124: sumnewm[cptcod]=0.;
9125: for (i=1; i<=nlstate;i++){
9126: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
9127: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9128: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
9129: }
9130: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
9131: }
9132: /* printf("\n"); */
9133: /* } */
1.266 brouard 9134:
1.222 brouard 9135: /* brutal averaging */
1.266 brouard 9136: /* for (i=1; i<=nlstate;i++){ */
9137: /* for (age=1; age<=bage; age++){ */
9138: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
9139: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9140: /* } */
9141: /* for (age=fage; age<=AGESUP; age++){ */
9142: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
9143: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9144: /* } */
9145: /* } /\* end i status *\/ */
9146: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
9147: /* for (age=1; age<=AGESUP; age++){ */
9148: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
9149: /* mobaverage[(int)age][i][cptcod]=0.; */
9150: /* } */
9151: /* } */
1.222 brouard 9152: }/* end cptcod */
1.266 brouard 9153: free_vector(agemaxgoodr,1, ncovcombmax);
9154: free_vector(agemaxgood,1, ncovcombmax);
9155: free_vector(agemingood,1, ncovcombmax);
9156: free_vector(agemingoodr,1, ncovcombmax);
9157: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 9158: free_vector(sumnewm,1, ncovcombmax);
9159: free_vector(sumnewp,1, ncovcombmax);
9160: return 0;
9161: }/* End movingaverage */
1.218 brouard 9162:
1.126 brouard 9163:
1.296 brouard 9164:
1.126 brouard 9165: /************** Forecasting ******************/
1.296 brouard 9166: /* 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)*/
9167: 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){
9168: /* dateintemean, mean date of interviews
9169: dateprojd, year, month, day of starting projection
9170: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 9171: agemin, agemax range of age
9172: dateprev1 dateprev2 range of dates during which prevalence is computed
9173: */
1.296 brouard 9174: /* double anprojd, mprojd, jprojd; */
9175: /* double anprojf, mprojf, jprojf; */
1.267 brouard 9176: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 9177: double agec; /* generic age */
1.296 brouard 9178: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 9179: double *popeffectif,*popcount;
9180: double ***p3mat;
1.218 brouard 9181: /* double ***mobaverage; */
1.126 brouard 9182: char fileresf[FILENAMELENGTH];
9183:
9184: agelim=AGESUP;
1.211 brouard 9185: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9186: in each health status at the date of interview (if between dateprev1 and dateprev2).
9187: We still use firstpass and lastpass as another selection.
9188: */
1.214 brouard 9189: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9190: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 9191:
1.201 brouard 9192: strcpy(fileresf,"F_");
9193: strcat(fileresf,fileresu);
1.126 brouard 9194: if((ficresf=fopen(fileresf,"w"))==NULL) {
9195: printf("Problem with forecast resultfile: %s\n", fileresf);
9196: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
9197: }
1.235 brouard 9198: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
9199: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 9200:
1.225 brouard 9201: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 9202:
9203:
9204: stepsize=(int) (stepm+YEARM-1)/YEARM;
9205: if (stepm<=12) stepsize=1;
9206: if(estepm < stepm){
9207: printf ("Problem %d lower than %d\n",estepm, stepm);
9208: }
1.270 brouard 9209: else{
9210: hstepm=estepm;
9211: }
9212: if(estepm > stepm){ /* Yes every two year */
9213: stepsize=2;
9214: }
1.296 brouard 9215: hstepm=hstepm/stepm;
1.126 brouard 9216:
1.296 brouard 9217:
9218: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9219: /* fractional in yp1 *\/ */
9220: /* aintmean=yp; */
9221: /* yp2=modf((yp1*12),&yp); */
9222: /* mintmean=yp; */
9223: /* yp1=modf((yp2*30.5),&yp); */
9224: /* jintmean=yp; */
9225: /* if(jintmean==0) jintmean=1; */
9226: /* if(mintmean==0) mintmean=1; */
1.126 brouard 9227:
1.296 brouard 9228:
9229: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
9230: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
9231: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 9232: i1=pow(2,cptcoveff);
1.126 brouard 9233: if (cptcovn < 1){i1=1;}
9234:
1.296 brouard 9235: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 9236:
9237: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 9238:
1.126 brouard 9239: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 9240: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9241: 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 9242: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9243: continue;
1.227 brouard 9244: if(invalidvarcomb[k]){
9245: printf("\nCombination (%d) projection ignored because no cases \n",k);
9246: continue;
9247: }
9248: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
9249: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9250: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); */
9251: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 9252: }
1.235 brouard 9253: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 9254: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 9255: }
1.227 brouard 9256: fprintf(ficresf," yearproj age");
9257: for(j=1; j<=nlstate+ndeath;j++){
9258: for(i=1; i<=nlstate;i++)
9259: fprintf(ficresf," p%d%d",i,j);
9260: fprintf(ficresf," wp.%d",j);
9261: }
1.296 brouard 9262: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 9263: fprintf(ficresf,"\n");
1.296 brouard 9264: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 9265: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
9266: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 9267: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
9268: nhstepm = nhstepm/hstepm;
9269: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9270: oldm=oldms;savm=savms;
1.268 brouard 9271: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 9272: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 9273: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 9274: for (h=0; h<=nhstepm; h++){
9275: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 9276: break;
9277: }
9278: }
9279: fprintf(ficresf,"\n");
9280: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9281: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
9282: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* TnsdVar[Tvaraff] correct */
1.296 brouard 9283: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9284:
9285: for(j=1; j<=nlstate+ndeath;j++) {
9286: ppij=0.;
9287: for(i=1; i<=nlstate;i++) {
1.278 brouard 9288: if (mobilav>=1)
9289: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9290: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9291: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9292: }
1.268 brouard 9293: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9294: } /* end i */
9295: fprintf(ficresf," %.3f", ppij);
9296: }/* end j */
1.227 brouard 9297: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9298: } /* end agec */
1.266 brouard 9299: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9300: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9301: } /* end yearp */
9302: } /* end k */
1.219 brouard 9303:
1.126 brouard 9304: fclose(ficresf);
1.215 brouard 9305: printf("End of Computing forecasting \n");
9306: fprintf(ficlog,"End of Computing forecasting\n");
9307:
1.126 brouard 9308: }
9309:
1.269 brouard 9310: /************** Back Forecasting ******************/
1.296 brouard 9311: /* 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){ */
9312: 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){
9313: /* back1, year, month, day of starting backprojection
1.267 brouard 9314: agemin, agemax range of age
9315: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9316: anback2 year of end of backprojection (same day and month as back1).
9317: prevacurrent and prev are prevalences.
1.267 brouard 9318: */
9319: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9320: double agec; /* generic age */
1.302 brouard 9321: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9322: double *popeffectif,*popcount;
9323: double ***p3mat;
9324: /* double ***mobaverage; */
9325: char fileresfb[FILENAMELENGTH];
9326:
1.268 brouard 9327: agelim=AGEINF;
1.267 brouard 9328: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9329: in each health status at the date of interview (if between dateprev1 and dateprev2).
9330: We still use firstpass and lastpass as another selection.
9331: */
9332: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9333: /* firstpass, lastpass, stepm, weightopt, model); */
9334:
9335: /*Do we need to compute prevalence again?*/
9336:
9337: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9338:
9339: strcpy(fileresfb,"FB_");
9340: strcat(fileresfb,fileresu);
9341: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9342: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9343: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9344: }
9345: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9346: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9347:
9348: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9349:
9350:
9351: stepsize=(int) (stepm+YEARM-1)/YEARM;
9352: if (stepm<=12) stepsize=1;
9353: if(estepm < stepm){
9354: printf ("Problem %d lower than %d\n",estepm, stepm);
9355: }
1.270 brouard 9356: else{
9357: hstepm=estepm;
9358: }
9359: if(estepm >= stepm){ /* Yes every two year */
9360: stepsize=2;
9361: }
1.267 brouard 9362:
9363: hstepm=hstepm/stepm;
1.296 brouard 9364: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9365: /* fractional in yp1 *\/ */
9366: /* aintmean=yp; */
9367: /* yp2=modf((yp1*12),&yp); */
9368: /* mintmean=yp; */
9369: /* yp1=modf((yp2*30.5),&yp); */
9370: /* jintmean=yp; */
9371: /* if(jintmean==0) jintmean=1; */
9372: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9373:
9374: i1=pow(2,cptcoveff);
9375: if (cptcovn < 1){i1=1;}
9376:
1.296 brouard 9377: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9378: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9379:
9380: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9381:
9382: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9383: for(k=1; k<=i1;k++){
9384: if(i1 != 1 && TKresult[nres]!= k)
9385: continue;
9386: if(invalidvarcomb[k]){
9387: printf("\nCombination (%d) projection ignored because no cases \n",k);
9388: continue;
9389: }
1.268 brouard 9390: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9391: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9392: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.267 brouard 9393: }
9394: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9395: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9396: }
9397: fprintf(ficresfb," yearbproj age");
9398: for(j=1; j<=nlstate+ndeath;j++){
9399: for(i=1; i<=nlstate;i++)
1.268 brouard 9400: fprintf(ficresfb," b%d%d",i,j);
9401: fprintf(ficresfb," b.%d",j);
1.267 brouard 9402: }
1.296 brouard 9403: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9404: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9405: fprintf(ficresfb,"\n");
1.296 brouard 9406: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9407: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9408: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9409: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9410: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9411: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9412: nhstepm = nhstepm/hstepm;
9413: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9414: oldm=oldms;savm=savms;
1.268 brouard 9415: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9416: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9417: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9418: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9419: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9420: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9421: for (h=0; h<=nhstepm; h++){
1.268 brouard 9422: if (h*hstepm/YEARM*stepm ==-yearp) {
9423: break;
9424: }
9425: }
9426: fprintf(ficresfb,"\n");
9427: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9428: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.296 brouard 9429: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9430: for(i=1; i<=nlstate+ndeath;i++) {
9431: ppij=0.;ppi=0.;
9432: for(j=1; j<=nlstate;j++) {
9433: /* if (mobilav==1) */
1.269 brouard 9434: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9435: ppi=ppi+prevacurrent[(int)agec][j][k];
9436: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9437: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9438: /* else { */
9439: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9440: /* } */
1.268 brouard 9441: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9442: } /* end j */
9443: if(ppi <0.99){
9444: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9445: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9446: }
9447: fprintf(ficresfb," %.3f", ppij);
9448: }/* end j */
1.267 brouard 9449: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9450: } /* end agec */
9451: } /* end yearp */
9452: } /* end k */
1.217 brouard 9453:
1.267 brouard 9454: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9455:
1.267 brouard 9456: fclose(ficresfb);
9457: printf("End of Computing Back forecasting \n");
9458: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9459:
1.267 brouard 9460: }
1.217 brouard 9461:
1.269 brouard 9462: /* Variance of prevalence limit: varprlim */
9463: 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 9464: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9465:
9466: char fileresvpl[FILENAMELENGTH];
9467: FILE *ficresvpl;
9468: double **oldm, **savm;
9469: double **varpl; /* Variances of prevalence limits by age */
9470: int i1, k, nres, j ;
9471:
9472: strcpy(fileresvpl,"VPL_");
9473: strcat(fileresvpl,fileresu);
9474: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9475: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9476: exit(0);
9477: }
1.288 brouard 9478: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9479: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9480:
9481: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9482: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9483:
9484: i1=pow(2,cptcoveff);
9485: if (cptcovn < 1){i1=1;}
9486:
9487: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9488: for(k=1; k<=i1;k++){ /* We find the combination equivalent to result line values of dummies */
1.269 brouard 9489: if(i1 != 1 && TKresult[nres]!= k)
9490: continue;
9491: fprintf(ficresvpl,"\n#****** ");
9492: printf("\n#****** ");
9493: fprintf(ficlog,"\n#****** ");
9494: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9495: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9496: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9497: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.269 brouard 9498: }
9499: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 9500: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9501: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9502: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.269 brouard 9503: }
9504: fprintf(ficresvpl,"******\n");
9505: printf("******\n");
9506: fprintf(ficlog,"******\n");
9507:
9508: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9509: oldm=oldms;savm=savms;
9510: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9511: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9512: /*}*/
9513: }
9514:
9515: fclose(ficresvpl);
1.288 brouard 9516: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9517: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9518:
9519: }
9520: /* Variance of back prevalence: varbprlim */
9521: 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){
9522: /*------- Variance of back (stable) prevalence------*/
9523:
9524: char fileresvbl[FILENAMELENGTH];
9525: FILE *ficresvbl;
9526:
9527: double **oldm, **savm;
9528: double **varbpl; /* Variances of back prevalence limits by age */
9529: int i1, k, nres, j ;
9530:
9531: strcpy(fileresvbl,"VBL_");
9532: strcat(fileresvbl,fileresu);
9533: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9534: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9535: exit(0);
9536: }
9537: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9538: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9539:
9540:
9541: i1=pow(2,cptcoveff);
9542: if (cptcovn < 1){i1=1;}
9543:
9544: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9545: for(k=1; k<=i1;k++){
9546: if(i1 != 1 && TKresult[nres]!= k)
9547: continue;
9548: fprintf(ficresvbl,"\n#****** ");
9549: printf("\n#****** ");
9550: fprintf(ficlog,"\n#****** ");
9551: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9552: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9553: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9554: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.269 brouard 9555: }
9556: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 9557: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9558: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9559: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.269 brouard 9560: }
9561: fprintf(ficresvbl,"******\n");
9562: printf("******\n");
9563: fprintf(ficlog,"******\n");
9564:
9565: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9566: oldm=oldms;savm=savms;
9567:
9568: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9569: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9570: /*}*/
9571: }
9572:
9573: fclose(ficresvbl);
9574: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9575: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9576:
9577: } /* End of varbprlim */
9578:
1.126 brouard 9579: /************** Forecasting *****not tested NB*************/
1.227 brouard 9580: /* 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 9581:
1.227 brouard 9582: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9583: /* int *popage; */
9584: /* double calagedatem, agelim, kk1, kk2; */
9585: /* double *popeffectif,*popcount; */
9586: /* double ***p3mat,***tabpop,***tabpopprev; */
9587: /* /\* double ***mobaverage; *\/ */
9588: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9589:
1.227 brouard 9590: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9591: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9592: /* agelim=AGESUP; */
9593: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9594:
1.227 brouard 9595: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9596:
9597:
1.227 brouard 9598: /* strcpy(filerespop,"POP_"); */
9599: /* strcat(filerespop,fileresu); */
9600: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9601: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9602: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9603: /* } */
9604: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9605: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9606:
1.227 brouard 9607: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9608:
1.227 brouard 9609: /* /\* if (mobilav!=0) { *\/ */
9610: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9611: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9612: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9613: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9614: /* /\* } *\/ */
9615: /* /\* } *\/ */
1.126 brouard 9616:
1.227 brouard 9617: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9618: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9619:
1.227 brouard 9620: /* agelim=AGESUP; */
1.126 brouard 9621:
1.227 brouard 9622: /* hstepm=1; */
9623: /* hstepm=hstepm/stepm; */
1.218 brouard 9624:
1.227 brouard 9625: /* if (popforecast==1) { */
9626: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9627: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9628: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9629: /* } */
9630: /* popage=ivector(0,AGESUP); */
9631: /* popeffectif=vector(0,AGESUP); */
9632: /* popcount=vector(0,AGESUP); */
1.126 brouard 9633:
1.227 brouard 9634: /* i=1; */
9635: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9636:
1.227 brouard 9637: /* imx=i; */
9638: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9639: /* } */
1.218 brouard 9640:
1.227 brouard 9641: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9642: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9643: /* k=k+1; */
9644: /* fprintf(ficrespop,"\n#******"); */
9645: /* for(j=1;j<=cptcoveff;j++) { */
9646: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9647: /* } */
9648: /* fprintf(ficrespop,"******\n"); */
9649: /* fprintf(ficrespop,"# Age"); */
9650: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9651: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9652:
1.227 brouard 9653: /* for (cpt=0; cpt<=0;cpt++) { */
9654: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9655:
1.227 brouard 9656: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9657: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9658: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9659:
1.227 brouard 9660: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9661: /* oldm=oldms;savm=savms; */
9662: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9663:
1.227 brouard 9664: /* for (h=0; h<=nhstepm; h++){ */
9665: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9666: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9667: /* } */
9668: /* for(j=1; j<=nlstate+ndeath;j++) { */
9669: /* kk1=0.;kk2=0; */
9670: /* for(i=1; i<=nlstate;i++) { */
9671: /* if (mobilav==1) */
9672: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9673: /* else { */
9674: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9675: /* } */
9676: /* } */
9677: /* if (h==(int)(calagedatem+12*cpt)){ */
9678: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9679: /* /\*fprintf(ficrespop," %.3f", kk1); */
9680: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9681: /* } */
9682: /* } */
9683: /* for(i=1; i<=nlstate;i++){ */
9684: /* kk1=0.; */
9685: /* for(j=1; j<=nlstate;j++){ */
9686: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9687: /* } */
9688: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9689: /* } */
1.218 brouard 9690:
1.227 brouard 9691: /* if (h==(int)(calagedatem+12*cpt)) */
9692: /* for(j=1; j<=nlstate;j++) */
9693: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9694: /* } */
9695: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9696: /* } */
9697: /* } */
1.218 brouard 9698:
1.227 brouard 9699: /* /\******\/ */
1.218 brouard 9700:
1.227 brouard 9701: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9702: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9703: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9704: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9705: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9706:
1.227 brouard 9707: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9708: /* oldm=oldms;savm=savms; */
9709: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9710: /* for (h=0; h<=nhstepm; h++){ */
9711: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9712: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9713: /* } */
9714: /* for(j=1; j<=nlstate+ndeath;j++) { */
9715: /* kk1=0.;kk2=0; */
9716: /* for(i=1; i<=nlstate;i++) { */
9717: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9718: /* } */
9719: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9720: /* } */
9721: /* } */
9722: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9723: /* } */
9724: /* } */
9725: /* } */
9726: /* } */
1.218 brouard 9727:
1.227 brouard 9728: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9729:
1.227 brouard 9730: /* if (popforecast==1) { */
9731: /* free_ivector(popage,0,AGESUP); */
9732: /* free_vector(popeffectif,0,AGESUP); */
9733: /* free_vector(popcount,0,AGESUP); */
9734: /* } */
9735: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9736: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9737: /* fclose(ficrespop); */
9738: /* } /\* End of popforecast *\/ */
1.218 brouard 9739:
1.126 brouard 9740: int fileappend(FILE *fichier, char *optionfich)
9741: {
9742: if((fichier=fopen(optionfich,"a"))==NULL) {
9743: printf("Problem with file: %s\n", optionfich);
9744: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9745: return (0);
9746: }
9747: fflush(fichier);
9748: return (1);
9749: }
9750:
9751:
9752: /**************** function prwizard **********************/
9753: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9754: {
9755:
9756: /* Wizard to print covariance matrix template */
9757:
1.164 brouard 9758: char ca[32], cb[32];
9759: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9760: int numlinepar;
9761:
9762: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9763: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9764: for(i=1; i <=nlstate; i++){
9765: jj=0;
9766: for(j=1; j <=nlstate+ndeath; j++){
9767: if(j==i) continue;
9768: jj++;
9769: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9770: printf("%1d%1d",i,j);
9771: fprintf(ficparo,"%1d%1d",i,j);
9772: for(k=1; k<=ncovmodel;k++){
9773: /* printf(" %lf",param[i][j][k]); */
9774: /* fprintf(ficparo," %lf",param[i][j][k]); */
9775: printf(" 0.");
9776: fprintf(ficparo," 0.");
9777: }
9778: printf("\n");
9779: fprintf(ficparo,"\n");
9780: }
9781: }
9782: printf("# Scales (for hessian or gradient estimation)\n");
9783: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9784: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9785: for(i=1; i <=nlstate; i++){
9786: jj=0;
9787: for(j=1; j <=nlstate+ndeath; j++){
9788: if(j==i) continue;
9789: jj++;
9790: fprintf(ficparo,"%1d%1d",i,j);
9791: printf("%1d%1d",i,j);
9792: fflush(stdout);
9793: for(k=1; k<=ncovmodel;k++){
9794: /* printf(" %le",delti3[i][j][k]); */
9795: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9796: printf(" 0.");
9797: fprintf(ficparo," 0.");
9798: }
9799: numlinepar++;
9800: printf("\n");
9801: fprintf(ficparo,"\n");
9802: }
9803: }
9804: printf("# Covariance matrix\n");
9805: /* # 121 Var(a12)\n\ */
9806: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9807: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9808: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9809: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9810: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9811: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9812: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9813: fflush(stdout);
9814: fprintf(ficparo,"# Covariance matrix\n");
9815: /* # 121 Var(a12)\n\ */
9816: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9817: /* # ...\n\ */
9818: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9819:
9820: for(itimes=1;itimes<=2;itimes++){
9821: jj=0;
9822: for(i=1; i <=nlstate; i++){
9823: for(j=1; j <=nlstate+ndeath; j++){
9824: if(j==i) continue;
9825: for(k=1; k<=ncovmodel;k++){
9826: jj++;
9827: ca[0]= k+'a'-1;ca[1]='\0';
9828: if(itimes==1){
9829: printf("#%1d%1d%d",i,j,k);
9830: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9831: }else{
9832: printf("%1d%1d%d",i,j,k);
9833: fprintf(ficparo,"%1d%1d%d",i,j,k);
9834: /* printf(" %.5le",matcov[i][j]); */
9835: }
9836: ll=0;
9837: for(li=1;li <=nlstate; li++){
9838: for(lj=1;lj <=nlstate+ndeath; lj++){
9839: if(lj==li) continue;
9840: for(lk=1;lk<=ncovmodel;lk++){
9841: ll++;
9842: if(ll<=jj){
9843: cb[0]= lk +'a'-1;cb[1]='\0';
9844: if(ll<jj){
9845: if(itimes==1){
9846: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9847: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9848: }else{
9849: printf(" 0.");
9850: fprintf(ficparo," 0.");
9851: }
9852: }else{
9853: if(itimes==1){
9854: printf(" Var(%s%1d%1d)",ca,i,j);
9855: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9856: }else{
9857: printf(" 0.");
9858: fprintf(ficparo," 0.");
9859: }
9860: }
9861: }
9862: } /* end lk */
9863: } /* end lj */
9864: } /* end li */
9865: printf("\n");
9866: fprintf(ficparo,"\n");
9867: numlinepar++;
9868: } /* end k*/
9869: } /*end j */
9870: } /* end i */
9871: } /* end itimes */
9872:
9873: } /* end of prwizard */
9874: /******************* Gompertz Likelihood ******************************/
9875: double gompertz(double x[])
9876: {
1.302 brouard 9877: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9878: int i,n=0; /* n is the size of the sample */
9879:
1.220 brouard 9880: for (i=1;i<=imx ; i++) {
1.126 brouard 9881: sump=sump+weight[i];
9882: /* sump=sump+1;*/
9883: num=num+1;
9884: }
1.302 brouard 9885: L=0.0;
9886: /* agegomp=AGEGOMP; */
1.126 brouard 9887: /* for (i=0; i<=imx; i++)
9888: 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]);*/
9889:
1.302 brouard 9890: for (i=1;i<=imx ; i++) {
9891: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9892: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9893: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9894: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9895: * +
9896: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9897: */
9898: if (wav[i] > 1 || agedc[i] < AGESUP) {
9899: if (cens[i] == 1){
9900: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9901: } else if (cens[i] == 0){
1.126 brouard 9902: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9903: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9904: } else
9905: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9906: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9907: L=L+A*weight[i];
1.126 brouard 9908: /* 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 9909: }
9910: }
1.126 brouard 9911:
1.302 brouard 9912: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9913:
9914: return -2*L*num/sump;
9915: }
9916:
1.136 brouard 9917: #ifdef GSL
9918: /******************* Gompertz_f Likelihood ******************************/
9919: double gompertz_f(const gsl_vector *v, void *params)
9920: {
1.302 brouard 9921: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9922: double *x= (double *) v->data;
9923: int i,n=0; /* n is the size of the sample */
9924:
9925: for (i=0;i<=imx-1 ; i++) {
9926: sump=sump+weight[i];
9927: /* sump=sump+1;*/
9928: num=num+1;
9929: }
9930:
9931:
9932: /* for (i=0; i<=imx; i++)
9933: 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]);*/
9934: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9935: for (i=1;i<=imx ; i++)
9936: {
9937: if (cens[i] == 1 && wav[i]>1)
9938: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9939:
9940: if (cens[i] == 0 && wav[i]>1)
9941: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9942: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9943:
9944: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9945: if (wav[i] > 1 ) { /* ??? */
9946: LL=LL+A*weight[i];
9947: /* 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]);*/
9948: }
9949: }
9950:
9951: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9952: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9953:
9954: return -2*LL*num/sump;
9955: }
9956: #endif
9957:
1.126 brouard 9958: /******************* Printing html file ***********/
1.201 brouard 9959: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9960: int lastpass, int stepm, int weightopt, char model[],\
9961: int imx, double p[],double **matcov,double agemortsup){
9962: int i,k;
9963:
9964: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9965: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9966: for (i=1;i<=2;i++)
9967: 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 9968: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9969: fprintf(fichtm,"</ul>");
9970:
9971: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9972:
9973: 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>");
9974:
9975: for (k=agegomp;k<(agemortsup-2);k++)
9976: 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]);
9977:
9978:
9979: fflush(fichtm);
9980: }
9981:
9982: /******************* Gnuplot file **************/
1.201 brouard 9983: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9984:
9985: char dirfileres[132],optfileres[132];
1.164 brouard 9986:
1.126 brouard 9987: int ng;
9988:
9989:
9990: /*#ifdef windows */
9991: fprintf(ficgp,"cd \"%s\" \n",pathc);
9992: /*#endif */
9993:
9994:
9995: strcpy(dirfileres,optionfilefiname);
9996: strcpy(optfileres,"vpl");
1.199 brouard 9997: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9998: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9999: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 10000: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 10001: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
10002:
10003: }
10004:
1.136 brouard 10005: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
10006: {
1.126 brouard 10007:
1.136 brouard 10008: /*-------- data file ----------*/
10009: FILE *fic;
10010: char dummy[]=" ";
1.240 brouard 10011: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 10012: int lstra;
1.136 brouard 10013: int linei, month, year,iout;
1.302 brouard 10014: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 10015: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 10016: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 10017: char *stratrunc;
1.223 brouard 10018:
1.240 brouard 10019: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
10020: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 10021: for(v=1;v<NCOVMAX;v++){
10022: DummyV[v]=0;
10023: FixedV[v]=0;
10024: }
1.126 brouard 10025:
1.240 brouard 10026: for(v=1; v <=ncovcol;v++){
10027: DummyV[v]=0;
10028: FixedV[v]=0;
10029: }
10030: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
10031: DummyV[v]=1;
10032: FixedV[v]=0;
10033: }
10034: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
10035: DummyV[v]=0;
10036: FixedV[v]=1;
10037: }
10038: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
10039: DummyV[v]=1;
10040: FixedV[v]=1;
10041: }
10042: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
10043: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10044: 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]);
10045: }
1.126 brouard 10046:
1.136 brouard 10047: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 10048: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10049: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 10050: }
1.126 brouard 10051:
1.302 brouard 10052: /* Is it a BOM UTF-8 Windows file? */
10053: /* First data line */
10054: linei=0;
10055: while(fgets(line, MAXLINE, fic)) {
10056: noffset=0;
10057: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10058: {
10059: noffset=noffset+3;
10060: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
10061: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
10062: fflush(ficlog); return 1;
10063: }
10064: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
10065: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
10066: {
10067: noffset=noffset+2;
1.304 brouard 10068: 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);
10069: 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 10070: fflush(ficlog); return 1;
10071: }
10072: else if( line[0] == 0 && line[1] == 0)
10073: {
10074: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10075: noffset=noffset+4;
1.304 brouard 10076: 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);
10077: 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 10078: fflush(ficlog); return 1;
10079: }
10080: } else{
10081: ;/*printf(" Not a BOM file\n");*/
10082: }
10083: /* If line starts with a # it is a comment */
10084: if (line[noffset] == '#') {
10085: linei=linei+1;
10086: break;
10087: }else{
10088: break;
10089: }
10090: }
10091: fclose(fic);
10092: if((fic=fopen(datafile,"r"))==NULL) {
10093: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10094: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
10095: }
10096: /* Not a Bom file */
10097:
1.136 brouard 10098: i=1;
10099: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
10100: linei=linei+1;
10101: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
10102: if(line[j] == '\t')
10103: line[j] = ' ';
10104: }
10105: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
10106: ;
10107: };
10108: line[j+1]=0; /* Trims blanks at end of line */
10109: if(line[0]=='#'){
10110: fprintf(ficlog,"Comment line\n%s\n",line);
10111: printf("Comment line\n%s\n",line);
10112: continue;
10113: }
10114: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 10115: strcpy(line, linetmp);
1.223 brouard 10116:
10117: /* Loops on waves */
10118: for (j=maxwav;j>=1;j--){
10119: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 10120: cutv(stra, strb, line, ' ');
10121: if(strb[0]=='.') { /* Missing value */
10122: lval=-1;
10123: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
10124: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
10125: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
10126: 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);
10127: 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);
10128: return 1;
10129: }
10130: }else{
10131: errno=0;
10132: /* what_kind_of_number(strb); */
10133: dval=strtod(strb,&endptr);
10134: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
10135: /* if(strb != endptr && *endptr == '\0') */
10136: /* dval=dlval; */
10137: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10138: if( strb[0]=='\0' || (*endptr != '\0')){
10139: 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);
10140: 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);
10141: return 1;
10142: }
10143: cotqvar[j][iv][i]=dval;
10144: cotvar[j][ntv+iv][i]=dval;
10145: }
10146: strcpy(line,stra);
1.223 brouard 10147: }/* end loop ntqv */
1.225 brouard 10148:
1.223 brouard 10149: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 10150: cutv(stra, strb, line, ' ');
10151: if(strb[0]=='.') { /* Missing value */
10152: lval=-1;
10153: }else{
10154: errno=0;
10155: lval=strtol(strb,&endptr,10);
10156: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10157: if( strb[0]=='\0' || (*endptr != '\0')){
10158: 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);
10159: 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);
10160: return 1;
10161: }
10162: }
10163: if(lval <-1 || lval >1){
10164: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10165: 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 10166: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10167: For example, for multinomial values like 1, 2 and 3,\n \
10168: build V1=0 V2=0 for the reference value (1),\n \
10169: V1=1 V2=0 for (2) \n \
1.223 brouard 10170: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10171: output of IMaCh is often meaningless.\n \
1.319 brouard 10172: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 10173: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10174: 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 10175: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10176: For example, for multinomial values like 1, 2 and 3,\n \
10177: build V1=0 V2=0 for the reference value (1),\n \
10178: V1=1 V2=0 for (2) \n \
1.223 brouard 10179: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10180: output of IMaCh is often meaningless.\n \
1.319 brouard 10181: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 10182: return 1;
10183: }
10184: cotvar[j][iv][i]=(double)(lval);
10185: strcpy(line,stra);
1.223 brouard 10186: }/* end loop ntv */
1.225 brouard 10187:
1.223 brouard 10188: /* Statuses at wave */
1.137 brouard 10189: cutv(stra, strb, line, ' ');
1.223 brouard 10190: if(strb[0]=='.') { /* Missing value */
1.238 brouard 10191: lval=-1;
1.136 brouard 10192: }else{
1.238 brouard 10193: errno=0;
10194: lval=strtol(strb,&endptr,10);
10195: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10196: if( strb[0]=='\0' || (*endptr != '\0')){
10197: 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);
10198: 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);
10199: return 1;
10200: }
1.136 brouard 10201: }
1.225 brouard 10202:
1.136 brouard 10203: s[j][i]=lval;
1.225 brouard 10204:
1.223 brouard 10205: /* Date of Interview */
1.136 brouard 10206: strcpy(line,stra);
10207: cutv(stra, strb,line,' ');
1.169 brouard 10208: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10209: }
1.169 brouard 10210: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 10211: month=99;
10212: year=9999;
1.136 brouard 10213: }else{
1.225 brouard 10214: 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);
10215: 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);
10216: return 1;
1.136 brouard 10217: }
10218: anint[j][i]= (double) year;
1.302 brouard 10219: mint[j][i]= (double)month;
10220: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
10221: /* 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]); */
10222: /* 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]); */
10223: /* } */
1.136 brouard 10224: strcpy(line,stra);
1.223 brouard 10225: } /* End loop on waves */
1.225 brouard 10226:
1.223 brouard 10227: /* Date of death */
1.136 brouard 10228: cutv(stra, strb,line,' ');
1.169 brouard 10229: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10230: }
1.169 brouard 10231: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 10232: month=99;
10233: year=9999;
10234: }else{
1.141 brouard 10235: 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 10236: 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);
10237: return 1;
1.136 brouard 10238: }
10239: andc[i]=(double) year;
10240: moisdc[i]=(double) month;
10241: strcpy(line,stra);
10242:
1.223 brouard 10243: /* Date of birth */
1.136 brouard 10244: cutv(stra, strb,line,' ');
1.169 brouard 10245: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10246: }
1.169 brouard 10247: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 10248: month=99;
10249: year=9999;
10250: }else{
1.141 brouard 10251: 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);
10252: 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 10253: return 1;
1.136 brouard 10254: }
10255: if (year==9999) {
1.141 brouard 10256: 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);
10257: 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 10258: return 1;
10259:
1.136 brouard 10260: }
10261: annais[i]=(double)(year);
1.302 brouard 10262: moisnais[i]=(double)(month);
10263: for (j=1;j<=maxwav;j++){
10264: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
10265: 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]);
10266: 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]);
10267: }
10268: }
10269:
1.136 brouard 10270: strcpy(line,stra);
1.225 brouard 10271:
1.223 brouard 10272: /* Sample weight */
1.136 brouard 10273: cutv(stra, strb,line,' ');
10274: errno=0;
10275: dval=strtod(strb,&endptr);
10276: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 10277: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
10278: 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 10279: fflush(ficlog);
10280: return 1;
10281: }
10282: weight[i]=dval;
10283: strcpy(line,stra);
1.225 brouard 10284:
1.223 brouard 10285: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10286: cutv(stra, strb, line, ' ');
10287: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10288: lval=-1;
1.311 brouard 10289: coqvar[iv][i]=NAN;
10290: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10291: }else{
1.225 brouard 10292: errno=0;
10293: /* what_kind_of_number(strb); */
10294: dval=strtod(strb,&endptr);
10295: /* if(strb != endptr && *endptr == '\0') */
10296: /* dval=dlval; */
10297: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10298: if( strb[0]=='\0' || (*endptr != '\0')){
10299: 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);
10300: 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);
10301: return 1;
10302: }
10303: coqvar[iv][i]=dval;
1.226 brouard 10304: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10305: }
10306: strcpy(line,stra);
10307: }/* end loop nqv */
1.136 brouard 10308:
1.223 brouard 10309: /* Covariate values */
1.136 brouard 10310: for (j=ncovcol;j>=1;j--){
10311: cutv(stra, strb,line,' ');
1.223 brouard 10312: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10313: lval=-1;
1.136 brouard 10314: }else{
1.225 brouard 10315: errno=0;
10316: lval=strtol(strb,&endptr,10);
10317: if( strb[0]=='\0' || (*endptr != '\0')){
10318: 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);
10319: 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);
10320: return 1;
10321: }
1.136 brouard 10322: }
10323: if(lval <-1 || lval >1){
1.225 brouard 10324: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10325: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10326: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10327: For example, for multinomial values like 1, 2 and 3,\n \
10328: build V1=0 V2=0 for the reference value (1),\n \
10329: V1=1 V2=0 for (2) \n \
1.136 brouard 10330: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10331: output of IMaCh is often meaningless.\n \
1.136 brouard 10332: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10333: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10334: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10335: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10336: For example, for multinomial values like 1, 2 and 3,\n \
10337: build V1=0 V2=0 for the reference value (1),\n \
10338: V1=1 V2=0 for (2) \n \
1.136 brouard 10339: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10340: output of IMaCh is often meaningless.\n \
1.136 brouard 10341: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10342: return 1;
1.136 brouard 10343: }
10344: covar[j][i]=(double)(lval);
10345: strcpy(line,stra);
10346: }
10347: lstra=strlen(stra);
1.225 brouard 10348:
1.136 brouard 10349: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10350: stratrunc = &(stra[lstra-9]);
10351: num[i]=atol(stratrunc);
10352: }
10353: else
10354: num[i]=atol(stra);
10355: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10356: 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;}*/
10357:
10358: i=i+1;
10359: } /* End loop reading data */
1.225 brouard 10360:
1.136 brouard 10361: *imax=i-1; /* Number of individuals */
10362: fclose(fic);
1.225 brouard 10363:
1.136 brouard 10364: return (0);
1.164 brouard 10365: /* endread: */
1.225 brouard 10366: printf("Exiting readdata: ");
10367: fclose(fic);
10368: return (1);
1.223 brouard 10369: }
1.126 brouard 10370:
1.234 brouard 10371: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10372: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10373: while (*p2 == ' ')
1.234 brouard 10374: p2++;
10375: /* while ((*p1++ = *p2++) !=0) */
10376: /* ; */
10377: /* do */
10378: /* while (*p2 == ' ') */
10379: /* p2++; */
10380: /* while (*p1++ == *p2++); */
10381: *stri=p2;
1.145 brouard 10382: }
10383:
1.330 brouard 10384: int decoderesult( char resultline[], int nres)
1.230 brouard 10385: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10386: {
1.235 brouard 10387: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10388: char resultsav[MAXLINE];
1.330 brouard 10389: /* int resultmodel[MAXLINE]; */
1.334 brouard 10390: /* int modelresult[MAXLINE]; */
1.230 brouard 10391: char stra[80], strb[80], strc[80], strd[80],stre[80];
10392:
1.234 brouard 10393: removefirstspace(&resultline);
1.332 brouard 10394: printf("decoderesult:%s\n",resultline);
1.230 brouard 10395:
1.332 brouard 10396: strcpy(resultsav,resultline);
10397: printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline);
1.230 brouard 10398: if (strlen(resultsav) >1){
1.334 brouard 10399: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 10400: }
1.253 brouard 10401: if(j == 0){ /* Resultline but no = */
10402: TKresult[nres]=0; /* Combination for the nresult and the model */
10403: return (0);
10404: }
1.234 brouard 10405: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.334 brouard 10406: 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);
10407: 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 10408: /* return 1;*/
1.234 brouard 10409: }
1.334 brouard 10410: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 10411: if(nbocc(resultsav,'=') >1){
1.318 brouard 10412: 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 10413: /* 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 10414: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 10415: /* If a blank, then strc="V4=" and strd='\0' */
10416: if(strc[0]=='\0'){
10417: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
10418: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
10419: return 1;
10420: }
1.234 brouard 10421: }else
10422: cutl(strc,strd,resultsav,'=');
1.318 brouard 10423: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10424:
1.230 brouard 10425: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10426: 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 10427: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10428: /* cptcovsel++; */
10429: if (nbocc(stra,'=') >0)
10430: strcpy(resultsav,stra); /* and analyzes it */
10431: }
1.235 brouard 10432: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10433: /* 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 10434: 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 10435: if(Typevar[k1]==0){ /* Single covariate in model */
10436: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 10437: match=0;
1.318 brouard 10438: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10439: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10440: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10441: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10442: break;
10443: }
10444: }
10445: if(match == 0){
1.332 brouard 10446: 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]);
10447: 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 10448: return 1;
1.234 brouard 10449: }
1.332 brouard 10450: }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*/
10451: /* We feed resultmodel[k1]=k2; */
10452: match=0;
10453: 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 */
10454: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10455: 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 10456: resultmodel[nres][k1]=k2; /* Added here */
10457: printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]);
10458: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10459: break;
10460: }
10461: }
10462: if(match == 0){
10463: 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 10464: 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 10465: return 1;
10466: }
10467: }else if(Typevar[k1]==2){ /* Product No age We want to get the position in the resultline of the product in the model line*/
10468: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
10469: match=0;
10470: 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]);
10471: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10472: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10473: /* modelresult[k2]=k1; */
10474: printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]);
10475: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10476: }
10477: }
10478: if(match == 0){
10479: 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 10480: 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 10481: return 1;
10482: }
10483: match=0;
10484: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10485: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10486: /* modelresult[k2]=k1;*/
10487: printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]);
10488: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10489: break;
10490: }
10491: }
10492: if(match == 0){
10493: 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 10494: 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 10495: return 1;
10496: }
10497: }/* End of testing */
1.333 brouard 10498: }/* End loop cptcovt */
1.235 brouard 10499: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10500: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 10501: 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)
10502: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 10503: match=0;
1.318 brouard 10504: 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 10505: if(Typevar[k1]==0){ /* Single only */
1.237 brouard 10506: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 brouard 10507: 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 10508: 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 10509: ++match;
10510: }
10511: }
10512: }
10513: if(match == 0){
1.332 brouard 10514: printf("Error in result line: variable V%d is missing in model; result: %s, model=%s\n",Tvarsel[k2], resultline, model);
10515: 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 10516: return 1;
1.234 brouard 10517: }else if(match > 1){
10518: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 10519: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
10520: return 1;
1.234 brouard 10521: }
10522: }
1.334 brouard 10523: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 10524: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10525: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 10526: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10527: /* 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*/
10528: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 10529: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10530: /* 1 0 0 0 */
10531: /* 2 1 0 0 */
10532: /* 3 0 1 0 */
1.330 brouard 10533: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 10534: /* 5 0 0 1 */
1.330 brouard 10535: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 10536: /* 7 0 1 1 */
10537: /* 8 1 1 1 */
1.237 brouard 10538: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10539: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10540: /* V5*age V5 known which value for nres? */
10541: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 10542: 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.
10543: * loop on position k1 in the MODEL LINE */
1.331 brouard 10544: /* k counting number of combination of single dummies in the equation model */
10545: /* k4 counting single dummies in the equation model */
10546: /* k4q counting single quantitatives in the equation model */
1.334 brouard 10547: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Dummy and Single, k1 is sorting according to MODEL, but k3 to resultline */
10548: /* 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 10549: /* 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 10550: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 10551: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
10552: /* k3 is the position in the nres result line of the k1th variable of the model equation */
10553: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
10554: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
10555: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 10556: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 10557: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 10558: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 10559: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
10560: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
10561: 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 10562: 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 10563: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 10564: /* Tinvresult[nres][4]=1 */
1.334 brouard 10565: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
10566: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
10567: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10568: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 10569: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 10570: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.332 brouard 10571: 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 10572: k4++;;
1.331 brouard 10573: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 10574: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 10575: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 10576: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 10577: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
10578: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
10579: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 10580: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
10581: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10582: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
10583: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10584: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10585: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 10586: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 10587: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 10588: precov[nres][k1]=Tvalsel[k3q];
10589: 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 10590: k4q++;;
1.331 brouard 10591: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
10592: /* Tvar[k1]; */ /* Age variable */
1.332 brouard 10593: /* Wrong we want the value of variable name Tvar[k1] */
10594:
10595: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
1.331 brouard 10596: 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 10597: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
1.332 brouard 10598: precov[nres][k1]=Tvalsel[k3];
10599: 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 10600: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.332 brouard 10601: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.331 brouard 10602: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 10603: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
1.332 brouard 10604: precov[nres][k1]=Tvalsel[k3q];
1.334 brouard 10605: 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 10606: }else if(Typevar[k1]==2 ){ /* For product quant or dummy (not with age) */
1.332 brouard 10607: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
10608: 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 10609: }else{
1.332 brouard 10610: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
10611: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 10612: }
10613: }
1.234 brouard 10614:
1.334 brouard 10615: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 10616: return (0);
10617: }
1.235 brouard 10618:
1.230 brouard 10619: int decodemodel( char model[], int lastobs)
10620: /**< This routine decodes the model and returns:
1.224 brouard 10621: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10622: * - nagesqr = 1 if age*age in the model, otherwise 0.
10623: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10624: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10625: * - cptcovage number of covariates with age*products =2
10626: * - cptcovs number of simple covariates
10627: * - 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
10628: * which is a new column after the 9 (ncovcol) variables.
1.319 brouard 10629: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10630: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10631: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10632: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10633: */
1.319 brouard 10634: /* 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 10635: {
1.238 brouard 10636: int i, j, k, ks, v;
1.227 brouard 10637: int j1, k1, k2, k3, k4;
1.136 brouard 10638: char modelsav[80];
1.145 brouard 10639: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10640: char *strpt;
1.136 brouard 10641:
1.145 brouard 10642: /*removespace(model);*/
1.136 brouard 10643: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10644: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10645: if (strstr(model,"AGE") !=0){
1.192 brouard 10646: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10647: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10648: return 1;
10649: }
1.141 brouard 10650: if (strstr(model,"v") !=0){
10651: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
10652: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
10653: return 1;
10654: }
1.187 brouard 10655: strcpy(modelsav,model);
10656: if ((strpt=strstr(model,"age*age")) !=0){
10657: printf(" strpt=%s, model=%s\n",strpt, model);
10658: if(strpt != model){
1.234 brouard 10659: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10660: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10661: corresponding column of parameters.\n",model);
1.234 brouard 10662: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10663: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10664: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10665: return 1;
1.225 brouard 10666: }
1.187 brouard 10667: nagesqr=1;
10668: if (strstr(model,"+age*age") !=0)
1.234 brouard 10669: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10670: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10671: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10672: else
1.234 brouard 10673: substrchaine(modelsav, model, "age*age");
1.187 brouard 10674: }else
10675: nagesqr=0;
10676: if (strlen(modelsav) >1){
10677: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10678: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10679: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10680: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10681: * cst, age and age*age
10682: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10683: /* including age products which are counted in cptcovage.
10684: * but the covariates which are products must be treated
10685: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10686: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10687: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10688:
10689:
1.187 brouard 10690: /* Design
10691: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10692: * < ncovcol=8 >
10693: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10694: * k= 1 2 3 4 5 6 7 8
10695: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10696: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10697: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10698: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10699: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10700: * Tage[++cptcovage]=k
10701: * if products, new covar are created after ncovcol with k1
10702: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10703: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10704: * 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
10705: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10706: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10707: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10708: * < ncovcol=8 >
10709: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10710: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10711: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10712: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10713: * p Tprod[1]@2={ 6, 5}
10714: *p Tvard[1][1]@4= {7, 8, 5, 6}
10715: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10716: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10717: *How to reorganize? Tvars(orted)
1.187 brouard 10718: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10719: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
10720: * {2, 1, 4, 8, 5, 6, 3, 7}
10721: * Struct []
10722: */
1.225 brouard 10723:
1.187 brouard 10724: /* This loop fills the array Tvar from the string 'model'.*/
10725: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
10726: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
10727: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
10728: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
10729: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
10730: /* k=1 Tvar[1]=2 (from V2) */
10731: /* k=5 Tvar[5] */
10732: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 10733: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 10734: /* } */
1.198 brouard 10735: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 10736: /*
10737: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 10738: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
10739: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
10740: }
1.187 brouard 10741: cptcovage=0;
1.319 brouard 10742: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
10743: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
10744: 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" */
10745: if (nbocc(modelsav,'+')==0)
10746: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 10747: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
10748: /*scanf("%d",i);*/
1.319 brouard 10749: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
10750: 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 10751: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
10752: /* covar is not filled and then is empty */
10753: cptcovprod--;
10754: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 10755: 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 10756: Typevar[k]=1; /* 1 for age product */
1.319 brouard 10757: cptcovage++; /* Counts the number of covariates which include age as a product */
10758: 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 10759: /*printf("stre=%s ", stre);*/
10760: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
10761: cptcovprod--;
10762: cutl(stre,strb,strc,'V');
10763: Tvar[k]=atoi(stre);
10764: Typevar[k]=1; /* 1 for age product */
10765: cptcovage++;
10766: Tage[cptcovage]=k;
10767: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
10768: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
10769: cptcovn++;
10770: cptcovprodnoage++;k1++;
10771: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10772: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10773: because this model-covariate is a construction we invent a new column
10774: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.335 brouard 10775: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
1.319 brouard 10776: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
10777: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=4 etc */
1.335 brouard 10778: /* Please remark that the new variables are model dependent */
10779: /* If we have 4 variable but the model uses only 3, like in
10780: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
10781: * k= 1 2 3 4 5 6 7 8
10782: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
10783: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
10784: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
10785: */
1.234 brouard 10786: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10787: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10788: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 10789: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 10790: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 brouard 10791: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 10792: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 brouard 10793: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 10794: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10795: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10796: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10797: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10798: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10799: for (i=1; i<=lastobs;i++){
10800: /* Computes the new covariate which is a product of
10801: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10802: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10803: }
10804: } /* End age is not in the model */
10805: } /* End if model includes a product */
1.319 brouard 10806: else { /* not a product */
1.234 brouard 10807: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10808: /* scanf("%d",i);*/
10809: cutl(strd,strc,strb,'V');
10810: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10811: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10812: Tvar[k]=atoi(strd);
10813: Typevar[k]=0; /* 0 for simple covariates */
10814: }
10815: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10816: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10817: scanf("%d",i);*/
1.187 brouard 10818: } /* end of loop + on total covariates */
10819: } /* end if strlen(modelsave == 0) age*age might exist */
10820: } /* end if strlen(model == 0) */
1.136 brouard 10821:
10822: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10823: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10824:
1.136 brouard 10825: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10826: printf("cptcovprod=%d ", cptcovprod);
10827: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10828: scanf("%d ",i);*/
10829:
10830:
1.230 brouard 10831: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10832: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10833: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10834: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10835: k = 1 2 3 4 5 6 7 8 9
10836: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 10837: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 10838: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10839: Dummy[k] 1 0 0 0 3 1 1 2 3
10840: Tmodelind[combination of covar]=k;
1.225 brouard 10841: */
10842: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10843: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10844: /* 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 10845: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 10846: printf("Model=1+age+%s\n\
1.227 brouard 10847: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10848: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10849: 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 10850: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 10851: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10852: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10853: 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 10854: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10855: 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 */
10856: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10857: Fixed[k]= 0;
10858: Dummy[k]= 0;
1.225 brouard 10859: ncoveff++;
1.232 brouard 10860: ncovf++;
1.234 brouard 10861: nsd++;
10862: modell[k].maintype= FTYPE;
10863: TvarsD[nsd]=Tvar[k];
10864: TvarsDind[nsd]=k;
1.330 brouard 10865: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 10866: TvarF[ncovf]=Tvar[k];
10867: TvarFind[ncovf]=k;
10868: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10869: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10870: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10871: Fixed[k]= 0;
10872: Dummy[k]= 0;
10873: ncoveff++;
10874: ncovf++;
10875: modell[k].maintype= FTYPE;
10876: TvarF[ncovf]=Tvar[k];
1.330 brouard 10877: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 10878: TvarFind[ncovf]=k;
1.230 brouard 10879: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10880: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10881: }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 10882: Fixed[k]= 0;
10883: Dummy[k]= 1;
1.230 brouard 10884: nqfveff++;
1.234 brouard 10885: modell[k].maintype= FTYPE;
10886: modell[k].subtype= FQ;
10887: nsq++;
1.334 brouard 10888: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
10889: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 10890: ncovf++;
1.234 brouard 10891: TvarF[ncovf]=Tvar[k];
10892: TvarFind[ncovf]=k;
1.231 brouard 10893: 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 10894: 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 10895: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10896: Fixed[k]= 1;
10897: Dummy[k]= 0;
1.225 brouard 10898: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10899: modell[k].maintype= VTYPE;
10900: modell[k].subtype= VD;
10901: nsd++;
10902: TvarsD[nsd]=Tvar[k];
10903: TvarsDind[nsd]=k;
1.330 brouard 10904: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 10905: ncovv++; /* Only simple time varying variables */
10906: TvarV[ncovv]=Tvar[k];
1.242 brouard 10907: 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 10908: 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 */
10909: 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 10910: 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);
10911: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10912: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10913: Fixed[k]= 1;
10914: Dummy[k]= 1;
10915: nqtveff++;
10916: modell[k].maintype= VTYPE;
10917: modell[k].subtype= VQ;
10918: ncovv++; /* Only simple time varying variables */
10919: nsq++;
1.334 brouard 10920: 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) */
10921: 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 10922: TvarV[ncovv]=Tvar[k];
1.242 brouard 10923: 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 10924: 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 */
10925: 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 10926: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10927: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10928: 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 10929: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10930: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10931: ncova++;
10932: TvarA[ncova]=Tvar[k];
10933: TvarAind[ncova]=k;
1.231 brouard 10934: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10935: Fixed[k]= 2;
10936: Dummy[k]= 2;
10937: modell[k].maintype= ATYPE;
10938: modell[k].subtype= APFD;
10939: /* ncoveff++; */
1.227 brouard 10940: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10941: Fixed[k]= 2;
10942: Dummy[k]= 3;
10943: modell[k].maintype= ATYPE;
10944: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10945: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10946: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10947: Fixed[k]= 3;
10948: Dummy[k]= 2;
10949: modell[k].maintype= ATYPE;
10950: modell[k].subtype= APVD; /* Product age * varying dummy */
10951: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10952: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10953: Fixed[k]= 3;
10954: Dummy[k]= 3;
10955: modell[k].maintype= ATYPE;
10956: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10957: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10958: }
10959: }else if (Typevar[k] == 2) { /* product without age */
10960: k1=Tposprod[k];
10961: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10962: if(Tvard[k1][2] <=ncovcol){
10963: Fixed[k]= 1;
10964: Dummy[k]= 0;
10965: modell[k].maintype= FTYPE;
10966: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10967: ncovf++; /* Fixed variables without age */
10968: TvarF[ncovf]=Tvar[k];
10969: TvarFind[ncovf]=k;
10970: }else if(Tvard[k1][2] <=ncovcol+nqv){
10971: Fixed[k]= 0; /* or 2 ?*/
10972: Dummy[k]= 1;
10973: modell[k].maintype= FTYPE;
10974: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10975: ncovf++; /* Varying variables without age */
10976: TvarF[ncovf]=Tvar[k];
10977: TvarFind[ncovf]=k;
10978: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10979: Fixed[k]= 1;
10980: Dummy[k]= 0;
10981: modell[k].maintype= VTYPE;
10982: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10983: ncovv++; /* Varying variables without age */
10984: TvarV[ncovv]=Tvar[k];
10985: TvarVind[ncovv]=k;
10986: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10987: Fixed[k]= 1;
10988: Dummy[k]= 1;
10989: modell[k].maintype= VTYPE;
10990: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10991: ncovv++; /* Varying variables without age */
10992: TvarV[ncovv]=Tvar[k];
10993: TvarVind[ncovv]=k;
10994: }
1.227 brouard 10995: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10996: if(Tvard[k1][2] <=ncovcol){
10997: Fixed[k]= 0; /* or 2 ?*/
10998: Dummy[k]= 1;
10999: modell[k].maintype= FTYPE;
11000: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
11001: ncovf++; /* Fixed variables without age */
11002: TvarF[ncovf]=Tvar[k];
11003: TvarFind[ncovf]=k;
11004: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11005: Fixed[k]= 1;
11006: Dummy[k]= 1;
11007: modell[k].maintype= VTYPE;
11008: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
11009: ncovv++; /* Varying variables without age */
11010: TvarV[ncovv]=Tvar[k];
11011: TvarVind[ncovv]=k;
11012: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11013: Fixed[k]= 1;
11014: Dummy[k]= 1;
11015: modell[k].maintype= VTYPE;
11016: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
11017: ncovv++; /* Varying variables without age */
11018: TvarV[ncovv]=Tvar[k];
11019: TvarVind[ncovv]=k;
11020: ncovv++; /* Varying variables without age */
11021: TvarV[ncovv]=Tvar[k];
11022: TvarVind[ncovv]=k;
11023: }
1.227 brouard 11024: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 11025: if(Tvard[k1][2] <=ncovcol){
11026: Fixed[k]= 1;
11027: Dummy[k]= 1;
11028: modell[k].maintype= VTYPE;
11029: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
11030: ncovv++; /* Varying variables without age */
11031: TvarV[ncovv]=Tvar[k];
11032: TvarVind[ncovv]=k;
11033: }else if(Tvard[k1][2] <=ncovcol+nqv){
11034: Fixed[k]= 1;
11035: Dummy[k]= 1;
11036: modell[k].maintype= VTYPE;
11037: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
11038: ncovv++; /* Varying variables without age */
11039: TvarV[ncovv]=Tvar[k];
11040: TvarVind[ncovv]=k;
11041: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11042: Fixed[k]= 1;
11043: Dummy[k]= 0;
11044: modell[k].maintype= VTYPE;
11045: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
11046: ncovv++; /* Varying variables without age */
11047: TvarV[ncovv]=Tvar[k];
11048: TvarVind[ncovv]=k;
11049: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11050: Fixed[k]= 1;
11051: Dummy[k]= 1;
11052: modell[k].maintype= VTYPE;
11053: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
11054: ncovv++; /* Varying variables without age */
11055: TvarV[ncovv]=Tvar[k];
11056: TvarVind[ncovv]=k;
11057: }
1.227 brouard 11058: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 11059: if(Tvard[k1][2] <=ncovcol){
11060: Fixed[k]= 1;
11061: Dummy[k]= 1;
11062: modell[k].maintype= VTYPE;
11063: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
11064: ncovv++; /* Varying variables without age */
11065: TvarV[ncovv]=Tvar[k];
11066: TvarVind[ncovv]=k;
11067: }else if(Tvard[k1][2] <=ncovcol+nqv){
11068: Fixed[k]= 1;
11069: Dummy[k]= 1;
11070: modell[k].maintype= VTYPE;
11071: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
11072: ncovv++; /* Varying variables without age */
11073: TvarV[ncovv]=Tvar[k];
11074: TvarVind[ncovv]=k;
11075: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11076: Fixed[k]= 1;
11077: Dummy[k]= 1;
11078: modell[k].maintype= VTYPE;
11079: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
11080: ncovv++; /* Varying variables without age */
11081: TvarV[ncovv]=Tvar[k];
11082: TvarVind[ncovv]=k;
11083: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11084: Fixed[k]= 1;
11085: Dummy[k]= 1;
11086: modell[k].maintype= VTYPE;
11087: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
11088: ncovv++; /* Varying variables without age */
11089: TvarV[ncovv]=Tvar[k];
11090: TvarVind[ncovv]=k;
11091: }
1.227 brouard 11092: }else{
1.240 brouard 11093: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11094: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11095: } /*end k1*/
1.225 brouard 11096: }else{
1.226 brouard 11097: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
11098: 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 11099: }
1.227 brouard 11100: 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 11101: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 11102: 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]);
11103: }
11104: /* Searching for doublons in the model */
11105: for(k1=1; k1<= cptcovt;k1++){
11106: for(k2=1; k2 <k1;k2++){
1.285 brouard 11107: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
11108: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 11109: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
11110: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 11111: 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]);
11112: 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 11113: return(1);
11114: }
11115: }else if (Typevar[k1] ==2){
11116: k3=Tposprod[k1];
11117: k4=Tposprod[k2];
11118: 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])) ){
11119: 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]]);
11120: 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);
11121: return(1);
11122: }
11123: }
1.227 brouard 11124: }
11125: }
1.225 brouard 11126: }
11127: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
11128: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 11129: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
11130: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 11131: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 11132: /*endread:*/
1.225 brouard 11133: printf("Exiting decodemodel: ");
11134: return (1);
1.136 brouard 11135: }
11136:
1.169 brouard 11137: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 11138: {/* Check ages at death */
1.136 brouard 11139: int i, m;
1.218 brouard 11140: int firstone=0;
11141:
1.136 brouard 11142: for (i=1; i<=imx; i++) {
11143: for(m=2; (m<= maxwav); m++) {
11144: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
11145: anint[m][i]=9999;
1.216 brouard 11146: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
11147: s[m][i]=-1;
1.136 brouard 11148: }
11149: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 11150: *nberr = *nberr + 1;
1.218 brouard 11151: if(firstone == 0){
11152: firstone=1;
1.260 brouard 11153: 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 11154: }
1.262 brouard 11155: 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 11156: s[m][i]=-1; /* Droping the death status */
1.136 brouard 11157: }
11158: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 11159: (*nberr)++;
1.259 brouard 11160: 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 11161: 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 11162: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 11163: }
11164: }
11165: }
11166:
11167: for (i=1; i<=imx; i++) {
11168: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
11169: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 11170: 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 11171: if (s[m][i] >= nlstate+1) {
1.169 brouard 11172: if(agedc[i]>0){
11173: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 11174: agev[m][i]=agedc[i];
1.214 brouard 11175: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 11176: }else {
1.136 brouard 11177: if ((int)andc[i]!=9999){
11178: nbwarn++;
11179: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
11180: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
11181: agev[m][i]=-1;
11182: }
11183: }
1.169 brouard 11184: } /* agedc > 0 */
1.214 brouard 11185: } /* end if */
1.136 brouard 11186: else if(s[m][i] !=9){ /* Standard case, age in fractional
11187: years but with the precision of a month */
11188: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
11189: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
11190: agev[m][i]=1;
11191: else if(agev[m][i] < *agemin){
11192: *agemin=agev[m][i];
11193: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
11194: }
11195: else if(agev[m][i] >*agemax){
11196: *agemax=agev[m][i];
1.156 brouard 11197: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 11198: }
11199: /*agev[m][i]=anint[m][i]-annais[i];*/
11200: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 11201: } /* en if 9*/
1.136 brouard 11202: else { /* =9 */
1.214 brouard 11203: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 11204: agev[m][i]=1;
11205: s[m][i]=-1;
11206: }
11207: }
1.214 brouard 11208: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 11209: agev[m][i]=1;
1.214 brouard 11210: else{
11211: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11212: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11213: agev[m][i]=0;
11214: }
11215: } /* End for lastpass */
11216: }
1.136 brouard 11217:
11218: for (i=1; i<=imx; i++) {
11219: for(m=firstpass; (m<=lastpass); m++){
11220: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 11221: (*nberr)++;
1.136 brouard 11222: 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);
11223: 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);
11224: return 1;
11225: }
11226: }
11227: }
11228:
11229: /*for (i=1; i<=imx; i++){
11230: for (m=firstpass; (m<lastpass); m++){
11231: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
11232: }
11233:
11234: }*/
11235:
11236:
1.139 brouard 11237: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
11238: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 11239:
11240: return (0);
1.164 brouard 11241: /* endread:*/
1.136 brouard 11242: printf("Exiting calandcheckages: ");
11243: return (1);
11244: }
11245:
1.172 brouard 11246: #if defined(_MSC_VER)
11247: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11248: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11249: //#include "stdafx.h"
11250: //#include <stdio.h>
11251: //#include <tchar.h>
11252: //#include <windows.h>
11253: //#include <iostream>
11254: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
11255:
11256: LPFN_ISWOW64PROCESS fnIsWow64Process;
11257:
11258: BOOL IsWow64()
11259: {
11260: BOOL bIsWow64 = FALSE;
11261:
11262: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
11263: // (HANDLE, PBOOL);
11264:
11265: //LPFN_ISWOW64PROCESS fnIsWow64Process;
11266:
11267: HMODULE module = GetModuleHandle(_T("kernel32"));
11268: const char funcName[] = "IsWow64Process";
11269: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
11270: GetProcAddress(module, funcName);
11271:
11272: if (NULL != fnIsWow64Process)
11273: {
11274: if (!fnIsWow64Process(GetCurrentProcess(),
11275: &bIsWow64))
11276: //throw std::exception("Unknown error");
11277: printf("Unknown error\n");
11278: }
11279: return bIsWow64 != FALSE;
11280: }
11281: #endif
1.177 brouard 11282:
1.191 brouard 11283: void syscompilerinfo(int logged)
1.292 brouard 11284: {
11285: #include <stdint.h>
11286:
11287: /* #include "syscompilerinfo.h"*/
1.185 brouard 11288: /* command line Intel compiler 32bit windows, XP compatible:*/
11289: /* /GS /W3 /Gy
11290: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
11291: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
11292: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 11293: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
11294: */
11295: /* 64 bits */
1.185 brouard 11296: /*
11297: /GS /W3 /Gy
11298: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
11299: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
11300: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
11301: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
11302: /* Optimization are useless and O3 is slower than O2 */
11303: /*
11304: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
11305: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
11306: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
11307: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
11308: */
1.186 brouard 11309: /* Link is */ /* /OUT:"visual studio
1.185 brouard 11310: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
11311: /PDB:"visual studio
11312: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
11313: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
11314: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
11315: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
11316: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
11317: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
11318: uiAccess='false'"
11319: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
11320: /NOLOGO /TLBID:1
11321: */
1.292 brouard 11322:
11323:
1.177 brouard 11324: #if defined __INTEL_COMPILER
1.178 brouard 11325: #if defined(__GNUC__)
11326: struct utsname sysInfo; /* For Intel on Linux and OS/X */
11327: #endif
1.177 brouard 11328: #elif defined(__GNUC__)
1.179 brouard 11329: #ifndef __APPLE__
1.174 brouard 11330: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 11331: #endif
1.177 brouard 11332: struct utsname sysInfo;
1.178 brouard 11333: int cross = CROSS;
11334: if (cross){
11335: printf("Cross-");
1.191 brouard 11336: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 11337: }
1.174 brouard 11338: #endif
11339:
1.191 brouard 11340: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 11341: #if defined(__clang__)
1.191 brouard 11342: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 11343: #endif
11344: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 11345: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 11346: #endif
11347: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 11348: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 11349: #endif
11350: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 11351: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 11352: #endif
11353: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 11354: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 11355: #endif
11356: #if defined(_MSC_VER)
1.191 brouard 11357: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 11358: #endif
11359: #if defined(__PGI)
1.191 brouard 11360: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 11361: #endif
11362: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 11363: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 11364: #endif
1.191 brouard 11365: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 11366:
1.167 brouard 11367: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
11368: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
11369: // Windows (x64 and x86)
1.191 brouard 11370: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 11371: #elif __unix__ // all unices, not all compilers
11372: // Unix
1.191 brouard 11373: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 11374: #elif __linux__
11375: // linux
1.191 brouard 11376: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 11377: #elif __APPLE__
1.174 brouard 11378: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 11379: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11380: #endif
11381:
11382: /* __MINGW32__ */
11383: /* __CYGWIN__ */
11384: /* __MINGW64__ */
11385: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11386: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11387: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11388: /* _WIN64 // Defined for applications for Win64. */
11389: /* _M_X64 // Defined for compilations that target x64 processors. */
11390: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11391:
1.167 brouard 11392: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11393: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11394: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11395: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11396: #else
1.191 brouard 11397: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11398: #endif
11399:
1.169 brouard 11400: #if defined(__GNUC__)
11401: # if defined(__GNUC_PATCHLEVEL__)
11402: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11403: + __GNUC_MINOR__ * 100 \
11404: + __GNUC_PATCHLEVEL__)
11405: # else
11406: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11407: + __GNUC_MINOR__ * 100)
11408: # endif
1.174 brouard 11409: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11410: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11411:
11412: if (uname(&sysInfo) != -1) {
11413: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11414: 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 11415: }
11416: else
11417: perror("uname() error");
1.179 brouard 11418: //#ifndef __INTEL_COMPILER
11419: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11420: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11421: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11422: #endif
1.169 brouard 11423: #endif
1.172 brouard 11424:
1.286 brouard 11425: // void main ()
1.172 brouard 11426: // {
1.169 brouard 11427: #if defined(_MSC_VER)
1.174 brouard 11428: if (IsWow64()){
1.191 brouard 11429: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11430: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11431: }
11432: else{
1.191 brouard 11433: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11434: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11435: }
1.172 brouard 11436: // printf("\nPress Enter to continue...");
11437: // getchar();
11438: // }
11439:
1.169 brouard 11440: #endif
11441:
1.167 brouard 11442:
1.219 brouard 11443: }
1.136 brouard 11444:
1.219 brouard 11445: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11446: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 11447: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 11448: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11449: /* double ftolpl = 1.e-10; */
1.180 brouard 11450: double age, agebase, agelim;
1.203 brouard 11451: double tot;
1.180 brouard 11452:
1.202 brouard 11453: strcpy(filerespl,"PL_");
11454: strcat(filerespl,fileresu);
11455: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11456: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11457: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11458: }
1.288 brouard 11459: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11460: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11461: pstamp(ficrespl);
1.288 brouard 11462: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11463: fprintf(ficrespl,"#Age ");
11464: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11465: fprintf(ficrespl,"\n");
1.180 brouard 11466:
1.219 brouard 11467: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11468:
1.219 brouard 11469: agebase=ageminpar;
11470: agelim=agemaxpar;
1.180 brouard 11471:
1.227 brouard 11472: /* i1=pow(2,ncoveff); */
1.234 brouard 11473: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11474: if (cptcovn < 1){i1=1;}
1.180 brouard 11475:
1.238 brouard 11476: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
11477: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 11478: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11479: continue;
1.235 brouard 11480:
1.238 brouard 11481: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11482: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11483: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11484: /* k=k+1; */
11485: /* to clean */
1.332 brouard 11486: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11487: fprintf(ficrespl,"#******");
11488: printf("#******");
11489: fprintf(ficlog,"#******");
11490: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
1.332 brouard 11491: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
11492: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* Here problem for varying dummy*/
11493: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11494: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11495: }
11496: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11497: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11498: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11499: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11500: }
11501: fprintf(ficrespl,"******\n");
11502: printf("******\n");
11503: fprintf(ficlog,"******\n");
11504: if(invalidvarcomb[k]){
11505: printf("\nCombination (%d) ignored because no case \n",k);
11506: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
11507: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
11508: continue;
11509: }
1.219 brouard 11510:
1.238 brouard 11511: fprintf(ficrespl,"#Age ");
11512: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 11513: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11514: }
11515: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
11516: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 11517:
1.238 brouard 11518: for (age=agebase; age<=agelim; age++){
11519: /* for (age=agebase; age<=agebase; age++){ */
11520: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
11521: fprintf(ficrespl,"%.0f ",age );
11522: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11523: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11524: tot=0.;
11525: for(i=1; i<=nlstate;i++){
11526: tot += prlim[i][i];
11527: fprintf(ficrespl," %.5f", prlim[i][i]);
11528: }
11529: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
11530: } /* Age */
11531: /* was end of cptcod */
11532: } /* cptcov */
11533: } /* nres */
1.219 brouard 11534: return 0;
1.180 brouard 11535: }
11536:
1.218 brouard 11537: 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 11538: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 11539:
11540: /* Computes the back prevalence limit for any combination of covariate values
11541: * at any age between ageminpar and agemaxpar
11542: */
1.235 brouard 11543: int i, j, k, i1, nres=0 ;
1.217 brouard 11544: /* double ftolpl = 1.e-10; */
11545: double age, agebase, agelim;
11546: double tot;
1.218 brouard 11547: /* double ***mobaverage; */
11548: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 11549:
11550: strcpy(fileresplb,"PLB_");
11551: strcat(fileresplb,fileresu);
11552: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 11553: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
11554: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 11555: }
1.288 brouard 11556: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
11557: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 11558: pstamp(ficresplb);
1.288 brouard 11559: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 11560: fprintf(ficresplb,"#Age ");
11561: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
11562: fprintf(ficresplb,"\n");
11563:
1.218 brouard 11564:
11565: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
11566:
11567: agebase=ageminpar;
11568: agelim=agemaxpar;
11569:
11570:
1.227 brouard 11571: i1=pow(2,cptcoveff);
1.218 brouard 11572: if (cptcovn < 1){i1=1;}
1.227 brouard 11573:
1.238 brouard 11574: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11575: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11576: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11577: continue;
1.332 brouard 11578: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11579: fprintf(ficresplb,"#******");
11580: printf("#******");
11581: fprintf(ficlog,"#******");
11582: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
1.332 brouard 11583: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11584: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11585: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11586: }
11587: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 11588: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
11589: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
11590: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 11591: }
11592: fprintf(ficresplb,"******\n");
11593: printf("******\n");
11594: fprintf(ficlog,"******\n");
11595: if(invalidvarcomb[k]){
11596: printf("\nCombination (%d) ignored because no cases \n",k);
11597: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
11598: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
11599: continue;
11600: }
1.218 brouard 11601:
1.238 brouard 11602: fprintf(ficresplb,"#Age ");
11603: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 11604: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11605: }
11606: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
11607: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 11608:
11609:
1.238 brouard 11610: for (age=agebase; age<=agelim; age++){
11611: /* for (age=agebase; age<=agebase; age++){ */
11612: if(mobilavproj > 0){
11613: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11614: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11615: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11616: }else if (mobilavproj == 0){
11617: 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);
11618: 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);
11619: exit(1);
11620: }else{
11621: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11622: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11623: /* printf("TOTOT\n"); */
11624: /* exit(1); */
1.238 brouard 11625: }
11626: fprintf(ficresplb,"%.0f ",age );
11627: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11628: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11629: tot=0.;
11630: for(i=1; i<=nlstate;i++){
11631: tot += bprlim[i][i];
11632: fprintf(ficresplb," %.5f", bprlim[i][i]);
11633: }
11634: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11635: } /* Age */
11636: /* was end of cptcod */
1.255 brouard 11637: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 11638: } /* end of any combination */
11639: } /* end of nres */
1.218 brouard 11640: /* hBijx(p, bage, fage); */
11641: /* fclose(ficrespijb); */
11642:
11643: return 0;
1.217 brouard 11644: }
1.218 brouard 11645:
1.180 brouard 11646: int hPijx(double *p, int bage, int fage){
11647: /*------------- h Pij x at various ages ------------*/
1.336 ! brouard 11648: /* to be optimized with precov */
1.180 brouard 11649: int stepsize;
11650: int agelim;
11651: int hstepm;
11652: int nhstepm;
1.235 brouard 11653: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11654:
11655: double agedeb;
11656: double ***p3mat;
11657:
1.201 brouard 11658: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 11659: if((ficrespij=fopen(filerespij,"w"))==NULL) {
11660: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
11661: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
11662: }
11663: printf("Computing pij: result on file '%s' \n", filerespij);
11664: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
11665:
11666: stepsize=(int) (stepm+YEARM-1)/YEARM;
11667: /*if (stepm<=24) stepsize=2;*/
11668:
11669: agelim=AGESUP;
11670: hstepm=stepsize*YEARM; /* Every year of age */
11671: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 11672:
1.180 brouard 11673: /* hstepm=1; aff par mois*/
11674: pstamp(ficrespij);
11675: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 11676: i1= pow(2,cptcoveff);
1.218 brouard 11677: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11678: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11679: /* k=k+1; */
1.235 brouard 11680: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11681: for(k=1; k<=i1;k++){
1.253 brouard 11682: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11683: continue;
1.183 brouard 11684: fprintf(ficrespij,"\n#****** ");
1.227 brouard 11685: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11686: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 11687: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11688: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11689: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11690: }
1.183 brouard 11691: fprintf(ficrespij,"******\n");
11692:
11693: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
11694: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
11695: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
11696:
11697: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 11698:
1.183 brouard 11699: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11700: oldm=oldms;savm=savms;
1.235 brouard 11701: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 11702: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
11703: for(i=1; i<=nlstate;i++)
11704: for(j=1; j<=nlstate+ndeath;j++)
11705: fprintf(ficrespij," %1d-%1d",i,j);
11706: fprintf(ficrespij,"\n");
11707: for (h=0; h<=nhstepm; h++){
11708: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11709: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 11710: for(i=1; i<=nlstate;i++)
11711: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 11712: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 11713: fprintf(ficrespij,"\n");
11714: }
1.183 brouard 11715: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11716: fprintf(ficrespij,"\n");
11717: }
1.180 brouard 11718: /*}*/
11719: }
1.218 brouard 11720: return 0;
1.180 brouard 11721: }
1.218 brouard 11722:
11723: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 11724: /*------------- h Bij x at various ages ------------*/
1.336 ! brouard 11725: /* To be optimized with precov */
1.217 brouard 11726: int stepsize;
1.218 brouard 11727: /* int agelim; */
11728: int ageminl;
1.217 brouard 11729: int hstepm;
11730: int nhstepm;
1.238 brouard 11731: int h, i, i1, j, k, nres;
1.218 brouard 11732:
1.217 brouard 11733: double agedeb;
11734: double ***p3mat;
1.218 brouard 11735:
11736: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
11737: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
11738: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11739: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11740: }
11741: printf("Computing pij back: result on file '%s' \n", filerespijb);
11742: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
11743:
11744: stepsize=(int) (stepm+YEARM-1)/YEARM;
11745: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 11746:
1.218 brouard 11747: /* agelim=AGESUP; */
1.289 brouard 11748: ageminl=AGEINF; /* was 30 */
1.218 brouard 11749: hstepm=stepsize*YEARM; /* Every year of age */
11750: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11751:
11752: /* hstepm=1; aff par mois*/
11753: pstamp(ficrespijb);
1.255 brouard 11754: 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 11755: i1= pow(2,cptcoveff);
1.218 brouard 11756: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11757: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11758: /* k=k+1; */
1.238 brouard 11759: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11760: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11761: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11762: continue;
11763: fprintf(ficrespijb,"\n#****** ");
11764: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11765: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11766: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 11767: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 11768: }
11769: fprintf(ficrespijb,"******\n");
1.264 brouard 11770: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 11771: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
11772: continue;
11773: }
11774:
11775: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
11776: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
11777: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 11778: 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 */
11779: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 11780:
11781: /* nhstepm=nhstepm*YEARM; aff par mois*/
11782:
1.266 brouard 11783: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
11784: /* and memory limitations if stepm is small */
11785:
1.238 brouard 11786: /* oldm=oldms;savm=savms; */
11787: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.325 brouard 11788: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
1.238 brouard 11789: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 11790: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 11791: for(i=1; i<=nlstate;i++)
11792: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 11793: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 11794: fprintf(ficrespijb,"\n");
1.238 brouard 11795: for (h=0; h<=nhstepm; h++){
11796: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11797: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
11798: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
11799: for(i=1; i<=nlstate;i++)
11800: for(j=1; j<=nlstate+ndeath;j++)
1.325 brouard 11801: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.238 brouard 11802: fprintf(ficrespijb,"\n");
11803: }
11804: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11805: fprintf(ficrespijb,"\n");
11806: } /* end age deb */
11807: } /* end combination */
11808: } /* end nres */
1.218 brouard 11809: return 0;
11810: } /* hBijx */
1.217 brouard 11811:
1.180 brouard 11812:
1.136 brouard 11813: /***********************************************/
11814: /**************** Main Program *****************/
11815: /***********************************************/
11816:
11817: int main(int argc, char *argv[])
11818: {
11819: #ifdef GSL
11820: const gsl_multimin_fminimizer_type *T;
11821: size_t iteri = 0, it;
11822: int rval = GSL_CONTINUE;
11823: int status = GSL_SUCCESS;
11824: double ssval;
11825: #endif
11826: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11827: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11828: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11829: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11830: int jj, ll, li, lj, lk;
1.136 brouard 11831: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11832: int num_filled;
1.136 brouard 11833: int itimes;
11834: int NDIM=2;
11835: int vpopbased=0;
1.235 brouard 11836: int nres=0;
1.258 brouard 11837: int endishere=0;
1.277 brouard 11838: int noffset=0;
1.274 brouard 11839: int ncurrv=0; /* Temporary variable */
11840:
1.164 brouard 11841: char ca[32], cb[32];
1.136 brouard 11842: /* FILE *fichtm; *//* Html File */
11843: /* FILE *ficgp;*/ /*Gnuplot File */
11844: struct stat info;
1.191 brouard 11845: double agedeb=0.;
1.194 brouard 11846:
11847: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11848: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11849:
1.165 brouard 11850: double fret;
1.191 brouard 11851: double dum=0.; /* Dummy variable */
1.136 brouard 11852: double ***p3mat;
1.218 brouard 11853: /* double ***mobaverage; */
1.319 brouard 11854: double wald;
1.164 brouard 11855:
11856: char line[MAXLINE];
1.197 brouard 11857: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11858:
1.234 brouard 11859: char modeltemp[MAXLINE];
1.332 brouard 11860: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 11861:
1.136 brouard 11862: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11863: char *tok, *val; /* pathtot */
1.334 brouard 11864: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.195 brouard 11865: int c, h , cpt, c2;
1.191 brouard 11866: int jl=0;
11867: int i1, j1, jk, stepsize=0;
1.194 brouard 11868: int count=0;
11869:
1.164 brouard 11870: int *tab;
1.136 brouard 11871: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11872: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11873: /* double anprojf, mprojf, jprojf; */
11874: /* double jintmean,mintmean,aintmean; */
11875: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11876: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11877: double yrfproj= 10.0; /* Number of years of forward projections */
11878: double yrbproj= 10.0; /* Number of years of backward projections */
11879: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11880: int mobilav=0,popforecast=0;
1.191 brouard 11881: int hstepm=0, nhstepm=0;
1.136 brouard 11882: int agemortsup;
11883: float sumlpop=0.;
11884: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11885: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11886:
1.191 brouard 11887: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11888: double ftolpl=FTOL;
11889: double **prlim;
1.217 brouard 11890: double **bprlim;
1.317 brouard 11891: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
11892: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11893: double ***paramstart; /* Matrix of starting parameter values */
11894: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11895: double **matcov; /* Matrix of covariance */
1.203 brouard 11896: double **hess; /* Hessian matrix */
1.136 brouard 11897: double ***delti3; /* Scale */
11898: double *delti; /* Scale */
11899: double ***eij, ***vareij;
11900: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11901:
1.136 brouard 11902: double *epj, vepp;
1.164 brouard 11903:
1.273 brouard 11904: double dateprev1, dateprev2;
1.296 brouard 11905: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11906: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11907:
1.217 brouard 11908:
1.136 brouard 11909: double **ximort;
1.145 brouard 11910: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11911: int *dcwave;
11912:
1.164 brouard 11913: char z[1]="c";
1.136 brouard 11914:
11915: /*char *strt;*/
11916: char strtend[80];
1.126 brouard 11917:
1.164 brouard 11918:
1.126 brouard 11919: /* setlocale (LC_ALL, ""); */
11920: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11921: /* textdomain (PACKAGE); */
11922: /* setlocale (LC_CTYPE, ""); */
11923: /* setlocale (LC_MESSAGES, ""); */
11924:
11925: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11926: rstart_time = time(NULL);
11927: /* (void) gettimeofday(&start_time,&tzp);*/
11928: start_time = *localtime(&rstart_time);
1.126 brouard 11929: curr_time=start_time;
1.157 brouard 11930: /*tml = *localtime(&start_time.tm_sec);*/
11931: /* strcpy(strstart,asctime(&tml)); */
11932: strcpy(strstart,asctime(&start_time));
1.126 brouard 11933:
11934: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11935: /* tp.tm_sec = tp.tm_sec +86400; */
11936: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11937: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11938: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11939: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11940: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11941: /* strt=asctime(&tmg); */
11942: /* printf("Time(after) =%s",strstart); */
11943: /* (void) time (&time_value);
11944: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11945: * tm = *localtime(&time_value);
11946: * strstart=asctime(&tm);
11947: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11948: */
11949:
11950: nberr=0; /* Number of errors and warnings */
11951: nbwarn=0;
1.184 brouard 11952: #ifdef WIN32
11953: _getcwd(pathcd, size);
11954: #else
1.126 brouard 11955: getcwd(pathcd, size);
1.184 brouard 11956: #endif
1.191 brouard 11957: syscompilerinfo(0);
1.196 brouard 11958: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11959: if(argc <=1){
11960: printf("\nEnter the parameter file name: ");
1.205 brouard 11961: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11962: printf("ERROR Empty parameter file name\n");
11963: goto end;
11964: }
1.126 brouard 11965: i=strlen(pathr);
11966: if(pathr[i-1]=='\n')
11967: pathr[i-1]='\0';
1.156 brouard 11968: i=strlen(pathr);
1.205 brouard 11969: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11970: pathr[i-1]='\0';
1.205 brouard 11971: }
11972: i=strlen(pathr);
11973: if( i==0 ){
11974: printf("ERROR Empty parameter file name\n");
11975: goto end;
11976: }
11977: for (tok = pathr; tok != NULL; ){
1.126 brouard 11978: printf("Pathr |%s|\n",pathr);
11979: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11980: printf("val= |%s| pathr=%s\n",val,pathr);
11981: strcpy (pathtot, val);
11982: if(pathr[0] == '\0') break; /* Dirty */
11983: }
11984: }
1.281 brouard 11985: else if (argc<=2){
11986: strcpy(pathtot,argv[1]);
11987: }
1.126 brouard 11988: else{
11989: strcpy(pathtot,argv[1]);
1.281 brouard 11990: strcpy(z,argv[2]);
11991: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11992: }
11993: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11994: /*cygwin_split_path(pathtot,path,optionfile);
11995: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11996: /* cutv(path,optionfile,pathtot,'\\');*/
11997:
11998: /* Split argv[0], imach program to get pathimach */
11999: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
12000: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12001: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12002: /* strcpy(pathimach,argv[0]); */
12003: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
12004: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
12005: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 12006: #ifdef WIN32
12007: _chdir(path); /* Can be a relative path */
12008: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
12009: #else
1.126 brouard 12010: chdir(path); /* Can be a relative path */
1.184 brouard 12011: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
12012: #endif
12013: printf("Current directory %s!\n",pathcd);
1.126 brouard 12014: strcpy(command,"mkdir ");
12015: strcat(command,optionfilefiname);
12016: if((outcmd=system(command)) != 0){
1.169 brouard 12017: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 12018: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
12019: /* fclose(ficlog); */
12020: /* exit(1); */
12021: }
12022: /* if((imk=mkdir(optionfilefiname))<0){ */
12023: /* perror("mkdir"); */
12024: /* } */
12025:
12026: /*-------- arguments in the command line --------*/
12027:
1.186 brouard 12028: /* Main Log file */
1.126 brouard 12029: strcat(filelog, optionfilefiname);
12030: strcat(filelog,".log"); /* */
12031: if((ficlog=fopen(filelog,"w"))==NULL) {
12032: printf("Problem with logfile %s\n",filelog);
12033: goto end;
12034: }
12035: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 12036: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 12037: fprintf(ficlog,"\nEnter the parameter file name: \n");
12038: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
12039: path=%s \n\
12040: optionfile=%s\n\
12041: optionfilext=%s\n\
1.156 brouard 12042: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 12043:
1.197 brouard 12044: syscompilerinfo(1);
1.167 brouard 12045:
1.126 brouard 12046: printf("Local time (at start):%s",strstart);
12047: fprintf(ficlog,"Local time (at start): %s",strstart);
12048: fflush(ficlog);
12049: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 12050: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 12051:
12052: /* */
12053: strcpy(fileres,"r");
12054: strcat(fileres, optionfilefiname);
1.201 brouard 12055: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 12056: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 12057: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 12058:
1.186 brouard 12059: /* Main ---------arguments file --------*/
1.126 brouard 12060:
12061: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 12062: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
12063: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 12064: fflush(ficlog);
1.149 brouard 12065: /* goto end; */
12066: exit(70);
1.126 brouard 12067: }
12068:
12069: strcpy(filereso,"o");
1.201 brouard 12070: strcat(filereso,fileresu);
1.126 brouard 12071: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
12072: printf("Problem with Output resultfile: %s\n", filereso);
12073: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
12074: fflush(ficlog);
12075: goto end;
12076: }
1.278 brouard 12077: /*-------- Rewriting parameter file ----------*/
12078: strcpy(rfileres,"r"); /* "Rparameterfile */
12079: strcat(rfileres,optionfilefiname); /* Parameter file first name */
12080: strcat(rfileres,"."); /* */
12081: strcat(rfileres,optionfilext); /* Other files have txt extension */
12082: if((ficres =fopen(rfileres,"w"))==NULL) {
12083: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
12084: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
12085: fflush(ficlog);
12086: goto end;
12087: }
12088: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 12089:
1.278 brouard 12090:
1.126 brouard 12091: /* Reads comments: lines beginning with '#' */
12092: numlinepar=0;
1.277 brouard 12093: /* Is it a BOM UTF-8 Windows file? */
12094: /* First parameter line */
1.197 brouard 12095: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 12096: noffset=0;
12097: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12098: {
12099: noffset=noffset+3;
12100: printf("# File is an UTF8 Bom.\n"); // 0xBF
12101: }
1.302 brouard 12102: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12103: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 12104: {
12105: noffset=noffset+2;
12106: printf("# File is an UTF16BE BOM file\n");
12107: }
12108: else if( line[0] == 0 && line[1] == 0)
12109: {
12110: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12111: noffset=noffset+4;
12112: printf("# File is an UTF16BE BOM file\n");
12113: }
12114: } else{
12115: ;/*printf(" Not a BOM file\n");*/
12116: }
12117:
1.197 brouard 12118: /* If line starts with a # it is a comment */
1.277 brouard 12119: if (line[noffset] == '#') {
1.197 brouard 12120: numlinepar++;
12121: fputs(line,stdout);
12122: fputs(line,ficparo);
1.278 brouard 12123: fputs(line,ficres);
1.197 brouard 12124: fputs(line,ficlog);
12125: continue;
12126: }else
12127: break;
12128: }
12129: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
12130: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
12131: if (num_filled != 5) {
12132: printf("Should be 5 parameters\n");
1.283 brouard 12133: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 12134: }
1.126 brouard 12135: numlinepar++;
1.197 brouard 12136: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 12137: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12138: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12139: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 12140: }
12141: /* Second parameter line */
12142: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 12143: /* while(fscanf(ficpar,"%[^\n]", line)) { */
12144: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 12145: if (line[0] == '#') {
12146: numlinepar++;
1.283 brouard 12147: printf("%s",line);
12148: fprintf(ficres,"%s",line);
12149: fprintf(ficparo,"%s",line);
12150: fprintf(ficlog,"%s",line);
1.197 brouard 12151: continue;
12152: }else
12153: break;
12154: }
1.223 brouard 12155: 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", \
12156: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
12157: if (num_filled != 11) {
12158: 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 12159: printf("but line=%s\n",line);
1.283 brouard 12160: 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");
12161: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 12162: }
1.286 brouard 12163: if( lastpass > maxwav){
12164: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12165: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12166: fflush(ficlog);
12167: goto end;
12168: }
12169: 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 12170: 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 12171: 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 12172: 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 12173: }
1.203 brouard 12174: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 12175: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 12176: /* Third parameter line */
12177: while(fgets(line, MAXLINE, ficpar)) {
12178: /* If line starts with a # it is a comment */
12179: if (line[0] == '#') {
12180: numlinepar++;
1.283 brouard 12181: printf("%s",line);
12182: fprintf(ficres,"%s",line);
12183: fprintf(ficparo,"%s",line);
12184: fprintf(ficlog,"%s",line);
1.197 brouard 12185: continue;
12186: }else
12187: break;
12188: }
1.201 brouard 12189: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 12190: if (num_filled != 1){
1.302 brouard 12191: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
12192: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 12193: model[0]='\0';
12194: goto end;
12195: }
12196: else{
12197: if (model[0]=='+'){
12198: for(i=1; i<=strlen(model);i++)
12199: modeltemp[i-1]=model[i];
1.201 brouard 12200: strcpy(model,modeltemp);
1.197 brouard 12201: }
12202: }
1.199 brouard 12203: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 12204: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 12205: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
12206: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
12207: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 12208: }
12209: /* 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); */
12210: /* numlinepar=numlinepar+3; /\* In general *\/ */
12211: /* 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 12212: /* 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); */
12213: /* 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 12214: fflush(ficlog);
1.190 brouard 12215: /* if(model[0]=='#'|| model[0]== '\0'){ */
12216: if(model[0]=='#'){
1.279 brouard 12217: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
12218: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
12219: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 12220: if(mle != -1){
1.279 brouard 12221: 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 12222: exit(1);
12223: }
12224: }
1.126 brouard 12225: while((c=getc(ficpar))=='#' && c!= EOF){
12226: ungetc(c,ficpar);
12227: fgets(line, MAXLINE, ficpar);
12228: numlinepar++;
1.195 brouard 12229: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
12230: z[0]=line[1];
12231: }
12232: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 12233: fputs(line, stdout);
12234: //puts(line);
1.126 brouard 12235: fputs(line,ficparo);
12236: fputs(line,ficlog);
12237: }
12238: ungetc(c,ficpar);
12239:
12240:
1.290 brouard 12241: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
12242: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
12243: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
12244: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 12245: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
12246: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
12247: v1+v2*age+v2*v3 makes cptcovn = 3
12248: */
12249: if (strlen(model)>1)
1.187 brouard 12250: 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 12251: else
1.187 brouard 12252: ncovmodel=2; /* Constant and age */
1.133 brouard 12253: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
12254: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 12255: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
12256: 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);
12257: 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);
12258: fflush(stdout);
12259: fclose (ficlog);
12260: goto end;
12261: }
1.126 brouard 12262: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12263: delti=delti3[1][1];
12264: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
12265: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 12266: /* We could also provide initial parameters values giving by simple logistic regression
12267: * only one way, that is without matrix product. We will have nlstate maximizations */
12268: /* for(i=1;i<nlstate;i++){ */
12269: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12270: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12271: /* } */
1.126 brouard 12272: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 12273: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
12274: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12275: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12276: fclose (ficparo);
12277: fclose (ficlog);
12278: goto end;
12279: exit(0);
1.220 brouard 12280: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 12281: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 12282: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
12283: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12284: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12285: matcov=matrix(1,npar,1,npar);
1.203 brouard 12286: hess=matrix(1,npar,1,npar);
1.220 brouard 12287: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 12288: /* Read guessed parameters */
1.126 brouard 12289: /* Reads comments: lines beginning with '#' */
12290: while((c=getc(ficpar))=='#' && c!= EOF){
12291: ungetc(c,ficpar);
12292: fgets(line, MAXLINE, ficpar);
12293: numlinepar++;
1.141 brouard 12294: fputs(line,stdout);
1.126 brouard 12295: fputs(line,ficparo);
12296: fputs(line,ficlog);
12297: }
12298: ungetc(c,ficpar);
12299:
12300: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 12301: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 12302: for(i=1; i <=nlstate; i++){
1.234 brouard 12303: j=0;
1.126 brouard 12304: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 12305: if(jj==i) continue;
12306: j++;
1.292 brouard 12307: while((c=getc(ficpar))=='#' && c!= EOF){
12308: ungetc(c,ficpar);
12309: fgets(line, MAXLINE, ficpar);
12310: numlinepar++;
12311: fputs(line,stdout);
12312: fputs(line,ficparo);
12313: fputs(line,ficlog);
12314: }
12315: ungetc(c,ficpar);
1.234 brouard 12316: fscanf(ficpar,"%1d%1d",&i1,&j1);
12317: if ((i1 != i) || (j1 != jj)){
12318: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 12319: It might be a problem of design; if ncovcol and the model are correct\n \
12320: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 12321: exit(1);
12322: }
12323: fprintf(ficparo,"%1d%1d",i1,j1);
12324: if(mle==1)
12325: printf("%1d%1d",i,jj);
12326: fprintf(ficlog,"%1d%1d",i,jj);
12327: for(k=1; k<=ncovmodel;k++){
12328: fscanf(ficpar," %lf",¶m[i][j][k]);
12329: if(mle==1){
12330: printf(" %lf",param[i][j][k]);
12331: fprintf(ficlog," %lf",param[i][j][k]);
12332: }
12333: else
12334: fprintf(ficlog," %lf",param[i][j][k]);
12335: fprintf(ficparo," %lf",param[i][j][k]);
12336: }
12337: fscanf(ficpar,"\n");
12338: numlinepar++;
12339: if(mle==1)
12340: printf("\n");
12341: fprintf(ficlog,"\n");
12342: fprintf(ficparo,"\n");
1.126 brouard 12343: }
12344: }
12345: fflush(ficlog);
1.234 brouard 12346:
1.251 brouard 12347: /* Reads parameters values */
1.126 brouard 12348: p=param[1][1];
1.251 brouard 12349: pstart=paramstart[1][1];
1.126 brouard 12350:
12351: /* Reads comments: lines beginning with '#' */
12352: while((c=getc(ficpar))=='#' && c!= EOF){
12353: ungetc(c,ficpar);
12354: fgets(line, MAXLINE, ficpar);
12355: numlinepar++;
1.141 brouard 12356: fputs(line,stdout);
1.126 brouard 12357: fputs(line,ficparo);
12358: fputs(line,ficlog);
12359: }
12360: ungetc(c,ficpar);
12361:
12362: for(i=1; i <=nlstate; i++){
12363: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 12364: fscanf(ficpar,"%1d%1d",&i1,&j1);
12365: if ( (i1-i) * (j1-j) != 0){
12366: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
12367: exit(1);
12368: }
12369: printf("%1d%1d",i,j);
12370: fprintf(ficparo,"%1d%1d",i1,j1);
12371: fprintf(ficlog,"%1d%1d",i1,j1);
12372: for(k=1; k<=ncovmodel;k++){
12373: fscanf(ficpar,"%le",&delti3[i][j][k]);
12374: printf(" %le",delti3[i][j][k]);
12375: fprintf(ficparo," %le",delti3[i][j][k]);
12376: fprintf(ficlog," %le",delti3[i][j][k]);
12377: }
12378: fscanf(ficpar,"\n");
12379: numlinepar++;
12380: printf("\n");
12381: fprintf(ficparo,"\n");
12382: fprintf(ficlog,"\n");
1.126 brouard 12383: }
12384: }
12385: fflush(ficlog);
1.234 brouard 12386:
1.145 brouard 12387: /* Reads covariance matrix */
1.126 brouard 12388: delti=delti3[1][1];
1.220 brouard 12389:
12390:
1.126 brouard 12391: /* 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 12392:
1.126 brouard 12393: /* Reads comments: lines beginning with '#' */
12394: while((c=getc(ficpar))=='#' && c!= EOF){
12395: ungetc(c,ficpar);
12396: fgets(line, MAXLINE, ficpar);
12397: numlinepar++;
1.141 brouard 12398: fputs(line,stdout);
1.126 brouard 12399: fputs(line,ficparo);
12400: fputs(line,ficlog);
12401: }
12402: ungetc(c,ficpar);
1.220 brouard 12403:
1.126 brouard 12404: matcov=matrix(1,npar,1,npar);
1.203 brouard 12405: hess=matrix(1,npar,1,npar);
1.131 brouard 12406: for(i=1; i <=npar; i++)
12407: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12408:
1.194 brouard 12409: /* Scans npar lines */
1.126 brouard 12410: for(i=1; i <=npar; i++){
1.226 brouard 12411: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12412: if(count != 3){
1.226 brouard 12413: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12414: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12415: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12416: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12417: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12418: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12419: exit(1);
1.220 brouard 12420: }else{
1.226 brouard 12421: if(mle==1)
12422: printf("%1d%1d%d",i1,j1,jk);
12423: }
12424: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12425: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12426: for(j=1; j <=i; j++){
1.226 brouard 12427: fscanf(ficpar," %le",&matcov[i][j]);
12428: if(mle==1){
12429: printf(" %.5le",matcov[i][j]);
12430: }
12431: fprintf(ficlog," %.5le",matcov[i][j]);
12432: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12433: }
12434: fscanf(ficpar,"\n");
12435: numlinepar++;
12436: if(mle==1)
1.220 brouard 12437: printf("\n");
1.126 brouard 12438: fprintf(ficlog,"\n");
12439: fprintf(ficparo,"\n");
12440: }
1.194 brouard 12441: /* End of read covariance matrix npar lines */
1.126 brouard 12442: for(i=1; i <=npar; i++)
12443: for(j=i+1;j<=npar;j++)
1.226 brouard 12444: matcov[i][j]=matcov[j][i];
1.126 brouard 12445:
12446: if(mle==1)
12447: printf("\n");
12448: fprintf(ficlog,"\n");
12449:
12450: fflush(ficlog);
12451:
12452: } /* End of mle != -3 */
1.218 brouard 12453:
1.186 brouard 12454: /* Main data
12455: */
1.290 brouard 12456: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12457: /* num=lvector(1,n); */
12458: /* moisnais=vector(1,n); */
12459: /* annais=vector(1,n); */
12460: /* moisdc=vector(1,n); */
12461: /* andc=vector(1,n); */
12462: /* weight=vector(1,n); */
12463: /* agedc=vector(1,n); */
12464: /* cod=ivector(1,n); */
12465: /* for(i=1;i<=n;i++){ */
12466: num=lvector(firstobs,lastobs);
12467: moisnais=vector(firstobs,lastobs);
12468: annais=vector(firstobs,lastobs);
12469: moisdc=vector(firstobs,lastobs);
12470: andc=vector(firstobs,lastobs);
12471: weight=vector(firstobs,lastobs);
12472: agedc=vector(firstobs,lastobs);
12473: cod=ivector(firstobs,lastobs);
12474: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 12475: num[i]=0;
12476: moisnais[i]=0;
12477: annais[i]=0;
12478: moisdc[i]=0;
12479: andc[i]=0;
12480: agedc[i]=0;
12481: cod[i]=0;
12482: weight[i]=1.0; /* Equal weights, 1 by default */
12483: }
1.290 brouard 12484: mint=matrix(1,maxwav,firstobs,lastobs);
12485: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 12486: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.336 ! brouard 12487: /* printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel)); */
1.126 brouard 12488: tab=ivector(1,NCOVMAX);
1.144 brouard 12489: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 12490: 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 12491:
1.136 brouard 12492: /* Reads data from file datafile */
12493: if (readdata(datafile, firstobs, lastobs, &imx)==1)
12494: goto end;
12495:
12496: /* Calculation of the number of parameters from char model */
1.234 brouard 12497: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 12498: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
12499: k=3 V4 Tvar[k=3]= 4 (from V4)
12500: k=2 V1 Tvar[k=2]= 1 (from V1)
12501: k=1 Tvar[1]=2 (from V2)
1.234 brouard 12502: */
12503:
12504: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
12505: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 12506: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 brouard 12507: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 12508: TvarsD=ivector(1,NCOVMAX); /* */
12509: TvarsQind=ivector(1,NCOVMAX); /* */
12510: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 12511: TvarF=ivector(1,NCOVMAX); /* */
12512: TvarFind=ivector(1,NCOVMAX); /* */
12513: TvarV=ivector(1,NCOVMAX); /* */
12514: TvarVind=ivector(1,NCOVMAX); /* */
12515: TvarA=ivector(1,NCOVMAX); /* */
12516: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 12517: TvarFD=ivector(1,NCOVMAX); /* */
12518: TvarFDind=ivector(1,NCOVMAX); /* */
12519: TvarFQ=ivector(1,NCOVMAX); /* */
12520: TvarFQind=ivector(1,NCOVMAX); /* */
12521: TvarVD=ivector(1,NCOVMAX); /* */
12522: TvarVDind=ivector(1,NCOVMAX); /* */
12523: TvarVQ=ivector(1,NCOVMAX); /* */
12524: TvarVQind=ivector(1,NCOVMAX); /* */
12525:
1.230 brouard 12526: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 12527: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 12528: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
12529: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
12530: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 12531: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
12532: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
12533: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
12534: */
12535: /* For model-covariate k tells which data-covariate to use but
12536: because this model-covariate is a construction we invent a new column
12537: ncovcol + k1
12538: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
12539: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 12540: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
12541: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 12542: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
12543: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 12544: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 12545: */
1.145 brouard 12546: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
12547: 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 12548: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
12549: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 brouard 12550: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 12551: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 12552: 4 covariates (3 plus signs)
12553: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 12554: */
12555: for(i=1;i<NCOVMAX;i++)
12556: Tage[i]=0;
1.230 brouard 12557: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 12558: * individual dummy, fixed or varying:
12559: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
12560: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 12561: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
12562: * V1 df, V2 qf, V3 & V4 dv, V5 qv
12563: * Tmodelind[1]@9={9,0,3,2,}*/
12564: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
12565: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 12566: * individual quantitative, fixed or varying:
12567: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
12568: * 3, 1, 0, 0, 0, 0, 0, 0},
12569: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 12570: /* Main decodemodel */
12571:
1.187 brouard 12572:
1.223 brouard 12573: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 12574: goto end;
12575:
1.137 brouard 12576: if((double)(lastobs-imx)/(double)imx > 1.10){
12577: nbwarn++;
12578: 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);
12579: 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);
12580: }
1.136 brouard 12581: /* if(mle==1){*/
1.137 brouard 12582: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
12583: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 12584: }
12585:
12586: /*-calculation of age at interview from date of interview and age at death -*/
12587: agev=matrix(1,maxwav,1,imx);
12588:
12589: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
12590: goto end;
12591:
1.126 brouard 12592:
1.136 brouard 12593: agegomp=(int)agemin;
1.290 brouard 12594: free_vector(moisnais,firstobs,lastobs);
12595: free_vector(annais,firstobs,lastobs);
1.126 brouard 12596: /* free_matrix(mint,1,maxwav,1,n);
12597: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 12598: /* free_vector(moisdc,1,n); */
12599: /* free_vector(andc,1,n); */
1.145 brouard 12600: /* */
12601:
1.126 brouard 12602: wav=ivector(1,imx);
1.214 brouard 12603: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
12604: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
12605: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
12606: 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.*/
12607: bh=imatrix(1,lastpass-firstpass+2,1,imx);
12608: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 12609:
12610: /* Concatenates waves */
1.214 brouard 12611: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
12612: Death is a valid wave (if date is known).
12613: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
12614: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
12615: and mw[mi+1][i]. dh depends on stepm.
12616: */
12617:
1.126 brouard 12618: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 12619: /* Concatenates waves */
1.145 brouard 12620:
1.290 brouard 12621: free_vector(moisdc,firstobs,lastobs);
12622: free_vector(andc,firstobs,lastobs);
1.215 brouard 12623:
1.126 brouard 12624: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12625: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12626: ncodemax[1]=1;
1.145 brouard 12627: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12628: cptcoveff=0;
1.220 brouard 12629: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 brouard 12630: 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 12631: }
12632:
12633: ncovcombmax=pow(2,cptcoveff);
12634: invalidvarcomb=ivector(1, ncovcombmax);
12635: for(i=1;i<ncovcombmax;i++)
12636: invalidvarcomb[i]=0;
12637:
1.211 brouard 12638: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12639: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12640: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12641:
1.200 brouard 12642: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12643: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12644: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12645: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12646: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12647: * (currently 0 or 1) in the data.
12648: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12649: * corresponding modality (h,j).
12650: */
12651:
1.145 brouard 12652: h=0;
12653: /*if (cptcovn > 0) */
1.126 brouard 12654: m=pow(2,cptcoveff);
12655:
1.144 brouard 12656: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12657: * For k=4 covariates, h goes from 1 to m=2**k
12658: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12659: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 12660: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
12661: *______________________________ *______________________
12662: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
12663: * 2 2 1 1 1 * 1 0 0 0 1
12664: * 3 i=2 1 2 1 1 * 2 0 0 1 0
12665: * 4 2 2 1 1 * 3 0 0 1 1
12666: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
12667: * 6 2 1 2 1 * 5 0 1 0 1
12668: * 7 i=4 1 2 2 1 * 6 0 1 1 0
12669: * 8 2 2 2 1 * 7 0 1 1 1
12670: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
12671: * 10 2 1 1 2 * 9 1 0 0 1
12672: * 11 i=6 1 2 1 2 * 10 1 0 1 0
12673: * 12 2 2 1 2 * 11 1 0 1 1
12674: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
12675: * 14 2 1 2 2 * 13 1 1 0 1
12676: * 15 i=8 1 2 2 2 * 14 1 1 1 0
12677: * 16 2 2 2 2 * 15 1 1 1 1
12678: */
1.212 brouard 12679: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 12680: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
12681: * and the value of each covariate?
12682: * V1=1, V2=1, V3=2, V4=1 ?
12683: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
12684: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
12685: * In order to get the real value in the data, we use nbcode
12686: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
12687: * We are keeping this crazy system in order to be able (in the future?)
12688: * to have more than 2 values (0 or 1) for a covariate.
12689: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
12690: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
12691: * bbbbbbbb
12692: * 76543210
12693: * h-1 00000101 (6-1=5)
1.219 brouard 12694: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 12695: * &
12696: * 1 00000001 (1)
1.219 brouard 12697: * 00000000 = 1 & ((h-1) >> (k-1))
12698: * +1= 00000001 =1
1.211 brouard 12699: *
12700: * h=14, k=3 => h'=h-1=13, k'=k-1=2
12701: * h' 1101 =2^3+2^2+0x2^1+2^0
12702: * >>k' 11
12703: * & 00000001
12704: * = 00000001
12705: * +1 = 00000010=2 = codtabm(14,3)
12706: * Reverse h=6 and m=16?
12707: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
12708: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
12709: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
12710: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
12711: * V3=decodtabm(14,3,2**4)=2
12712: * h'=13 1101 =2^3+2^2+0x2^1+2^0
12713: *(h-1) >> (j-1) 0011 =13 >> 2
12714: * &1 000000001
12715: * = 000000001
12716: * +1= 000000010 =2
12717: * 2211
12718: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
12719: * V3=2
1.220 brouard 12720: * codtabm and decodtabm are identical
1.211 brouard 12721: */
12722:
1.145 brouard 12723:
12724: free_ivector(Ndum,-1,NCOVMAX);
12725:
12726:
1.126 brouard 12727:
1.186 brouard 12728: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 12729: strcpy(optionfilegnuplot,optionfilefiname);
12730: if(mle==-3)
1.201 brouard 12731: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 12732: strcat(optionfilegnuplot,".gp");
12733:
12734: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
12735: printf("Problem with file %s",optionfilegnuplot);
12736: }
12737: else{
1.204 brouard 12738: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 12739: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 12740: //fprintf(ficgp,"set missing 'NaNq'\n");
12741: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 12742: }
12743: /* fclose(ficgp);*/
1.186 brouard 12744:
12745:
12746: /* Initialisation of --------- index.htm --------*/
1.126 brouard 12747:
12748: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
12749: if(mle==-3)
1.201 brouard 12750: strcat(optionfilehtm,"-MORT_");
1.126 brouard 12751: strcat(optionfilehtm,".htm");
12752: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 12753: printf("Problem with %s \n",optionfilehtm);
12754: exit(0);
1.126 brouard 12755: }
12756:
12757: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
12758: strcat(optionfilehtmcov,"-cov.htm");
12759: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
12760: printf("Problem with %s \n",optionfilehtmcov), exit(0);
12761: }
12762: else{
12763: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
12764: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12765: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 12766: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
12767: }
12768:
1.335 brouard 12769: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
12770: <title>IMaCh %s</title></head>\n\
12771: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
12772: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
12773: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
12774: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
12775: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
12776:
12777: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12778: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 12779: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.335 brouard 12780: 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 12781: \n\
12782: <hr size=\"2\" color=\"#EC5E5E\">\
12783: <ul><li><h4>Parameter files</h4>\n\
12784: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
12785: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
12786: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
12787: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
12788: - Date and time at start: %s</ul>\n",\
1.335 brouard 12789: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 12790: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
12791: fileres,fileres,\
12792: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
12793: fflush(fichtm);
12794:
12795: strcpy(pathr,path);
12796: strcat(pathr,optionfilefiname);
1.184 brouard 12797: #ifdef WIN32
12798: _chdir(optionfilefiname); /* Move to directory named optionfile */
12799: #else
1.126 brouard 12800: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12801: #endif
12802:
1.126 brouard 12803:
1.220 brouard 12804: /* Calculates basic frequencies. Computes observed prevalence at single age
12805: and for any valid combination of covariates
1.126 brouard 12806: and prints on file fileres'p'. */
1.251 brouard 12807: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12808: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12809:
12810: fprintf(fichtm,"\n");
1.286 brouard 12811: 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 12812: ftol, stepm);
12813: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12814: ncurrv=1;
12815: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12816: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12817: ncurrv=i;
12818: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12819: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12820: ncurrv=i;
12821: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12822: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12823: ncurrv=i;
12824: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12825: 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", \
12826: nlstate, ndeath, maxwav, mle, weightopt);
12827:
12828: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12829: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12830:
12831:
1.317 brouard 12832: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12833: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12834: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12835: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12836: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12837: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12838: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12839: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12840: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12841:
1.126 brouard 12842: /* For Powell, parameters are in a vector p[] starting at p[1]
12843: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12844: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12845:
12846: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12847: /* For mortality only */
1.126 brouard 12848: if (mle==-3){
1.136 brouard 12849: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12850: for(i=1;i<=NDIM;i++)
12851: for(j=1;j<=NDIM;j++)
12852: ximort[i][j]=0.;
1.186 brouard 12853: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12854: cens=ivector(firstobs,lastobs);
12855: ageexmed=vector(firstobs,lastobs);
12856: agecens=vector(firstobs,lastobs);
12857: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12858:
1.126 brouard 12859: for (i=1; i<=imx; i++){
12860: dcwave[i]=-1;
12861: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12862: if (s[m][i]>nlstate) {
12863: dcwave[i]=m;
12864: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12865: break;
12866: }
1.126 brouard 12867: }
1.226 brouard 12868:
1.126 brouard 12869: for (i=1; i<=imx; i++) {
12870: if (wav[i]>0){
1.226 brouard 12871: ageexmed[i]=agev[mw[1][i]][i];
12872: j=wav[i];
12873: agecens[i]=1.;
12874:
12875: if (ageexmed[i]> 1 && wav[i] > 0){
12876: agecens[i]=agev[mw[j][i]][i];
12877: cens[i]= 1;
12878: }else if (ageexmed[i]< 1)
12879: cens[i]= -1;
12880: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12881: cens[i]=0 ;
1.126 brouard 12882: }
12883: else cens[i]=-1;
12884: }
12885:
12886: for (i=1;i<=NDIM;i++) {
12887: for (j=1;j<=NDIM;j++)
1.226 brouard 12888: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12889: }
12890:
1.302 brouard 12891: p[1]=0.0268; p[NDIM]=0.083;
12892: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12893:
12894:
1.136 brouard 12895: #ifdef GSL
12896: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12897: #else
1.126 brouard 12898: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12899: #endif
1.201 brouard 12900: strcpy(filerespow,"POW-MORT_");
12901: strcat(filerespow,fileresu);
1.126 brouard 12902: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12903: printf("Problem with resultfile: %s\n", filerespow);
12904: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12905: }
1.136 brouard 12906: #ifdef GSL
12907: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12908: #else
1.126 brouard 12909: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12910: #endif
1.126 brouard 12911: /* for (i=1;i<=nlstate;i++)
12912: for(j=1;j<=nlstate+ndeath;j++)
12913: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12914: */
12915: fprintf(ficrespow,"\n");
1.136 brouard 12916: #ifdef GSL
12917: /* gsl starts here */
12918: T = gsl_multimin_fminimizer_nmsimplex;
12919: gsl_multimin_fminimizer *sfm = NULL;
12920: gsl_vector *ss, *x;
12921: gsl_multimin_function minex_func;
12922:
12923: /* Initial vertex size vector */
12924: ss = gsl_vector_alloc (NDIM);
12925:
12926: if (ss == NULL){
12927: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12928: }
12929: /* Set all step sizes to 1 */
12930: gsl_vector_set_all (ss, 0.001);
12931:
12932: /* Starting point */
1.126 brouard 12933:
1.136 brouard 12934: x = gsl_vector_alloc (NDIM);
12935:
12936: if (x == NULL){
12937: gsl_vector_free(ss);
12938: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12939: }
12940:
12941: /* Initialize method and iterate */
12942: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12943: /* gsl_vector_set(x, 0, 0.0268); */
12944: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12945: gsl_vector_set(x, 0, p[1]);
12946: gsl_vector_set(x, 1, p[2]);
12947:
12948: minex_func.f = &gompertz_f;
12949: minex_func.n = NDIM;
12950: minex_func.params = (void *)&p; /* ??? */
12951:
12952: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12953: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12954:
12955: printf("Iterations beginning .....\n\n");
12956: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12957:
12958: iteri=0;
12959: while (rval == GSL_CONTINUE){
12960: iteri++;
12961: status = gsl_multimin_fminimizer_iterate(sfm);
12962:
12963: if (status) printf("error: %s\n", gsl_strerror (status));
12964: fflush(0);
12965:
12966: if (status)
12967: break;
12968:
12969: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12970: ssval = gsl_multimin_fminimizer_size (sfm);
12971:
12972: if (rval == GSL_SUCCESS)
12973: printf ("converged to a local maximum at\n");
12974:
12975: printf("%5d ", iteri);
12976: for (it = 0; it < NDIM; it++){
12977: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12978: }
12979: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12980: }
12981:
12982: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12983:
12984: gsl_vector_free(x); /* initial values */
12985: gsl_vector_free(ss); /* inital step size */
12986: for (it=0; it<NDIM; it++){
12987: p[it+1]=gsl_vector_get(sfm->x,it);
12988: fprintf(ficrespow," %.12lf", p[it]);
12989: }
12990: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12991: #endif
12992: #ifdef POWELL
12993: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12994: #endif
1.126 brouard 12995: fclose(ficrespow);
12996:
1.203 brouard 12997: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12998:
12999: for(i=1; i <=NDIM; i++)
13000: for(j=i+1;j<=NDIM;j++)
1.220 brouard 13001: matcov[i][j]=matcov[j][i];
1.126 brouard 13002:
13003: printf("\nCovariance matrix\n ");
1.203 brouard 13004: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 13005: for(i=1; i <=NDIM; i++) {
13006: for(j=1;j<=NDIM;j++){
1.220 brouard 13007: printf("%f ",matcov[i][j]);
13008: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 13009: }
1.203 brouard 13010: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 13011: }
13012:
13013: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 13014: for (i=1;i<=NDIM;i++) {
1.126 brouard 13015: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 13016: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
13017: }
1.302 brouard 13018: lsurv=vector(agegomp,AGESUP);
13019: lpop=vector(agegomp,AGESUP);
13020: tpop=vector(agegomp,AGESUP);
1.126 brouard 13021: lsurv[agegomp]=100000;
13022:
13023: for (k=agegomp;k<=AGESUP;k++) {
13024: agemortsup=k;
13025: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
13026: }
13027:
13028: for (k=agegomp;k<agemortsup;k++)
13029: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
13030:
13031: for (k=agegomp;k<agemortsup;k++){
13032: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
13033: sumlpop=sumlpop+lpop[k];
13034: }
13035:
13036: tpop[agegomp]=sumlpop;
13037: for (k=agegomp;k<(agemortsup-3);k++){
13038: /* tpop[k+1]=2;*/
13039: tpop[k+1]=tpop[k]-lpop[k];
13040: }
13041:
13042:
13043: printf("\nAge lx qx dx Lx Tx e(x)\n");
13044: for (k=agegomp;k<(agemortsup-2);k++)
13045: 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]);
13046:
13047:
13048: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 13049: ageminpar=50;
13050: agemaxpar=100;
1.194 brouard 13051: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
13052: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13053: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13054: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
13055: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13056: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13057: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13058: }else{
13059: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
13060: 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 13061: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 13062: }
1.201 brouard 13063: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 13064: stepm, weightopt,\
13065: model,imx,p,matcov,agemortsup);
13066:
1.302 brouard 13067: free_vector(lsurv,agegomp,AGESUP);
13068: free_vector(lpop,agegomp,AGESUP);
13069: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 13070: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 13071: free_ivector(dcwave,firstobs,lastobs);
13072: free_vector(agecens,firstobs,lastobs);
13073: free_vector(ageexmed,firstobs,lastobs);
13074: free_ivector(cens,firstobs,lastobs);
1.220 brouard 13075: #ifdef GSL
1.136 brouard 13076: #endif
1.186 brouard 13077: } /* Endof if mle==-3 mortality only */
1.205 brouard 13078: /* Standard */
13079: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
13080: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13081: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 13082: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 13083: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
13084: for (k=1; k<=npar;k++)
13085: printf(" %d %8.5f",k,p[k]);
13086: printf("\n");
1.205 brouard 13087: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
13088: /* mlikeli uses func not funcone */
1.247 brouard 13089: /* for(i=1;i<nlstate;i++){ */
13090: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
13091: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
13092: /* } */
1.205 brouard 13093: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
13094: }
13095: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
13096: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13097: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
13098: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13099: }
13100: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 13101: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13102: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 brouard 13103: /* exit(0); */
1.126 brouard 13104: for (k=1; k<=npar;k++)
13105: printf(" %d %8.5f",k,p[k]);
13106: printf("\n");
13107:
13108: /*--------- results files --------------*/
1.283 brouard 13109: /* 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 13110:
13111:
13112: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13113: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 13114: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13115:
13116: printf("#model= 1 + age ");
13117: fprintf(ficres,"#model= 1 + age ");
13118: fprintf(ficlog,"#model= 1 + age ");
13119: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
13120: </ul>", model);
13121:
13122: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
13123: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
13124: if(nagesqr==1){
13125: printf(" + age*age ");
13126: fprintf(ficres," + age*age ");
13127: fprintf(ficlog," + age*age ");
13128: fprintf(fichtm, "<th>+ age*age</th>");
13129: }
13130: for(j=1;j <=ncovmodel-2;j++){
13131: if(Typevar[j]==0) {
13132: printf(" + V%d ",Tvar[j]);
13133: fprintf(ficres," + V%d ",Tvar[j]);
13134: fprintf(ficlog," + V%d ",Tvar[j]);
13135: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13136: }else if(Typevar[j]==1) {
13137: printf(" + V%d*age ",Tvar[j]);
13138: fprintf(ficres," + V%d*age ",Tvar[j]);
13139: fprintf(ficlog," + V%d*age ",Tvar[j]);
13140: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13141: }else if(Typevar[j]==2) {
13142: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13143: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13144: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13145: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13146: }
13147: }
13148: printf("\n");
13149: fprintf(ficres,"\n");
13150: fprintf(ficlog,"\n");
13151: fprintf(fichtm, "</tr>");
13152: fprintf(fichtm, "\n");
13153:
13154:
1.126 brouard 13155: for(i=1,jk=1; i <=nlstate; i++){
13156: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 13157: if (k != i) {
1.319 brouard 13158: fprintf(fichtm, "<tr>");
1.225 brouard 13159: printf("%d%d ",i,k);
13160: fprintf(ficlog,"%d%d ",i,k);
13161: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 13162: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13163: for(j=1; j <=ncovmodel; j++){
13164: printf("%12.7f ",p[jk]);
13165: fprintf(ficlog,"%12.7f ",p[jk]);
13166: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 13167: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 13168: jk++;
13169: }
13170: printf("\n");
13171: fprintf(ficlog,"\n");
13172: fprintf(ficres,"\n");
1.319 brouard 13173: fprintf(fichtm, "</tr>\n");
1.225 brouard 13174: }
1.126 brouard 13175: }
13176: }
1.319 brouard 13177: /* fprintf(fichtm,"</tr>\n"); */
13178: fprintf(fichtm,"</table>\n");
13179: fprintf(fichtm, "\n");
13180:
1.203 brouard 13181: if(mle != 0){
13182: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 13183: ftolhess=ftol; /* Usually correct */
1.203 brouard 13184: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
13185: 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");
13186: 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 13187: 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 13188: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
13189: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
13190: if(nagesqr==1){
13191: printf(" + age*age ");
13192: fprintf(ficres," + age*age ");
13193: fprintf(ficlog," + age*age ");
13194: fprintf(fichtm, "<th>+ age*age</th>");
13195: }
13196: for(j=1;j <=ncovmodel-2;j++){
13197: if(Typevar[j]==0) {
13198: printf(" + V%d ",Tvar[j]);
13199: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13200: }else if(Typevar[j]==1) {
13201: printf(" + V%d*age ",Tvar[j]);
13202: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13203: }else if(Typevar[j]==2) {
13204: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13205: }
13206: }
13207: fprintf(fichtm, "</tr>\n");
13208:
1.203 brouard 13209: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 13210: for(k=1; k <=(nlstate+ndeath); k++){
13211: if (k != i) {
1.319 brouard 13212: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 13213: printf("%d%d ",i,k);
13214: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 13215: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13216: for(j=1; j <=ncovmodel; j++){
1.319 brouard 13217: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 13218: 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]));
13219: 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 13220: if(fabs(wald) > 1.96){
1.321 brouard 13221: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 13222: }else{
13223: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
13224: }
1.324 brouard 13225: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 13226: 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 13227: jk++;
13228: }
13229: printf("\n");
13230: fprintf(ficlog,"\n");
1.319 brouard 13231: fprintf(fichtm, "</tr>\n");
1.225 brouard 13232: }
13233: }
1.193 brouard 13234: }
1.203 brouard 13235: } /* end of hesscov and Wald tests */
1.319 brouard 13236: fprintf(fichtm,"</table>\n");
1.225 brouard 13237:
1.203 brouard 13238: /* */
1.126 brouard 13239: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
13240: printf("# Scales (for hessian or gradient estimation)\n");
13241: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
13242: for(i=1,jk=1; i <=nlstate; i++){
13243: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 13244: if (j!=i) {
13245: fprintf(ficres,"%1d%1d",i,j);
13246: printf("%1d%1d",i,j);
13247: fprintf(ficlog,"%1d%1d",i,j);
13248: for(k=1; k<=ncovmodel;k++){
13249: printf(" %.5e",delti[jk]);
13250: fprintf(ficlog," %.5e",delti[jk]);
13251: fprintf(ficres," %.5e",delti[jk]);
13252: jk++;
13253: }
13254: printf("\n");
13255: fprintf(ficlog,"\n");
13256: fprintf(ficres,"\n");
13257: }
1.126 brouard 13258: }
13259: }
13260:
13261: 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 13262: if(mle >= 1) /* To big for the screen */
1.126 brouard 13263: 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");
13264: 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");
13265: /* # 121 Var(a12)\n\ */
13266: /* # 122 Cov(b12,a12) Var(b12)\n\ */
13267: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
13268: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
13269: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
13270: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
13271: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
13272: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
13273:
13274:
13275: /* Just to have a covariance matrix which will be more understandable
13276: even is we still don't want to manage dictionary of variables
13277: */
13278: for(itimes=1;itimes<=2;itimes++){
13279: jj=0;
13280: for(i=1; i <=nlstate; i++){
1.225 brouard 13281: for(j=1; j <=nlstate+ndeath; j++){
13282: if(j==i) continue;
13283: for(k=1; k<=ncovmodel;k++){
13284: jj++;
13285: ca[0]= k+'a'-1;ca[1]='\0';
13286: if(itimes==1){
13287: if(mle>=1)
13288: printf("#%1d%1d%d",i,j,k);
13289: fprintf(ficlog,"#%1d%1d%d",i,j,k);
13290: fprintf(ficres,"#%1d%1d%d",i,j,k);
13291: }else{
13292: if(mle>=1)
13293: printf("%1d%1d%d",i,j,k);
13294: fprintf(ficlog,"%1d%1d%d",i,j,k);
13295: fprintf(ficres,"%1d%1d%d",i,j,k);
13296: }
13297: ll=0;
13298: for(li=1;li <=nlstate; li++){
13299: for(lj=1;lj <=nlstate+ndeath; lj++){
13300: if(lj==li) continue;
13301: for(lk=1;lk<=ncovmodel;lk++){
13302: ll++;
13303: if(ll<=jj){
13304: cb[0]= lk +'a'-1;cb[1]='\0';
13305: if(ll<jj){
13306: if(itimes==1){
13307: if(mle>=1)
13308: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13309: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13310: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13311: }else{
13312: if(mle>=1)
13313: printf(" %.5e",matcov[jj][ll]);
13314: fprintf(ficlog," %.5e",matcov[jj][ll]);
13315: fprintf(ficres," %.5e",matcov[jj][ll]);
13316: }
13317: }else{
13318: if(itimes==1){
13319: if(mle>=1)
13320: printf(" Var(%s%1d%1d)",ca,i,j);
13321: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
13322: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
13323: }else{
13324: if(mle>=1)
13325: printf(" %.7e",matcov[jj][ll]);
13326: fprintf(ficlog," %.7e",matcov[jj][ll]);
13327: fprintf(ficres," %.7e",matcov[jj][ll]);
13328: }
13329: }
13330: }
13331: } /* end lk */
13332: } /* end lj */
13333: } /* end li */
13334: if(mle>=1)
13335: printf("\n");
13336: fprintf(ficlog,"\n");
13337: fprintf(ficres,"\n");
13338: numlinepar++;
13339: } /* end k*/
13340: } /*end j */
1.126 brouard 13341: } /* end i */
13342: } /* end itimes */
13343:
13344: fflush(ficlog);
13345: fflush(ficres);
1.225 brouard 13346: while(fgets(line, MAXLINE, ficpar)) {
13347: /* If line starts with a # it is a comment */
13348: if (line[0] == '#') {
13349: numlinepar++;
13350: fputs(line,stdout);
13351: fputs(line,ficparo);
13352: fputs(line,ficlog);
1.299 brouard 13353: fputs(line,ficres);
1.225 brouard 13354: continue;
13355: }else
13356: break;
13357: }
13358:
1.209 brouard 13359: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
13360: /* ungetc(c,ficpar); */
13361: /* fgets(line, MAXLINE, ficpar); */
13362: /* fputs(line,stdout); */
13363: /* fputs(line,ficparo); */
13364: /* } */
13365: /* ungetc(c,ficpar); */
1.126 brouard 13366:
13367: estepm=0;
1.209 brouard 13368: 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 13369:
13370: if (num_filled != 6) {
13371: 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);
13372: 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);
13373: goto end;
13374: }
13375: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
13376: }
13377: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
13378: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
13379:
1.209 brouard 13380: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 13381: if (estepm==0 || estepm < stepm) estepm=stepm;
13382: if (fage <= 2) {
13383: bage = ageminpar;
13384: fage = agemaxpar;
13385: }
13386:
13387: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 13388: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
13389: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 13390:
1.186 brouard 13391: /* Other stuffs, more or less useful */
1.254 brouard 13392: while(fgets(line, MAXLINE, ficpar)) {
13393: /* If line starts with a # it is a comment */
13394: if (line[0] == '#') {
13395: numlinepar++;
13396: fputs(line,stdout);
13397: fputs(line,ficparo);
13398: fputs(line,ficlog);
1.299 brouard 13399: fputs(line,ficres);
1.254 brouard 13400: continue;
13401: }else
13402: break;
13403: }
13404:
13405: 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){
13406:
13407: if (num_filled != 7) {
13408: 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);
13409: 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);
13410: goto end;
13411: }
13412: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13413: 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);
13414: 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);
13415: 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 13416: }
1.254 brouard 13417:
13418: while(fgets(line, MAXLINE, ficpar)) {
13419: /* If line starts with a # it is a comment */
13420: if (line[0] == '#') {
13421: numlinepar++;
13422: fputs(line,stdout);
13423: fputs(line,ficparo);
13424: fputs(line,ficlog);
1.299 brouard 13425: fputs(line,ficres);
1.254 brouard 13426: continue;
13427: }else
13428: break;
1.126 brouard 13429: }
13430:
13431:
13432: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13433: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13434:
1.254 brouard 13435: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13436: if (num_filled != 1) {
13437: 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);
13438: 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);
13439: goto end;
13440: }
13441: printf("pop_based=%d\n",popbased);
13442: fprintf(ficlog,"pop_based=%d\n",popbased);
13443: fprintf(ficparo,"pop_based=%d\n",popbased);
13444: fprintf(ficres,"pop_based=%d\n",popbased);
13445: }
13446:
1.258 brouard 13447: /* Results */
1.332 brouard 13448: /* Value of covariate in each resultine will be compututed (if product) and sorted according to model rank */
13449: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
13450: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 13451: endishere=0;
1.258 brouard 13452: nresult=0;
1.308 brouard 13453: parameterline=0;
1.258 brouard 13454: do{
13455: if(!fgets(line, MAXLINE, ficpar)){
13456: endishere=1;
1.308 brouard 13457: parameterline=15;
1.258 brouard 13458: }else if (line[0] == '#') {
13459: /* If line starts with a # it is a comment */
1.254 brouard 13460: numlinepar++;
13461: fputs(line,stdout);
13462: fputs(line,ficparo);
13463: fputs(line,ficlog);
1.299 brouard 13464: fputs(line,ficres);
1.254 brouard 13465: continue;
1.258 brouard 13466: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13467: parameterline=11;
1.296 brouard 13468: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 13469: parameterline=12;
1.307 brouard 13470: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 13471: parameterline=13;
1.307 brouard 13472: }
1.258 brouard 13473: else{
13474: parameterline=14;
1.254 brouard 13475: }
1.308 brouard 13476: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 13477: case 11:
1.296 brouard 13478: 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)){
13479: 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 13480: 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);
13481: 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);
13482: 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);
13483: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 13484: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
13485: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 13486: prvforecast = 1;
13487: }
13488: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 13489: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13490: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13491: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 13492: prvforecast = 2;
13493: }
13494: else {
13495: 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);
13496: 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);
13497: goto end;
1.258 brouard 13498: }
1.254 brouard 13499: break;
1.258 brouard 13500: case 12:
1.296 brouard 13501: 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)){
13502: 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);
13503: 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);
13504: 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);
13505: 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);
13506: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 13507: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
13508: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 13509: prvbackcast = 1;
13510: }
13511: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 13512: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13513: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13514: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 13515: prvbackcast = 2;
13516: }
13517: else {
13518: 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);
13519: 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);
13520: goto end;
1.258 brouard 13521: }
1.230 brouard 13522: break;
1.258 brouard 13523: case 13:
1.332 brouard 13524: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 13525: nresult++; /* Sum of resultlines */
1.332 brouard 13526: printf("Result %d: result:%s\n",nresult, resultlineori);
13527: /* removefirstspace(&resultlineori); */
13528:
13529: if(strstr(resultlineori,"v") !=0){
13530: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
13531: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
13532: return 1;
13533: }
13534: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
13535: printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori);
1.318 brouard 13536: if(nresult > MAXRESULTLINESPONE-1){
13537: 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);
13538: 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 13539: goto end;
13540: }
1.332 brouard 13541:
1.310 brouard 13542: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 13543: fprintf(ficparo,"result: %s\n",resultline);
13544: fprintf(ficres,"result: %s\n",resultline);
13545: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 13546: } else
13547: goto end;
1.307 brouard 13548: break;
13549: case 14:
13550: printf("Error: Unknown command '%s'\n",line);
13551: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 13552: if(line[0] == ' ' || line[0] == '\n'){
13553: printf("It should not be an empty line '%s'\n",line);
13554: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
13555: }
1.307 brouard 13556: if(ncovmodel >=2 && nresult==0 ){
13557: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
13558: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 13559: }
1.307 brouard 13560: /* goto end; */
13561: break;
1.308 brouard 13562: case 15:
13563: printf("End of resultlines.\n");
13564: fprintf(ficlog,"End of resultlines.\n");
13565: break;
13566: default: /* parameterline =0 */
1.307 brouard 13567: nresult=1;
13568: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 13569: } /* End switch parameterline */
13570: }while(endishere==0); /* End do */
1.126 brouard 13571:
1.230 brouard 13572: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 13573: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 13574:
13575: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 13576: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 13577: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13578: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13579: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 13580: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13581: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13582: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13583: }else{
1.270 brouard 13584: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 13585: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
13586: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
13587: if(prvforecast==1){
13588: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
13589: jprojd=jproj1;
13590: mprojd=mproj1;
13591: anprojd=anproj1;
13592: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
13593: jprojf=jproj2;
13594: mprojf=mproj2;
13595: anprojf=anproj2;
13596: } else if(prvforecast == 2){
13597: dateprojd=dateintmean;
13598: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
13599: dateprojf=dateintmean+yrfproj;
13600: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
13601: }
13602: if(prvbackcast==1){
13603: datebackd=(jback1+12*mback1+365*anback1)/365;
13604: jbackd=jback1;
13605: mbackd=mback1;
13606: anbackd=anback1;
13607: datebackf=(jback2+12*mback2+365*anback2)/365;
13608: jbackf=jback2;
13609: mbackf=mback2;
13610: anbackf=anback2;
13611: } else if(prvbackcast == 2){
13612: datebackd=dateintmean;
13613: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
13614: datebackf=dateintmean-yrbproj;
13615: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
13616: }
13617:
13618: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 13619: }
13620: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 13621: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
13622: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 13623:
1.225 brouard 13624: /*------------ free_vector -------------*/
13625: /* chdir(path); */
1.220 brouard 13626:
1.215 brouard 13627: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
13628: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
13629: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
13630: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 13631: free_lvector(num,firstobs,lastobs);
13632: free_vector(agedc,firstobs,lastobs);
1.126 brouard 13633: /*free_matrix(covar,0,NCOVMAX,1,n);*/
13634: /*free_matrix(covar,1,NCOVMAX,1,n);*/
13635: fclose(ficparo);
13636: fclose(ficres);
1.220 brouard 13637:
13638:
1.186 brouard 13639: /* Other results (useful)*/
1.220 brouard 13640:
13641:
1.126 brouard 13642: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13643: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13644: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 13645: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 13646: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13647: fclose(ficrespl);
13648:
13649: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13650: /*#include "hpijx.h"*/
1.332 brouard 13651: /** 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?*/
13652: /* calls hpxij with combination k */
1.180 brouard 13653: hPijx(p, bage, fage);
1.145 brouard 13654: fclose(ficrespij);
1.227 brouard 13655:
1.220 brouard 13656: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 13657: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 13658: k=1;
1.126 brouard 13659: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13660:
1.269 brouard 13661: /* Prevalence for each covariate combination in probs[age][status][cov] */
13662: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13663: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 13664: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 13665: for(k=1;k<=ncovcombmax;k++)
13666: probs[i][j][k]=0.;
1.269 brouard 13667: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
13668: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 13669: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 13670: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13671: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 13672: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 13673: for(k=1;k<=ncovcombmax;k++)
13674: mobaverages[i][j][k]=0.;
1.219 brouard 13675: mobaverage=mobaverages;
13676: if (mobilav!=0) {
1.235 brouard 13677: printf("Movingaveraging observed prevalence\n");
1.258 brouard 13678: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 13679: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
13680: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
13681: printf(" Error in movingaverage mobilav=%d\n",mobilav);
13682: }
1.269 brouard 13683: } else if (mobilavproj !=0) {
1.235 brouard 13684: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 13685: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 13686: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
13687: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
13688: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
13689: }
1.269 brouard 13690: }else{
13691: printf("Internal error moving average\n");
13692: fflush(stdout);
13693: exit(1);
1.219 brouard 13694: }
13695: }/* end if moving average */
1.227 brouard 13696:
1.126 brouard 13697: /*---------- Forecasting ------------------*/
1.296 brouard 13698: if(prevfcast==1){
13699: /* /\* if(stepm ==1){*\/ */
13700: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13701: /*This done previously after freqsummary.*/
13702: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
13703: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
13704:
13705: /* } else if (prvforecast==2){ */
13706: /* /\* if(stepm ==1){*\/ */
13707: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13708: /* } */
13709: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
13710: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 13711: }
1.269 brouard 13712:
1.296 brouard 13713: /* Prevbcasting */
13714: if(prevbcast==1){
1.219 brouard 13715: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13716: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13717: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13718:
13719: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
13720:
13721: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 13722:
1.219 brouard 13723: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
13724: fclose(ficresplb);
13725:
1.222 brouard 13726: hBijx(p, bage, fage, mobaverage);
13727: fclose(ficrespijb);
1.219 brouard 13728:
1.296 brouard 13729: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
13730: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
13731: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
13732: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
13733: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
13734: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
13735:
13736:
1.269 brouard 13737: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13738:
13739:
1.269 brouard 13740: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 13741: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13742: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13743: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 13744: } /* end Prevbcasting */
1.268 brouard 13745:
1.186 brouard 13746:
13747: /* ------ Other prevalence ratios------------ */
1.126 brouard 13748:
1.215 brouard 13749: free_ivector(wav,1,imx);
13750: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
13751: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
13752: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 13753:
13754:
1.127 brouard 13755: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 13756:
1.201 brouard 13757: strcpy(filerese,"E_");
13758: strcat(filerese,fileresu);
1.126 brouard 13759: if((ficreseij=fopen(filerese,"w"))==NULL) {
13760: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13761: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13762: }
1.208 brouard 13763: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
13764: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 13765:
13766: pstamp(ficreseij);
1.219 brouard 13767:
1.235 brouard 13768: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13769: if (cptcovn < 1){i1=1;}
13770:
13771: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13772: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13773: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13774: continue;
1.219 brouard 13775: fprintf(ficreseij,"\n#****** ");
1.235 brouard 13776: printf("\n#****** ");
1.225 brouard 13777: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 13778: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
13779: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 13780: }
13781: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.334 brouard 13782: printf(" V%d=%f ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
13783: fprintf(ficreseij,"V%d=%f ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 13784: }
13785: fprintf(ficreseij,"******\n");
1.235 brouard 13786: printf("******\n");
1.219 brouard 13787:
13788: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13789: oldm=oldms;savm=savms;
1.330 brouard 13790: /* 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 13791: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 13792:
1.219 brouard 13793: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 13794: }
13795: fclose(ficreseij);
1.208 brouard 13796: printf("done evsij\n");fflush(stdout);
13797: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 13798:
1.218 brouard 13799:
1.227 brouard 13800: /*---------- State-specific expectancies and variances ------------*/
1.336 ! brouard 13801: /* Should be moved in a function */
1.201 brouard 13802: strcpy(filerest,"T_");
13803: strcat(filerest,fileresu);
1.127 brouard 13804: if((ficrest=fopen(filerest,"w"))==NULL) {
13805: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
13806: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
13807: }
1.208 brouard 13808: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
13809: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 13810: strcpy(fileresstde,"STDE_");
13811: strcat(fileresstde,fileresu);
1.126 brouard 13812: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 13813: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
13814: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 13815: }
1.227 brouard 13816: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
13817: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 13818:
1.201 brouard 13819: strcpy(filerescve,"CVE_");
13820: strcat(filerescve,fileresu);
1.126 brouard 13821: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 13822: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
13823: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 13824: }
1.227 brouard 13825: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
13826: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 13827:
1.201 brouard 13828: strcpy(fileresv,"V_");
13829: strcat(fileresv,fileresu);
1.126 brouard 13830: if((ficresvij=fopen(fileresv,"w"))==NULL) {
13831: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
13832: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
13833: }
1.227 brouard 13834: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
13835: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 13836:
1.235 brouard 13837: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13838: if (cptcovn < 1){i1=1;}
13839:
1.334 brouard 13840: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
13841: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
13842: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
13843: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
13844: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
13845: /* */
13846: 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 13847: continue;
1.321 brouard 13848: printf("\n# model %s \n#****** Result for:", model);
13849: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
13850: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.334 brouard 13851: /* It might not be a good idea to mix dummies and quantitative */
13852: /* 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 *\/ */
13853: 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 */
13854: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
13855: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
13856: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
13857: * (V5 is quanti) V4 and V3 are dummies
13858: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
13859: * l=1 l=2
13860: * k=1 1 1 0 0
13861: * k=2 2 1 1 0
13862: * k=3 [1] [2] 0 1
13863: * k=4 2 2 1 1
13864: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
13865: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
13866: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
13867: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
13868: */
13869: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
13870: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
13871: /* We give up with the combinations!! */
13872: 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 */
13873:
13874: if(Dummy[modelresult[nres][j]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to j in resultline */
13875: 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 */
13876: 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 */
13877: 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 */
13878: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
13879: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
13880: }else{
13881: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
13882: }
13883: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13884: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13885: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
13886: /* For each selected (single) quantitative value */
13887: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13888: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
13889: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
13890: }else{
13891: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
13892: }
13893: }else{
13894: 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 */
13895: 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 */
13896: exit(1);
13897: }
1.335 brouard 13898: } /* End loop for each variable in the resultline */
1.334 brouard 13899: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
13900: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
13901: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
13902: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
13903: /* } */
1.208 brouard 13904: fprintf(ficrest,"******\n");
1.227 brouard 13905: fprintf(ficlog,"******\n");
13906: printf("******\n");
1.208 brouard 13907:
13908: fprintf(ficresstdeij,"\n#****** ");
13909: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 13910: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 13911: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
13912: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13913: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13914: }
13915: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value, TvarsQind gives the position of a quantitative in model equation */
13916: fprintf(ficresstdeij," V%d=%f ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
13917: fprintf(ficrescveij," V%d=%f ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 13918: }
1.208 brouard 13919: fprintf(ficresstdeij,"******\n");
13920: fprintf(ficrescveij,"******\n");
13921:
13922: fprintf(ficresvij,"\n#****** ");
1.238 brouard 13923: /* pstamp(ficresvij); */
1.225 brouard 13924: for(j=1;j<=cptcoveff;j++)
1.335 brouard 13925: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
13926: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 13927: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 13928: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
13929: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 13930: }
1.208 brouard 13931: fprintf(ficresvij,"******\n");
13932:
13933: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13934: oldm=oldms;savm=savms;
1.235 brouard 13935: printf(" cvevsij ");
13936: fprintf(ficlog, " cvevsij ");
13937: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 13938: printf(" end cvevsij \n ");
13939: fprintf(ficlog, " end cvevsij \n ");
13940:
13941: /*
13942: */
13943: /* goto endfree; */
13944:
13945: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13946: pstamp(ficrest);
13947:
1.269 brouard 13948: epj=vector(1,nlstate+1);
1.208 brouard 13949: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 13950: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
13951: cptcod= 0; /* To be deleted */
13952: printf("varevsij vpopbased=%d \n",vpopbased);
13953: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 13954: 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 13955: 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 ");
13956: if(vpopbased==1)
13957: 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);
13958: else
1.288 brouard 13959: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.335 brouard 13960: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 13961: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
13962: fprintf(ficrest,"\n");
13963: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 13964: printf("Computing age specific forward period (stable) prevalences in each health state \n");
13965: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13966: for(age=bage; age <=fage ;age++){
1.235 brouard 13967: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 13968: if (vpopbased==1) {
13969: if(mobilav ==0){
13970: for(i=1; i<=nlstate;i++)
13971: prlim[i][i]=probs[(int)age][i][k];
13972: }else{ /* mobilav */
13973: for(i=1; i<=nlstate;i++)
13974: prlim[i][i]=mobaverage[(int)age][i][k];
13975: }
13976: }
1.219 brouard 13977:
1.227 brouard 13978: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
13979: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
13980: /* printf(" age %4.0f ",age); */
13981: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
13982: for(i=1, epj[j]=0.;i <=nlstate;i++) {
13983: epj[j] += prlim[i][i]*eij[i][j][(int)age];
13984: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
13985: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
13986: }
13987: epj[nlstate+1] +=epj[j];
13988: }
13989: /* printf(" age %4.0f \n",age); */
1.219 brouard 13990:
1.227 brouard 13991: for(i=1, vepp=0.;i <=nlstate;i++)
13992: for(j=1;j <=nlstate;j++)
13993: vepp += vareij[i][j][(int)age];
13994: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
13995: for(j=1;j <=nlstate;j++){
13996: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
13997: }
13998: fprintf(ficrest,"\n");
13999: }
1.208 brouard 14000: } /* End vpopbased */
1.269 brouard 14001: free_vector(epj,1,nlstate+1);
1.208 brouard 14002: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
14003: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 14004: printf("done selection\n");fflush(stdout);
14005: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 14006:
1.335 brouard 14007: } /* End k selection or end covariate selection for nres */
1.227 brouard 14008:
14009: printf("done State-specific expectancies\n");fflush(stdout);
14010: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
14011:
1.335 brouard 14012: /* variance-covariance of forward period prevalence */
1.269 brouard 14013: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14014:
1.227 brouard 14015:
1.290 brouard 14016: free_vector(weight,firstobs,lastobs);
1.330 brouard 14017: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 14018: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 14019: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
14020: free_matrix(anint,1,maxwav,firstobs,lastobs);
14021: free_matrix(mint,1,maxwav,firstobs,lastobs);
14022: free_ivector(cod,firstobs,lastobs);
1.227 brouard 14023: free_ivector(tab,1,NCOVMAX);
14024: fclose(ficresstdeij);
14025: fclose(ficrescveij);
14026: fclose(ficresvij);
14027: fclose(ficrest);
14028: fclose(ficpar);
14029:
14030:
1.126 brouard 14031: /*---------- End : free ----------------*/
1.219 brouard 14032: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 14033: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
14034: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 14035: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
14036: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 14037: } /* mle==-3 arrives here for freeing */
1.227 brouard 14038: /* endfree:*/
14039: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
14040: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
14041: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 14042: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
14043: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
14044: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
14045: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 14046: free_matrix(matcov,1,npar,1,npar);
14047: free_matrix(hess,1,npar,1,npar);
14048: /*free_vector(delti,1,npar);*/
14049: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14050: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 14051: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 14052: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14053:
14054: free_ivector(ncodemax,1,NCOVMAX);
14055: free_ivector(ncodemaxwundef,1,NCOVMAX);
14056: free_ivector(Dummy,-1,NCOVMAX);
14057: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 14058: free_ivector(DummyV,1,NCOVMAX);
14059: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 14060: free_ivector(Typevar,-1,NCOVMAX);
14061: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 14062: free_ivector(TvarsQ,1,NCOVMAX);
14063: free_ivector(TvarsQind,1,NCOVMAX);
14064: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 14065: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 14066: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 14067: free_ivector(TvarFD,1,NCOVMAX);
14068: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 14069: free_ivector(TvarF,1,NCOVMAX);
14070: free_ivector(TvarFind,1,NCOVMAX);
14071: free_ivector(TvarV,1,NCOVMAX);
14072: free_ivector(TvarVind,1,NCOVMAX);
14073: free_ivector(TvarA,1,NCOVMAX);
14074: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 14075: free_ivector(TvarFQ,1,NCOVMAX);
14076: free_ivector(TvarFQind,1,NCOVMAX);
14077: free_ivector(TvarVD,1,NCOVMAX);
14078: free_ivector(TvarVDind,1,NCOVMAX);
14079: free_ivector(TvarVQ,1,NCOVMAX);
14080: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 14081: free_ivector(Tvarsel,1,NCOVMAX);
14082: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 14083: free_ivector(Tposprod,1,NCOVMAX);
14084: free_ivector(Tprod,1,NCOVMAX);
14085: free_ivector(Tvaraff,1,NCOVMAX);
14086: free_ivector(invalidvarcomb,1,ncovcombmax);
14087: free_ivector(Tage,1,NCOVMAX);
14088: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 14089: free_ivector(TmodelInvind,1,NCOVMAX);
14090: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 14091:
14092: free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
14093:
1.227 brouard 14094: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
14095: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 14096: fflush(fichtm);
14097: fflush(ficgp);
14098:
1.227 brouard 14099:
1.126 brouard 14100: if((nberr >0) || (nbwarn>0)){
1.216 brouard 14101: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
14102: 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 14103: }else{
14104: printf("End of Imach\n");
14105: fprintf(ficlog,"End of Imach\n");
14106: }
14107: printf("See log file on %s\n",filelog);
14108: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 14109: /*(void) gettimeofday(&end_time,&tzp);*/
14110: rend_time = time(NULL);
14111: end_time = *localtime(&rend_time);
14112: /* tml = *localtime(&end_time.tm_sec); */
14113: strcpy(strtend,asctime(&end_time));
1.126 brouard 14114: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
14115: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 14116: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 14117:
1.157 brouard 14118: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
14119: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
14120: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 14121: /* printf("Total time was %d uSec.\n", total_usecs);*/
14122: /* if(fileappend(fichtm,optionfilehtm)){ */
14123: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14124: fclose(fichtm);
14125: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14126: fclose(fichtmcov);
14127: fclose(ficgp);
14128: fclose(ficlog);
14129: /*------ End -----------*/
1.227 brouard 14130:
1.281 brouard 14131:
14132: /* Executes gnuplot */
1.227 brouard 14133:
14134: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 14135: #ifdef WIN32
1.227 brouard 14136: if (_chdir(pathcd) != 0)
14137: printf("Can't move to directory %s!\n",path);
14138: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 14139: #else
1.227 brouard 14140: if(chdir(pathcd) != 0)
14141: printf("Can't move to directory %s!\n", path);
14142: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 14143: #endif
1.126 brouard 14144: printf("Current directory %s!\n",pathcd);
14145: /*strcat(plotcmd,CHARSEPARATOR);*/
14146: sprintf(plotcmd,"gnuplot");
1.157 brouard 14147: #ifdef _WIN32
1.126 brouard 14148: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
14149: #endif
14150: if(!stat(plotcmd,&info)){
1.158 brouard 14151: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14152: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 14153: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 14154: }else
14155: strcpy(pplotcmd,plotcmd);
1.157 brouard 14156: #ifdef __unix
1.126 brouard 14157: strcpy(plotcmd,GNUPLOTPROGRAM);
14158: if(!stat(plotcmd,&info)){
1.158 brouard 14159: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14160: }else
14161: strcpy(pplotcmd,plotcmd);
14162: #endif
14163: }else
14164: strcpy(pplotcmd,plotcmd);
14165:
14166: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 14167: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 14168: strcpy(pplotcmd,plotcmd);
1.227 brouard 14169:
1.126 brouard 14170: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 14171: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 14172: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 14173: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 14174: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 14175: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 14176: strcpy(plotcmd,pplotcmd);
14177: }
1.126 brouard 14178: }
1.158 brouard 14179: printf(" Successful, please wait...");
1.126 brouard 14180: while (z[0] != 'q') {
14181: /* chdir(path); */
1.154 brouard 14182: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 14183: scanf("%s",z);
14184: /* if (z[0] == 'c') system("./imach"); */
14185: if (z[0] == 'e') {
1.158 brouard 14186: #ifdef __APPLE__
1.152 brouard 14187: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 14188: #elif __linux
14189: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 14190: #else
1.152 brouard 14191: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 14192: #endif
14193: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
14194: system(pplotcmd);
1.126 brouard 14195: }
14196: else if (z[0] == 'g') system(plotcmd);
14197: else if (z[0] == 'q') exit(0);
14198: }
1.227 brouard 14199: end:
1.126 brouard 14200: while (z[0] != 'q') {
1.195 brouard 14201: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 14202: scanf("%s",z);
14203: }
1.283 brouard 14204: printf("End\n");
1.282 brouard 14205: exit(0);
1.126 brouard 14206: }
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