Annotation of imach/src/imach.c, revision 1.337
1.337 ! brouard 1: /* $Id: imach.c,v 1.336 2022/08/31 09:52:36 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.337 ! brouard 4: Revision 1.336 2022/08/31 09:52:36 brouard
! 5: *** empty log message ***
! 6:
1.336 brouard 7: Revision 1.335 2022/08/31 08:23:16 brouard
8: Summary: improvements...
9:
1.335 brouard 10: Revision 1.334 2022/08/25 09:08:41 brouard
11: Summary: In progress for quantitative
12:
1.334 brouard 13: Revision 1.333 2022/08/21 09:10:30 brouard
14: * src/imach.c (Module): Version 0.99r33 A lot of changes in
15: reassigning covariates: my first idea was that people will always
16: use the first covariate V1 into the model but in fact they are
17: producing data with many covariates and can use an equation model
18: with some of the covariate; it means that in a model V2+V3 instead
19: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
20: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
21: the equation model is restricted to two variables only (V2, V3)
22: and the combination for V2 should be codtabm(k,1) instead of
23: (codtabm(k,2), and the code should be
24: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
25: made. All of these should be simplified once a day like we did in
26: hpxij() for example by using precov[nres] which is computed in
27: decoderesult for each nres of each resultline. Loop should be done
28: on the equation model globally by distinguishing only product with
29: age (which are changing with age) and no more on type of
30: covariates, single dummies, single covariates.
31:
1.333 brouard 32: Revision 1.332 2022/08/21 09:06:25 brouard
33: Summary: Version 0.99r33
34:
35: * src/imach.c (Module): Version 0.99r33 A lot of changes in
36: reassigning covariates: my first idea was that people will always
37: use the first covariate V1 into the model but in fact they are
38: producing data with many covariates and can use an equation model
39: with some of the covariate; it means that in a model V2+V3 instead
40: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
41: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
42: the equation model is restricted to two variables only (V2, V3)
43: and the combination for V2 should be codtabm(k,1) instead of
44: (codtabm(k,2), and the code should be
45: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
46: made. All of these should be simplified once a day like we did in
47: hpxij() for example by using precov[nres] which is computed in
48: decoderesult for each nres of each resultline. Loop should be done
49: on the equation model globally by distinguishing only product with
50: age (which are changing with age) and no more on type of
51: covariates, single dummies, single covariates.
52:
1.332 brouard 53: Revision 1.331 2022/08/07 05:40:09 brouard
54: *** empty log message ***
55:
1.331 brouard 56: Revision 1.330 2022/08/06 07:18:25 brouard
57: Summary: last 0.99r31
58:
59: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
60:
1.330 brouard 61: Revision 1.329 2022/08/03 17:29:54 brouard
62: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
63:
1.329 brouard 64: Revision 1.328 2022/07/27 17:40:48 brouard
65: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
66:
1.328 brouard 67: Revision 1.327 2022/07/27 14:47:35 brouard
68: Summary: Still a problem for one-step probabilities in case of quantitative variables
69:
1.327 brouard 70: Revision 1.326 2022/07/26 17:33:55 brouard
71: Summary: some test with nres=1
72:
1.326 brouard 73: Revision 1.325 2022/07/25 14:27:23 brouard
74: Summary: r30
75:
76: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
77: coredumped, revealed by Feiuno, thank you.
78:
1.325 brouard 79: Revision 1.324 2022/07/23 17:44:26 brouard
80: *** empty log message ***
81:
1.324 brouard 82: Revision 1.323 2022/07/22 12:30:08 brouard
83: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
84:
1.323 brouard 85: Revision 1.322 2022/07/22 12:27:48 brouard
86: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
87:
1.322 brouard 88: Revision 1.321 2022/07/22 12:04:24 brouard
89: Summary: r28
90:
91: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
92:
1.321 brouard 93: Revision 1.320 2022/06/02 05:10:11 brouard
94: *** empty log message ***
95:
1.320 brouard 96: Revision 1.319 2022/06/02 04:45:11 brouard
97: * imach.c (Module): Adding the Wald tests from the log to the main
98: htm for better display of the maximum likelihood estimators.
99:
1.319 brouard 100: Revision 1.318 2022/05/24 08:10:59 brouard
101: * imach.c (Module): Some attempts to find a bug of wrong estimates
102: of confidencce intervals with product in the equation modelC
103:
1.318 brouard 104: Revision 1.317 2022/05/15 15:06:23 brouard
105: * imach.c (Module): Some minor improvements
106:
1.317 brouard 107: Revision 1.316 2022/05/11 15:11:31 brouard
108: Summary: r27
109:
1.316 brouard 110: Revision 1.315 2022/05/11 15:06:32 brouard
111: *** empty log message ***
112:
1.315 brouard 113: Revision 1.314 2022/04/13 17:43:09 brouard
114: * imach.c (Module): Adding link to text data files
115:
1.314 brouard 116: Revision 1.313 2022/04/11 15:57:42 brouard
117: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
118:
1.313 brouard 119: Revision 1.312 2022/04/05 21:24:39 brouard
120: *** empty log message ***
121:
1.312 brouard 122: Revision 1.311 2022/04/05 21:03:51 brouard
123: Summary: Fixed quantitative covariates
124:
125: Fixed covariates (dummy or quantitative)
126: with missing values have never been allowed but are ERRORS and
127: program quits. Standard deviations of fixed covariates were
128: wrongly computed. Mean and standard deviations of time varying
129: covariates are still not computed.
130:
1.311 brouard 131: Revision 1.310 2022/03/17 08:45:53 brouard
132: Summary: 99r25
133:
134: Improving detection of errors: result lines should be compatible with
135: the model.
136:
1.310 brouard 137: Revision 1.309 2021/05/20 12:39:14 brouard
138: Summary: Version 0.99r24
139:
1.309 brouard 140: Revision 1.308 2021/03/31 13:11:57 brouard
141: Summary: Version 0.99r23
142:
143:
144: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
145:
1.308 brouard 146: Revision 1.307 2021/03/08 18:11:32 brouard
147: Summary: 0.99r22 fixed bug on result:
148:
1.307 brouard 149: Revision 1.306 2021/02/20 15:44:02 brouard
150: Summary: Version 0.99r21
151:
152: * imach.c (Module): Fix bug on quitting after result lines!
153: (Module): Version 0.99r21
154:
1.306 brouard 155: Revision 1.305 2021/02/20 15:28:30 brouard
156: * imach.c (Module): Fix bug on quitting after result lines!
157:
1.305 brouard 158: Revision 1.304 2021/02/12 11:34:20 brouard
159: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
160:
1.304 brouard 161: Revision 1.303 2021/02/11 19:50:15 brouard
162: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
163:
1.303 brouard 164: Revision 1.302 2020/02/22 21:00:05 brouard
165: * (Module): imach.c Update mle=-3 (for computing Life expectancy
166: and life table from the data without any state)
167:
1.302 brouard 168: Revision 1.301 2019/06/04 13:51:20 brouard
169: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
170:
1.301 brouard 171: Revision 1.300 2019/05/22 19:09:45 brouard
172: Summary: version 0.99r19 of May 2019
173:
1.300 brouard 174: Revision 1.299 2019/05/22 18:37:08 brouard
175: Summary: Cleaned 0.99r19
176:
1.299 brouard 177: Revision 1.298 2019/05/22 18:19:56 brouard
178: *** empty log message ***
179:
1.298 brouard 180: Revision 1.297 2019/05/22 17:56:10 brouard
181: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
182:
1.297 brouard 183: Revision 1.296 2019/05/20 13:03:18 brouard
184: Summary: Projection syntax simplified
185:
186:
187: We can now start projections, forward or backward, from the mean date
188: of inteviews up to or down to a number of years of projection:
189: prevforecast=1 yearsfproj=15.3 mobil_average=0
190: or
191: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
192: or
193: prevbackcast=1 yearsbproj=12.3 mobil_average=1
194: or
195: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
196:
1.296 brouard 197: Revision 1.295 2019/05/18 09:52:50 brouard
198: Summary: doxygen tex bug
199:
1.295 brouard 200: Revision 1.294 2019/05/16 14:54:33 brouard
201: Summary: There was some wrong lines added
202:
1.294 brouard 203: Revision 1.293 2019/05/09 15:17:34 brouard
204: *** empty log message ***
205:
1.293 brouard 206: Revision 1.292 2019/05/09 14:17:20 brouard
207: Summary: Some updates
208:
1.292 brouard 209: Revision 1.291 2019/05/09 13:44:18 brouard
210: Summary: Before ncovmax
211:
1.291 brouard 212: Revision 1.290 2019/05/09 13:39:37 brouard
213: Summary: 0.99r18 unlimited number of individuals
214:
215: 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.
216:
1.290 brouard 217: Revision 1.289 2018/12/13 09:16:26 brouard
218: Summary: Bug for young ages (<-30) will be in r17
219:
1.289 brouard 220: Revision 1.288 2018/05/02 20:58:27 brouard
221: Summary: Some bugs fixed
222:
1.288 brouard 223: Revision 1.287 2018/05/01 17:57:25 brouard
224: Summary: Bug fixed by providing frequencies only for non missing covariates
225:
1.287 brouard 226: Revision 1.286 2018/04/27 14:27:04 brouard
227: Summary: some minor bugs
228:
1.286 brouard 229: Revision 1.285 2018/04/21 21:02:16 brouard
230: Summary: Some bugs fixed, valgrind tested
231:
1.285 brouard 232: Revision 1.284 2018/04/20 05:22:13 brouard
233: Summary: Computing mean and stdeviation of fixed quantitative variables
234:
1.284 brouard 235: Revision 1.283 2018/04/19 14:49:16 brouard
236: Summary: Some minor bugs fixed
237:
1.283 brouard 238: Revision 1.282 2018/02/27 22:50:02 brouard
239: *** empty log message ***
240:
1.282 brouard 241: Revision 1.281 2018/02/27 19:25:23 brouard
242: Summary: Adding second argument for quitting
243:
1.281 brouard 244: Revision 1.280 2018/02/21 07:58:13 brouard
245: Summary: 0.99r15
246:
247: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
248:
1.280 brouard 249: Revision 1.279 2017/07/20 13:35:01 brouard
250: Summary: temporary working
251:
1.279 brouard 252: Revision 1.278 2017/07/19 14:09:02 brouard
253: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
254:
1.278 brouard 255: Revision 1.277 2017/07/17 08:53:49 brouard
256: Summary: BOM files can be read now
257:
1.277 brouard 258: Revision 1.276 2017/06/30 15:48:31 brouard
259: Summary: Graphs improvements
260:
1.276 brouard 261: Revision 1.275 2017/06/30 13:39:33 brouard
262: Summary: Saito's color
263:
1.275 brouard 264: Revision 1.274 2017/06/29 09:47:08 brouard
265: Summary: Version 0.99r14
266:
1.274 brouard 267: Revision 1.273 2017/06/27 11:06:02 brouard
268: Summary: More documentation on projections
269:
1.273 brouard 270: Revision 1.272 2017/06/27 10:22:40 brouard
271: Summary: Color of backprojection changed from 6 to 5(yellow)
272:
1.272 brouard 273: Revision 1.271 2017/06/27 10:17:50 brouard
274: Summary: Some bug with rint
275:
1.271 brouard 276: Revision 1.270 2017/05/24 05:45:29 brouard
277: *** empty log message ***
278:
1.270 brouard 279: Revision 1.269 2017/05/23 08:39:25 brouard
280: Summary: Code into subroutine, cleanings
281:
1.269 brouard 282: Revision 1.268 2017/05/18 20:09:32 brouard
283: Summary: backprojection and confidence intervals of backprevalence
284:
1.268 brouard 285: Revision 1.267 2017/05/13 10:25:05 brouard
286: Summary: temporary save for backprojection
287:
1.267 brouard 288: Revision 1.266 2017/05/13 07:26:12 brouard
289: Summary: Version 0.99r13 (improvements and bugs fixed)
290:
1.266 brouard 291: Revision 1.265 2017/04/26 16:22:11 brouard
292: Summary: imach 0.99r13 Some bugs fixed
293:
1.265 brouard 294: Revision 1.264 2017/04/26 06:01:29 brouard
295: Summary: Labels in graphs
296:
1.264 brouard 297: Revision 1.263 2017/04/24 15:23:15 brouard
298: Summary: to save
299:
1.263 brouard 300: Revision 1.262 2017/04/18 16:48:12 brouard
301: *** empty log message ***
302:
1.262 brouard 303: Revision 1.261 2017/04/05 10:14:09 brouard
304: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
305:
1.261 brouard 306: Revision 1.260 2017/04/04 17:46:59 brouard
307: Summary: Gnuplot indexations fixed (humm)
308:
1.260 brouard 309: Revision 1.259 2017/04/04 13:01:16 brouard
310: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
311:
1.259 brouard 312: Revision 1.258 2017/04/03 10:17:47 brouard
313: Summary: Version 0.99r12
314:
315: Some cleanings, conformed with updated documentation.
316:
1.258 brouard 317: Revision 1.257 2017/03/29 16:53:30 brouard
318: Summary: Temp
319:
1.257 brouard 320: Revision 1.256 2017/03/27 05:50:23 brouard
321: Summary: Temporary
322:
1.256 brouard 323: Revision 1.255 2017/03/08 16:02:28 brouard
324: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
325:
1.255 brouard 326: Revision 1.254 2017/03/08 07:13:00 brouard
327: Summary: Fixing data parameter line
328:
1.254 brouard 329: Revision 1.253 2016/12/15 11:59:41 brouard
330: Summary: 0.99 in progress
331:
1.253 brouard 332: Revision 1.252 2016/09/15 21:15:37 brouard
333: *** empty log message ***
334:
1.252 brouard 335: Revision 1.251 2016/09/15 15:01:13 brouard
336: Summary: not working
337:
1.251 brouard 338: Revision 1.250 2016/09/08 16:07:27 brouard
339: Summary: continue
340:
1.250 brouard 341: Revision 1.249 2016/09/07 17:14:18 brouard
342: Summary: Starting values from frequencies
343:
1.249 brouard 344: Revision 1.248 2016/09/07 14:10:18 brouard
345: *** empty log message ***
346:
1.248 brouard 347: Revision 1.247 2016/09/02 11:11:21 brouard
348: *** empty log message ***
349:
1.247 brouard 350: Revision 1.246 2016/09/02 08:49:22 brouard
351: *** empty log message ***
352:
1.246 brouard 353: Revision 1.245 2016/09/02 07:25:01 brouard
354: *** empty log message ***
355:
1.245 brouard 356: Revision 1.244 2016/09/02 07:17:34 brouard
357: *** empty log message ***
358:
1.244 brouard 359: Revision 1.243 2016/09/02 06:45:35 brouard
360: *** empty log message ***
361:
1.243 brouard 362: Revision 1.242 2016/08/30 15:01:20 brouard
363: Summary: Fixing a lots
364:
1.242 brouard 365: Revision 1.241 2016/08/29 17:17:25 brouard
366: Summary: gnuplot problem in Back projection to fix
367:
1.241 brouard 368: Revision 1.240 2016/08/29 07:53:18 brouard
369: Summary: Better
370:
1.240 brouard 371: Revision 1.239 2016/08/26 15:51:03 brouard
372: Summary: Improvement in Powell output in order to copy and paste
373:
374: Author:
375:
1.239 brouard 376: Revision 1.238 2016/08/26 14:23:35 brouard
377: Summary: Starting tests of 0.99
378:
1.238 brouard 379: Revision 1.237 2016/08/26 09:20:19 brouard
380: Summary: to valgrind
381:
1.237 brouard 382: Revision 1.236 2016/08/25 10:50:18 brouard
383: *** empty log message ***
384:
1.236 brouard 385: Revision 1.235 2016/08/25 06:59:23 brouard
386: *** empty log message ***
387:
1.235 brouard 388: Revision 1.234 2016/08/23 16:51:20 brouard
389: *** empty log message ***
390:
1.234 brouard 391: Revision 1.233 2016/08/23 07:40:50 brouard
392: Summary: not working
393:
1.233 brouard 394: Revision 1.232 2016/08/22 14:20:21 brouard
395: Summary: not working
396:
1.232 brouard 397: Revision 1.231 2016/08/22 07:17:15 brouard
398: Summary: not working
399:
1.231 brouard 400: Revision 1.230 2016/08/22 06:55:53 brouard
401: Summary: Not working
402:
1.230 brouard 403: Revision 1.229 2016/07/23 09:45:53 brouard
404: Summary: Completing for func too
405:
1.229 brouard 406: Revision 1.228 2016/07/22 17:45:30 brouard
407: Summary: Fixing some arrays, still debugging
408:
1.227 brouard 409: Revision 1.226 2016/07/12 18:42:34 brouard
410: Summary: temp
411:
1.226 brouard 412: Revision 1.225 2016/07/12 08:40:03 brouard
413: Summary: saving but not running
414:
1.225 brouard 415: Revision 1.224 2016/07/01 13:16:01 brouard
416: Summary: Fixes
417:
1.224 brouard 418: Revision 1.223 2016/02/19 09:23:35 brouard
419: Summary: temporary
420:
1.223 brouard 421: Revision 1.222 2016/02/17 08:14:50 brouard
422: Summary: Probably last 0.98 stable version 0.98r6
423:
1.222 brouard 424: Revision 1.221 2016/02/15 23:35:36 brouard
425: Summary: minor bug
426:
1.220 brouard 427: Revision 1.219 2016/02/15 00:48:12 brouard
428: *** empty log message ***
429:
1.219 brouard 430: Revision 1.218 2016/02/12 11:29:23 brouard
431: Summary: 0.99 Back projections
432:
1.218 brouard 433: Revision 1.217 2015/12/23 17:18:31 brouard
434: Summary: Experimental backcast
435:
1.217 brouard 436: Revision 1.216 2015/12/18 17:32:11 brouard
437: Summary: 0.98r4 Warning and status=-2
438:
439: Version 0.98r4 is now:
440: - displaying an error when status is -1, date of interview unknown and date of death known;
441: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
442: Older changes concerning s=-2, dating from 2005 have been supersed.
443:
1.216 brouard 444: Revision 1.215 2015/12/16 08:52:24 brouard
445: Summary: 0.98r4 working
446:
1.215 brouard 447: Revision 1.214 2015/12/16 06:57:54 brouard
448: Summary: temporary not working
449:
1.214 brouard 450: Revision 1.213 2015/12/11 18:22:17 brouard
451: Summary: 0.98r4
452:
1.213 brouard 453: Revision 1.212 2015/11/21 12:47:24 brouard
454: Summary: minor typo
455:
1.212 brouard 456: Revision 1.211 2015/11/21 12:41:11 brouard
457: Summary: 0.98r3 with some graph of projected cross-sectional
458:
459: Author: Nicolas Brouard
460:
1.211 brouard 461: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 462: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 463: Summary: Adding ftolpl parameter
464: Author: N Brouard
465:
466: We had difficulties to get smoothed confidence intervals. It was due
467: to the period prevalence which wasn't computed accurately. The inner
468: parameter ftolpl is now an outer parameter of the .imach parameter
469: file after estepm. If ftolpl is small 1.e-4 and estepm too,
470: computation are long.
471:
1.209 brouard 472: Revision 1.208 2015/11/17 14:31:57 brouard
473: Summary: temporary
474:
1.208 brouard 475: Revision 1.207 2015/10/27 17:36:57 brouard
476: *** empty log message ***
477:
1.207 brouard 478: Revision 1.206 2015/10/24 07:14:11 brouard
479: *** empty log message ***
480:
1.206 brouard 481: Revision 1.205 2015/10/23 15:50:53 brouard
482: Summary: 0.98r3 some clarification for graphs on likelihood contributions
483:
1.205 brouard 484: Revision 1.204 2015/10/01 16:20:26 brouard
485: Summary: Some new graphs of contribution to likelihood
486:
1.204 brouard 487: Revision 1.203 2015/09/30 17:45:14 brouard
488: Summary: looking at better estimation of the hessian
489:
490: Also a better criteria for convergence to the period prevalence And
491: therefore adding the number of years needed to converge. (The
492: prevalence in any alive state shold sum to one
493:
1.203 brouard 494: Revision 1.202 2015/09/22 19:45:16 brouard
495: Summary: Adding some overall graph on contribution to likelihood. Might change
496:
1.202 brouard 497: Revision 1.201 2015/09/15 17:34:58 brouard
498: Summary: 0.98r0
499:
500: - Some new graphs like suvival functions
501: - Some bugs fixed like model=1+age+V2.
502:
1.201 brouard 503: Revision 1.200 2015/09/09 16:53:55 brouard
504: Summary: Big bug thanks to Flavia
505:
506: Even model=1+age+V2. did not work anymore
507:
1.200 brouard 508: Revision 1.199 2015/09/07 14:09:23 brouard
509: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
510:
1.199 brouard 511: Revision 1.198 2015/09/03 07:14:39 brouard
512: Summary: 0.98q5 Flavia
513:
1.198 brouard 514: Revision 1.197 2015/09/01 18:24:39 brouard
515: *** empty log message ***
516:
1.197 brouard 517: Revision 1.196 2015/08/18 23:17:52 brouard
518: Summary: 0.98q5
519:
1.196 brouard 520: Revision 1.195 2015/08/18 16:28:39 brouard
521: Summary: Adding a hack for testing purpose
522:
523: After reading the title, ftol and model lines, if the comment line has
524: a q, starting with #q, the answer at the end of the run is quit. It
525: permits to run test files in batch with ctest. The former workaround was
526: $ echo q | imach foo.imach
527:
1.195 brouard 528: Revision 1.194 2015/08/18 13:32:00 brouard
529: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
530:
1.194 brouard 531: Revision 1.193 2015/08/04 07:17:42 brouard
532: Summary: 0.98q4
533:
1.193 brouard 534: Revision 1.192 2015/07/16 16:49:02 brouard
535: Summary: Fixing some outputs
536:
1.192 brouard 537: Revision 1.191 2015/07/14 10:00:33 brouard
538: Summary: Some fixes
539:
1.191 brouard 540: Revision 1.190 2015/05/05 08:51:13 brouard
541: Summary: Adding digits in output parameters (7 digits instead of 6)
542:
543: Fix 1+age+.
544:
1.190 brouard 545: Revision 1.189 2015/04/30 14:45:16 brouard
546: Summary: 0.98q2
547:
1.189 brouard 548: Revision 1.188 2015/04/30 08:27:53 brouard
549: *** empty log message ***
550:
1.188 brouard 551: Revision 1.187 2015/04/29 09:11:15 brouard
552: *** empty log message ***
553:
1.187 brouard 554: Revision 1.186 2015/04/23 12:01:52 brouard
555: Summary: V1*age is working now, version 0.98q1
556:
557: Some codes had been disabled in order to simplify and Vn*age was
558: working in the optimization phase, ie, giving correct MLE parameters,
559: but, as usual, outputs were not correct and program core dumped.
560:
1.186 brouard 561: Revision 1.185 2015/03/11 13:26:42 brouard
562: Summary: Inclusion of compile and links command line for Intel Compiler
563:
1.185 brouard 564: Revision 1.184 2015/03/11 11:52:39 brouard
565: Summary: Back from Windows 8. Intel Compiler
566:
1.184 brouard 567: Revision 1.183 2015/03/10 20:34:32 brouard
568: Summary: 0.98q0, trying with directest, mnbrak fixed
569:
570: We use directest instead of original Powell test; probably no
571: incidence on the results, but better justifications;
572: We fixed Numerical Recipes mnbrak routine which was wrong and gave
573: wrong results.
574:
1.183 brouard 575: Revision 1.182 2015/02/12 08:19:57 brouard
576: Summary: Trying to keep directest which seems simpler and more general
577: Author: Nicolas Brouard
578:
1.182 brouard 579: Revision 1.181 2015/02/11 23:22:24 brouard
580: Summary: Comments on Powell added
581:
582: Author:
583:
1.181 brouard 584: Revision 1.180 2015/02/11 17:33:45 brouard
585: Summary: Finishing move from main to function (hpijx and prevalence_limit)
586:
1.180 brouard 587: Revision 1.179 2015/01/04 09:57:06 brouard
588: Summary: back to OS/X
589:
1.179 brouard 590: Revision 1.178 2015/01/04 09:35:48 brouard
591: *** empty log message ***
592:
1.178 brouard 593: Revision 1.177 2015/01/03 18:40:56 brouard
594: Summary: Still testing ilc32 on OSX
595:
1.177 brouard 596: Revision 1.176 2015/01/03 16:45:04 brouard
597: *** empty log message ***
598:
1.176 brouard 599: Revision 1.175 2015/01/03 16:33:42 brouard
600: *** empty log message ***
601:
1.175 brouard 602: Revision 1.174 2015/01/03 16:15:49 brouard
603: Summary: Still in cross-compilation
604:
1.174 brouard 605: Revision 1.173 2015/01/03 12:06:26 brouard
606: Summary: trying to detect cross-compilation
607:
1.173 brouard 608: Revision 1.172 2014/12/27 12:07:47 brouard
609: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
610:
1.172 brouard 611: Revision 1.171 2014/12/23 13:26:59 brouard
612: Summary: Back from Visual C
613:
614: Still problem with utsname.h on Windows
615:
1.171 brouard 616: Revision 1.170 2014/12/23 11:17:12 brouard
617: Summary: Cleaning some \%% back to %%
618:
619: The escape was mandatory for a specific compiler (which one?), but too many warnings.
620:
1.170 brouard 621: Revision 1.169 2014/12/22 23:08:31 brouard
622: Summary: 0.98p
623:
624: Outputs some informations on compiler used, OS etc. Testing on different platforms.
625:
1.169 brouard 626: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 627: Summary: update
1.169 brouard 628:
1.168 brouard 629: Revision 1.167 2014/12/22 13:50:56 brouard
630: Summary: Testing uname and compiler version and if compiled 32 or 64
631:
632: Testing on Linux 64
633:
1.167 brouard 634: Revision 1.166 2014/12/22 11:40:47 brouard
635: *** empty log message ***
636:
1.166 brouard 637: Revision 1.165 2014/12/16 11:20:36 brouard
638: Summary: After compiling on Visual C
639:
640: * imach.c (Module): Merging 1.61 to 1.162
641:
1.165 brouard 642: Revision 1.164 2014/12/16 10:52:11 brouard
643: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
644:
645: * imach.c (Module): Merging 1.61 to 1.162
646:
1.164 brouard 647: Revision 1.163 2014/12/16 10:30:11 brouard
648: * imach.c (Module): Merging 1.61 to 1.162
649:
1.163 brouard 650: Revision 1.162 2014/09/25 11:43:39 brouard
651: Summary: temporary backup 0.99!
652:
1.162 brouard 653: Revision 1.1 2014/09/16 11:06:58 brouard
654: Summary: With some code (wrong) for nlopt
655:
656: Author:
657:
658: Revision 1.161 2014/09/15 20:41:41 brouard
659: Summary: Problem with macro SQR on Intel compiler
660:
1.161 brouard 661: Revision 1.160 2014/09/02 09:24:05 brouard
662: *** empty log message ***
663:
1.160 brouard 664: Revision 1.159 2014/09/01 10:34:10 brouard
665: Summary: WIN32
666: Author: Brouard
667:
1.159 brouard 668: Revision 1.158 2014/08/27 17:11:51 brouard
669: *** empty log message ***
670:
1.158 brouard 671: Revision 1.157 2014/08/27 16:26:55 brouard
672: Summary: Preparing windows Visual studio version
673: Author: Brouard
674:
675: In order to compile on Visual studio, time.h is now correct and time_t
676: and tm struct should be used. difftime should be used but sometimes I
677: just make the differences in raw time format (time(&now).
678: Trying to suppress #ifdef LINUX
679: Add xdg-open for __linux in order to open default browser.
680:
1.157 brouard 681: Revision 1.156 2014/08/25 20:10:10 brouard
682: *** empty log message ***
683:
1.156 brouard 684: Revision 1.155 2014/08/25 18:32:34 brouard
685: Summary: New compile, minor changes
686: Author: Brouard
687:
1.155 brouard 688: Revision 1.154 2014/06/20 17:32:08 brouard
689: Summary: Outputs now all graphs of convergence to period prevalence
690:
1.154 brouard 691: Revision 1.153 2014/06/20 16:45:46 brouard
692: Summary: If 3 live state, convergence to period prevalence on same graph
693: Author: Brouard
694:
1.153 brouard 695: Revision 1.152 2014/06/18 17:54:09 brouard
696: Summary: open browser, use gnuplot on same dir than imach if not found in the path
697:
1.152 brouard 698: Revision 1.151 2014/06/18 16:43:30 brouard
699: *** empty log message ***
700:
1.151 brouard 701: Revision 1.150 2014/06/18 16:42:35 brouard
702: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
703: Author: brouard
704:
1.150 brouard 705: Revision 1.149 2014/06/18 15:51:14 brouard
706: Summary: Some fixes in parameter files errors
707: Author: Nicolas Brouard
708:
1.149 brouard 709: Revision 1.148 2014/06/17 17:38:48 brouard
710: Summary: Nothing new
711: Author: Brouard
712:
713: Just a new packaging for OS/X version 0.98nS
714:
1.148 brouard 715: Revision 1.147 2014/06/16 10:33:11 brouard
716: *** empty log message ***
717:
1.147 brouard 718: Revision 1.146 2014/06/16 10:20:28 brouard
719: Summary: Merge
720: Author: Brouard
721:
722: Merge, before building revised version.
723:
1.146 brouard 724: Revision 1.145 2014/06/10 21:23:15 brouard
725: Summary: Debugging with valgrind
726: Author: Nicolas Brouard
727:
728: Lot of changes in order to output the results with some covariates
729: After the Edimburgh REVES conference 2014, it seems mandatory to
730: improve the code.
731: No more memory valgrind error but a lot has to be done in order to
732: continue the work of splitting the code into subroutines.
733: Also, decodemodel has been improved. Tricode is still not
734: optimal. nbcode should be improved. Documentation has been added in
735: the source code.
736:
1.144 brouard 737: Revision 1.143 2014/01/26 09:45:38 brouard
738: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
739:
740: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
741: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
742:
1.143 brouard 743: Revision 1.142 2014/01/26 03:57:36 brouard
744: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
745:
746: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
747:
1.142 brouard 748: Revision 1.141 2014/01/26 02:42:01 brouard
749: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
750:
1.141 brouard 751: Revision 1.140 2011/09/02 10:37:54 brouard
752: Summary: times.h is ok with mingw32 now.
753:
1.140 brouard 754: Revision 1.139 2010/06/14 07:50:17 brouard
755: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
756: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
757:
1.139 brouard 758: Revision 1.138 2010/04/30 18:19:40 brouard
759: *** empty log message ***
760:
1.138 brouard 761: Revision 1.137 2010/04/29 18:11:38 brouard
762: (Module): Checking covariates for more complex models
763: than V1+V2. A lot of change to be done. Unstable.
764:
1.137 brouard 765: Revision 1.136 2010/04/26 20:30:53 brouard
766: (Module): merging some libgsl code. Fixing computation
767: of likelione (using inter/intrapolation if mle = 0) in order to
768: get same likelihood as if mle=1.
769: Some cleaning of code and comments added.
770:
1.136 brouard 771: Revision 1.135 2009/10/29 15:33:14 brouard
772: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
773:
1.135 brouard 774: Revision 1.134 2009/10/29 13:18:53 brouard
775: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
776:
1.134 brouard 777: Revision 1.133 2009/07/06 10:21:25 brouard
778: just nforces
779:
1.133 brouard 780: Revision 1.132 2009/07/06 08:22:05 brouard
781: Many tings
782:
1.132 brouard 783: Revision 1.131 2009/06/20 16:22:47 brouard
784: Some dimensions resccaled
785:
1.131 brouard 786: Revision 1.130 2009/05/26 06:44:34 brouard
787: (Module): Max Covariate is now set to 20 instead of 8. A
788: lot of cleaning with variables initialized to 0. Trying to make
789: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
790:
1.130 brouard 791: Revision 1.129 2007/08/31 13:49:27 lievre
792: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
793:
1.129 lievre 794: Revision 1.128 2006/06/30 13:02:05 brouard
795: (Module): Clarifications on computing e.j
796:
1.128 brouard 797: Revision 1.127 2006/04/28 18:11:50 brouard
798: (Module): Yes the sum of survivors was wrong since
799: imach-114 because nhstepm was no more computed in the age
800: loop. Now we define nhstepma in the age loop.
801: (Module): In order to speed up (in case of numerous covariates) we
802: compute health expectancies (without variances) in a first step
803: and then all the health expectancies with variances or standard
804: deviation (needs data from the Hessian matrices) which slows the
805: computation.
806: In the future we should be able to stop the program is only health
807: expectancies and graph are needed without standard deviations.
808:
1.127 brouard 809: Revision 1.126 2006/04/28 17:23:28 brouard
810: (Module): Yes the sum of survivors was wrong since
811: imach-114 because nhstepm was no more computed in the age
812: loop. Now we define nhstepma in the age loop.
813: Version 0.98h
814:
1.126 brouard 815: Revision 1.125 2006/04/04 15:20:31 lievre
816: Errors in calculation of health expectancies. Age was not initialized.
817: Forecasting file added.
818:
819: Revision 1.124 2006/03/22 17:13:53 lievre
820: Parameters are printed with %lf instead of %f (more numbers after the comma).
821: The log-likelihood is printed in the log file
822:
823: Revision 1.123 2006/03/20 10:52:43 brouard
824: * imach.c (Module): <title> changed, corresponds to .htm file
825: name. <head> headers where missing.
826:
827: * imach.c (Module): Weights can have a decimal point as for
828: English (a comma might work with a correct LC_NUMERIC environment,
829: otherwise the weight is truncated).
830: Modification of warning when the covariates values are not 0 or
831: 1.
832: Version 0.98g
833:
834: Revision 1.122 2006/03/20 09:45:41 brouard
835: (Module): Weights can have a decimal point as for
836: English (a comma might work with a correct LC_NUMERIC environment,
837: otherwise the weight is truncated).
838: Modification of warning when the covariates values are not 0 or
839: 1.
840: Version 0.98g
841:
842: Revision 1.121 2006/03/16 17:45:01 lievre
843: * imach.c (Module): Comments concerning covariates added
844:
845: * imach.c (Module): refinements in the computation of lli if
846: status=-2 in order to have more reliable computation if stepm is
847: not 1 month. Version 0.98f
848:
849: Revision 1.120 2006/03/16 15:10:38 lievre
850: (Module): refinements in the computation of lli if
851: status=-2 in order to have more reliable computation if stepm is
852: not 1 month. Version 0.98f
853:
854: Revision 1.119 2006/03/15 17:42:26 brouard
855: (Module): Bug if status = -2, the loglikelihood was
856: computed as likelihood omitting the logarithm. Version O.98e
857:
858: Revision 1.118 2006/03/14 18:20:07 brouard
859: (Module): varevsij Comments added explaining the second
860: table of variances if popbased=1 .
861: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
862: (Module): Function pstamp added
863: (Module): Version 0.98d
864:
865: Revision 1.117 2006/03/14 17:16:22 brouard
866: (Module): varevsij Comments added explaining the second
867: table of variances if popbased=1 .
868: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
869: (Module): Function pstamp added
870: (Module): Version 0.98d
871:
872: Revision 1.116 2006/03/06 10:29:27 brouard
873: (Module): Variance-covariance wrong links and
874: varian-covariance of ej. is needed (Saito).
875:
876: Revision 1.115 2006/02/27 12:17:45 brouard
877: (Module): One freematrix added in mlikeli! 0.98c
878:
879: Revision 1.114 2006/02/26 12:57:58 brouard
880: (Module): Some improvements in processing parameter
881: filename with strsep.
882:
883: Revision 1.113 2006/02/24 14:20:24 brouard
884: (Module): Memory leaks checks with valgrind and:
885: datafile was not closed, some imatrix were not freed and on matrix
886: allocation too.
887:
888: Revision 1.112 2006/01/30 09:55:26 brouard
889: (Module): Back to gnuplot.exe instead of wgnuplot.exe
890:
891: Revision 1.111 2006/01/25 20:38:18 brouard
892: (Module): Lots of cleaning and bugs added (Gompertz)
893: (Module): Comments can be added in data file. Missing date values
894: can be a simple dot '.'.
895:
896: Revision 1.110 2006/01/25 00:51:50 brouard
897: (Module): Lots of cleaning and bugs added (Gompertz)
898:
899: Revision 1.109 2006/01/24 19:37:15 brouard
900: (Module): Comments (lines starting with a #) are allowed in data.
901:
902: Revision 1.108 2006/01/19 18:05:42 lievre
903: Gnuplot problem appeared...
904: To be fixed
905:
906: Revision 1.107 2006/01/19 16:20:37 brouard
907: Test existence of gnuplot in imach path
908:
909: Revision 1.106 2006/01/19 13:24:36 brouard
910: Some cleaning and links added in html output
911:
912: Revision 1.105 2006/01/05 20:23:19 lievre
913: *** empty log message ***
914:
915: Revision 1.104 2005/09/30 16:11:43 lievre
916: (Module): sump fixed, loop imx fixed, and simplifications.
917: (Module): If the status is missing at the last wave but we know
918: that the person is alive, then we can code his/her status as -2
919: (instead of missing=-1 in earlier versions) and his/her
920: contributions to the likelihood is 1 - Prob of dying from last
921: health status (= 1-p13= p11+p12 in the easiest case of somebody in
922: the healthy state at last known wave). Version is 0.98
923:
924: Revision 1.103 2005/09/30 15:54:49 lievre
925: (Module): sump fixed, loop imx fixed, and simplifications.
926:
927: Revision 1.102 2004/09/15 17:31:30 brouard
928: Add the possibility to read data file including tab characters.
929:
930: Revision 1.101 2004/09/15 10:38:38 brouard
931: Fix on curr_time
932:
933: Revision 1.100 2004/07/12 18:29:06 brouard
934: Add version for Mac OS X. Just define UNIX in Makefile
935:
936: Revision 1.99 2004/06/05 08:57:40 brouard
937: *** empty log message ***
938:
939: Revision 1.98 2004/05/16 15:05:56 brouard
940: New version 0.97 . First attempt to estimate force of mortality
941: directly from the data i.e. without the need of knowing the health
942: state at each age, but using a Gompertz model: log u =a + b*age .
943: This is the basic analysis of mortality and should be done before any
944: other analysis, in order to test if the mortality estimated from the
945: cross-longitudinal survey is different from the mortality estimated
946: from other sources like vital statistic data.
947:
948: The same imach parameter file can be used but the option for mle should be -3.
949:
1.324 brouard 950: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 951: former routines in order to include the new code within the former code.
952:
953: The output is very simple: only an estimate of the intercept and of
954: the slope with 95% confident intervals.
955:
956: Current limitations:
957: A) Even if you enter covariates, i.e. with the
958: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
959: B) There is no computation of Life Expectancy nor Life Table.
960:
961: Revision 1.97 2004/02/20 13:25:42 lievre
962: Version 0.96d. Population forecasting command line is (temporarily)
963: suppressed.
964:
965: Revision 1.96 2003/07/15 15:38:55 brouard
966: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
967: rewritten within the same printf. Workaround: many printfs.
968:
969: Revision 1.95 2003/07/08 07:54:34 brouard
970: * imach.c (Repository):
971: (Repository): Using imachwizard code to output a more meaningful covariance
972: matrix (cov(a12,c31) instead of numbers.
973:
974: Revision 1.94 2003/06/27 13:00:02 brouard
975: Just cleaning
976:
977: Revision 1.93 2003/06/25 16:33:55 brouard
978: (Module): On windows (cygwin) function asctime_r doesn't
979: exist so I changed back to asctime which exists.
980: (Module): Version 0.96b
981:
982: Revision 1.92 2003/06/25 16:30:45 brouard
983: (Module): On windows (cygwin) function asctime_r doesn't
984: exist so I changed back to asctime which exists.
985:
986: Revision 1.91 2003/06/25 15:30:29 brouard
987: * imach.c (Repository): Duplicated warning errors corrected.
988: (Repository): Elapsed time after each iteration is now output. It
989: helps to forecast when convergence will be reached. Elapsed time
990: is stamped in powell. We created a new html file for the graphs
991: concerning matrix of covariance. It has extension -cov.htm.
992:
993: Revision 1.90 2003/06/24 12:34:15 brouard
994: (Module): Some bugs corrected for windows. Also, when
995: mle=-1 a template is output in file "or"mypar.txt with the design
996: of the covariance matrix to be input.
997:
998: Revision 1.89 2003/06/24 12:30:52 brouard
999: (Module): Some bugs corrected for windows. Also, when
1000: mle=-1 a template is output in file "or"mypar.txt with the design
1001: of the covariance matrix to be input.
1002:
1003: Revision 1.88 2003/06/23 17:54:56 brouard
1004: * 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.
1005:
1006: Revision 1.87 2003/06/18 12:26:01 brouard
1007: Version 0.96
1008:
1009: Revision 1.86 2003/06/17 20:04:08 brouard
1010: (Module): Change position of html and gnuplot routines and added
1011: routine fileappend.
1012:
1013: Revision 1.85 2003/06/17 13:12:43 brouard
1014: * imach.c (Repository): Check when date of death was earlier that
1015: current date of interview. It may happen when the death was just
1016: prior to the death. In this case, dh was negative and likelihood
1017: was wrong (infinity). We still send an "Error" but patch by
1018: assuming that the date of death was just one stepm after the
1019: interview.
1020: (Repository): Because some people have very long ID (first column)
1021: we changed int to long in num[] and we added a new lvector for
1022: memory allocation. But we also truncated to 8 characters (left
1023: truncation)
1024: (Repository): No more line truncation errors.
1025:
1026: Revision 1.84 2003/06/13 21:44:43 brouard
1027: * imach.c (Repository): Replace "freqsummary" at a correct
1028: place. It differs from routine "prevalence" which may be called
1029: many times. Probs is memory consuming and must be used with
1030: parcimony.
1031: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1032:
1033: Revision 1.83 2003/06/10 13:39:11 lievre
1034: *** empty log message ***
1035:
1036: Revision 1.82 2003/06/05 15:57:20 brouard
1037: Add log in imach.c and fullversion number is now printed.
1038:
1039: */
1040: /*
1041: Interpolated Markov Chain
1042:
1043: Short summary of the programme:
1044:
1.227 brouard 1045: This program computes Healthy Life Expectancies or State-specific
1046: (if states aren't health statuses) Expectancies from
1047: cross-longitudinal data. Cross-longitudinal data consist in:
1048:
1049: -1- a first survey ("cross") where individuals from different ages
1050: are interviewed on their health status or degree of disability (in
1051: the case of a health survey which is our main interest)
1052:
1053: -2- at least a second wave of interviews ("longitudinal") which
1054: measure each change (if any) in individual health status. Health
1055: expectancies are computed from the time spent in each health state
1056: according to a model. More health states you consider, more time is
1057: necessary to reach the Maximum Likelihood of the parameters involved
1058: in the model. The simplest model is the multinomial logistic model
1059: where pij is the probability to be observed in state j at the second
1060: wave conditional to be observed in state i at the first
1061: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1062: etc , where 'age' is age and 'sex' is a covariate. If you want to
1063: have a more complex model than "constant and age", you should modify
1064: the program where the markup *Covariates have to be included here
1065: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1066: convergence.
1067:
1068: The advantage of this computer programme, compared to a simple
1069: multinomial logistic model, is clear when the delay between waves is not
1070: identical for each individual. Also, if a individual missed an
1071: intermediate interview, the information is lost, but taken into
1072: account using an interpolation or extrapolation.
1073:
1074: hPijx is the probability to be observed in state i at age x+h
1075: conditional to the observed state i at age x. The delay 'h' can be
1076: split into an exact number (nh*stepm) of unobserved intermediate
1077: states. This elementary transition (by month, quarter,
1078: semester or year) is modelled as a multinomial logistic. The hPx
1079: matrix is simply the matrix product of nh*stepm elementary matrices
1080: and the contribution of each individual to the likelihood is simply
1081: hPijx.
1082:
1083: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1084: of the life expectancies. It also computes the period (stable) prevalence.
1085:
1086: Back prevalence and projections:
1.227 brouard 1087:
1088: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1089: double agemaxpar, double ftolpl, int *ncvyearp, double
1090: dateprev1,double dateprev2, int firstpass, int lastpass, int
1091: mobilavproj)
1092:
1093: Computes the back prevalence limit for any combination of
1094: covariate values k at any age between ageminpar and agemaxpar and
1095: returns it in **bprlim. In the loops,
1096:
1097: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1098: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1099:
1100: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1101: Computes for any combination of covariates k and any age between bage and fage
1102: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1103: oldm=oldms;savm=savms;
1.227 brouard 1104:
1.267 brouard 1105: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1106: Computes the transition matrix starting at age 'age' over
1107: 'nhstepm*hstepm*stepm' months (i.e. until
1108: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1109: nhstepm*hstepm matrices.
1110:
1111: Returns p3mat[i][j][h] after calling
1112: p3mat[i][j][h]=matprod2(newm,
1113: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1114: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1115: oldm);
1.226 brouard 1116:
1117: Important routines
1118:
1119: - func (or funcone), computes logit (pij) distinguishing
1120: o fixed variables (single or product dummies or quantitative);
1121: o varying variables by:
1122: (1) wave (single, product dummies, quantitative),
1123: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1124: % fixed dummy (treated) or quantitative (not done because time-consuming);
1125: % varying dummy (not done) or quantitative (not done);
1126: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1127: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1128: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1129: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1130: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1131:
1.226 brouard 1132:
1133:
1.324 brouard 1134: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1135: Institut national d'études démographiques, Paris.
1.126 brouard 1136: This software have been partly granted by Euro-REVES, a concerted action
1137: from the European Union.
1138: It is copyrighted identically to a GNU software product, ie programme and
1139: software can be distributed freely for non commercial use. Latest version
1140: can be accessed at http://euroreves.ined.fr/imach .
1141:
1142: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1143: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1144:
1145: **********************************************************************/
1146: /*
1147: main
1148: read parameterfile
1149: read datafile
1150: concatwav
1151: freqsummary
1152: if (mle >= 1)
1153: mlikeli
1154: print results files
1155: if mle==1
1156: computes hessian
1157: read end of parameter file: agemin, agemax, bage, fage, estepm
1158: begin-prev-date,...
1159: open gnuplot file
1160: open html file
1.145 brouard 1161: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1162: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1163: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1164: freexexit2 possible for memory heap.
1165:
1166: h Pij x | pij_nom ficrestpij
1167: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1168: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1169: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1170:
1171: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1172: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1173: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1174: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1175: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1176:
1.126 brouard 1177: forecasting if prevfcast==1 prevforecast call prevalence()
1178: health expectancies
1179: Variance-covariance of DFLE
1180: prevalence()
1181: movingaverage()
1182: varevsij()
1183: if popbased==1 varevsij(,popbased)
1184: total life expectancies
1185: Variance of period (stable) prevalence
1186: end
1187: */
1188:
1.187 brouard 1189: /* #define DEBUG */
1190: /* #define DEBUGBRENT */
1.203 brouard 1191: /* #define DEBUGLINMIN */
1192: /* #define DEBUGHESS */
1193: #define DEBUGHESSIJ
1.224 brouard 1194: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1195: #define POWELL /* Instead of NLOPT */
1.224 brouard 1196: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1197: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1198: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1199: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1200:
1201: #include <math.h>
1202: #include <stdio.h>
1203: #include <stdlib.h>
1204: #include <string.h>
1.226 brouard 1205: #include <ctype.h>
1.159 brouard 1206:
1207: #ifdef _WIN32
1208: #include <io.h>
1.172 brouard 1209: #include <windows.h>
1210: #include <tchar.h>
1.159 brouard 1211: #else
1.126 brouard 1212: #include <unistd.h>
1.159 brouard 1213: #endif
1.126 brouard 1214:
1215: #include <limits.h>
1216: #include <sys/types.h>
1.171 brouard 1217:
1218: #if defined(__GNUC__)
1219: #include <sys/utsname.h> /* Doesn't work on Windows */
1220: #endif
1221:
1.126 brouard 1222: #include <sys/stat.h>
1223: #include <errno.h>
1.159 brouard 1224: /* extern int errno; */
1.126 brouard 1225:
1.157 brouard 1226: /* #ifdef LINUX */
1227: /* #include <time.h> */
1228: /* #include "timeval.h" */
1229: /* #else */
1230: /* #include <sys/time.h> */
1231: /* #endif */
1232:
1.126 brouard 1233: #include <time.h>
1234:
1.136 brouard 1235: #ifdef GSL
1236: #include <gsl/gsl_errno.h>
1237: #include <gsl/gsl_multimin.h>
1238: #endif
1239:
1.167 brouard 1240:
1.162 brouard 1241: #ifdef NLOPT
1242: #include <nlopt.h>
1243: typedef struct {
1244: double (* function)(double [] );
1245: } myfunc_data ;
1246: #endif
1247:
1.126 brouard 1248: /* #include <libintl.h> */
1249: /* #define _(String) gettext (String) */
1250:
1.251 brouard 1251: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1252:
1253: #define GNUPLOTPROGRAM "gnuplot"
1254: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1255: #define FILENAMELENGTH 256
1.126 brouard 1256:
1257: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1258: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1259:
1.144 brouard 1260: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1261: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1262:
1263: #define NINTERVMAX 8
1.144 brouard 1264: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1265: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1266: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1267: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1268: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1269: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1270: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1271: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1272: /* #define AGESUP 130 */
1.288 brouard 1273: /* #define AGESUP 150 */
1274: #define AGESUP 200
1.268 brouard 1275: #define AGEINF 0
1.218 brouard 1276: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1277: #define AGEBASE 40
1.194 brouard 1278: #define AGEOVERFLOW 1.e20
1.164 brouard 1279: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1280: #ifdef _WIN32
1281: #define DIRSEPARATOR '\\'
1282: #define CHARSEPARATOR "\\"
1283: #define ODIRSEPARATOR '/'
1284: #else
1.126 brouard 1285: #define DIRSEPARATOR '/'
1286: #define CHARSEPARATOR "/"
1287: #define ODIRSEPARATOR '\\'
1288: #endif
1289:
1.337 ! brouard 1290: /* $Id: imach.c,v 1.336 2022/08/31 09:52:36 brouard Exp $ */
1.126 brouard 1291: /* $State: Exp $ */
1.196 brouard 1292: #include "version.h"
1293: char version[]=__IMACH_VERSION__;
1.337 ! brouard 1294: char copyright[]="September 2022,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2020, Nihon University 2021-202, INED 2000-2022";
! 1295: char fullversion[]="$Revision: 1.336 $ $Date: 2022/08/31 09:52:36 $";
1.126 brouard 1296: char strstart[80];
1297: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1298: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1299: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1300: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1301: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 brouard 1302: 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 1303: 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 1304: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1305: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1306: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1307: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 brouard 1308: 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 1309: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1310: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1311: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1312: int nsd=0; /**< Total number of single dummy variables (output) */
1313: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1314: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1315: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1316: int ntveff=0; /**< ntveff number of effective time varying variables */
1317: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1318: int cptcov=0; /* Working variable */
1.334 brouard 1319: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1320: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1321: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1322: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1323: int nlstate=2; /* Number of live states */
1324: int ndeath=1; /* Number of dead states */
1.130 brouard 1325: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1326: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1327: int popbased=0;
1328:
1329: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1330: int maxwav=0; /* Maxim number of waves */
1331: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1332: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1333: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1334: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1335: int mle=1, weightopt=0;
1.126 brouard 1336: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1337: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1338: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1339: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1340: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1341: int selected(int kvar); /* Is covariate kvar selected for printing results */
1342:
1.130 brouard 1343: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1344: double **matprod2(); /* test */
1.126 brouard 1345: double **oldm, **newm, **savm; /* Working pointers to matrices */
1346: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1347: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1348:
1.136 brouard 1349: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1350: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1351: FILE *ficlog, *ficrespow;
1.130 brouard 1352: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1353: double fretone; /* Only one call to likelihood */
1.130 brouard 1354: long ipmx=0; /* Number of contributions */
1.126 brouard 1355: double sw; /* Sum of weights */
1356: char filerespow[FILENAMELENGTH];
1357: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1358: FILE *ficresilk;
1359: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1360: FILE *ficresprobmorprev;
1361: FILE *fichtm, *fichtmcov; /* Html File */
1362: FILE *ficreseij;
1363: char filerese[FILENAMELENGTH];
1364: FILE *ficresstdeij;
1365: char fileresstde[FILENAMELENGTH];
1366: FILE *ficrescveij;
1367: char filerescve[FILENAMELENGTH];
1368: FILE *ficresvij;
1369: char fileresv[FILENAMELENGTH];
1.269 brouard 1370:
1.126 brouard 1371: char title[MAXLINE];
1.234 brouard 1372: char model[MAXLINE]; /**< The model line */
1.217 brouard 1373: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1374: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1375: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1376: char command[FILENAMELENGTH];
1377: int outcmd=0;
1378:
1.217 brouard 1379: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1380: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1381: char filelog[FILENAMELENGTH]; /* Log file */
1382: char filerest[FILENAMELENGTH];
1383: char fileregp[FILENAMELENGTH];
1384: char popfile[FILENAMELENGTH];
1385:
1386: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1387:
1.157 brouard 1388: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1389: /* struct timezone tzp; */
1390: /* extern int gettimeofday(); */
1391: struct tm tml, *gmtime(), *localtime();
1392:
1393: extern time_t time();
1394:
1395: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1396: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1397: struct tm tm;
1398:
1.126 brouard 1399: char strcurr[80], strfor[80];
1400:
1401: char *endptr;
1402: long lval;
1403: double dval;
1404:
1405: #define NR_END 1
1406: #define FREE_ARG char*
1407: #define FTOL 1.0e-10
1408:
1409: #define NRANSI
1.240 brouard 1410: #define ITMAX 200
1411: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1412:
1413: #define TOL 2.0e-4
1414:
1415: #define CGOLD 0.3819660
1416: #define ZEPS 1.0e-10
1417: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1418:
1419: #define GOLD 1.618034
1420: #define GLIMIT 100.0
1421: #define TINY 1.0e-20
1422:
1423: static double maxarg1,maxarg2;
1424: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1425: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1426:
1427: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1428: #define rint(a) floor(a+0.5)
1.166 brouard 1429: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1430: #define mytinydouble 1.0e-16
1.166 brouard 1431: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1432: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1433: /* static double dsqrarg; */
1434: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1435: static double sqrarg;
1436: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1437: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1438: int agegomp= AGEGOMP;
1439:
1440: int imx;
1441: int stepm=1;
1442: /* Stepm, step in month: minimum step interpolation*/
1443:
1444: int estepm;
1445: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1446:
1447: int m,nb;
1448: long *num;
1.197 brouard 1449: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1450: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1451: covariate for which somebody answered excluding
1452: undefined. Usually 2: 0 and 1. */
1453: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1454: covariate for which somebody answered including
1455: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1456: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1457: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1458: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1459: 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 1460: double *ageexmed,*agecens;
1461: double dateintmean=0;
1.296 brouard 1462: double anprojd, mprojd, jprojd; /* For eventual projections */
1463: double anprojf, mprojf, jprojf;
1.126 brouard 1464:
1.296 brouard 1465: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1466: double anbackf, mbackf, jbackf;
1467: double jintmean,mintmean,aintmean;
1.126 brouard 1468: double *weight;
1469: int **s; /* Status */
1.141 brouard 1470: double *agedc;
1.145 brouard 1471: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1472: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1473: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1474: double **coqvar; /* Fixed quantitative covariate nqv */
1475: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1476: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1477: double idx;
1478: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1479: /* Some documentation */
1480: /* Design original data
1481: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1482: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1483: * ntv=3 nqtv=1
1.330 brouard 1484: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1485: * For time varying covariate, quanti or dummies
1486: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1487: * cotvar[wav][ntv+iv][i]= [3+(1 to nqtv)][i]=(V12) quanti
1488: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1489: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1490: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1491: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1492: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1493: * k= 1 2 3 4 5 6 7 8 9 10 11
1494: */
1495: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1496: /* 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
1497: # States 1=Coresidence, 2 Living alone, 3 Institution
1498: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1499: */
1.319 brouard 1500: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1501: /* k 1 2 3 4 5 6 7 8 9 */
1502: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1503: /* fixed or varying), 1 for age product, 2 for*/
1504: /* product */
1505: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1506: /*(single or product without age), 2 dummy*/
1507: /* with age product, 3 quant with age product*/
1508: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1509: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 brouard 1510: /*TnsdVar[Tvar] 1 2 3 */
1.337 ! brouard 1511: /*Tvaraff[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.319 brouard 1512: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1513: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1514: /* nsq 1 2 */ /* Counting single quantit tv */
1515: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1516: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1517: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1518: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1519: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1520: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1521: /* 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 1522: /* 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 1523: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1524: /* Type */
1525: /* V 1 2 3 4 5 */
1526: /* F F V V V */
1527: /* D Q D D Q */
1528: /* */
1529: int *TvarsD;
1.330 brouard 1530: int *TnsdVar;
1.234 brouard 1531: int *TvarsDind;
1532: int *TvarsQ;
1533: int *TvarsQind;
1534:
1.318 brouard 1535: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1536: int nresult=0;
1.258 brouard 1537: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1538: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1539: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1540: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1541: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1542: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1543: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1544: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1545: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1546: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1547: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1548:
1549: /* 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
1550: # States 1=Coresidence, 2 Living alone, 3 Institution
1551: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1552: */
1.234 brouard 1553: /* 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 1554: 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 */
1555: 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 */
1556: 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 */
1557: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1558: 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 */
1559: 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 1560: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1561: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1562: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1563: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1564: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1565: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1566: 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 */
1567: 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 */
1568:
1.230 brouard 1569: int *Tvarsel; /**< Selected covariates for output */
1570: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1571: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1572: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1573: 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 1574: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1575: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1576: int *Tage;
1.227 brouard 1577: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1578: 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 1579: 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*/
1580: 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 1581: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1582: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1583: int **Tvard;
1.330 brouard 1584: int **Tvardk;
1.227 brouard 1585: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1586: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1587: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1588: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1589: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1590: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1591: double *lsurv, *lpop, *tpop;
1592:
1.231 brouard 1593: #define FD 1; /* Fixed dummy covariate */
1594: #define FQ 2; /* Fixed quantitative covariate */
1595: #define FP 3; /* Fixed product covariate */
1596: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1597: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1598: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1599: #define VD 10; /* Varying dummy covariate */
1600: #define VQ 11; /* Varying quantitative covariate */
1601: #define VP 12; /* Varying product covariate */
1602: #define VPDD 13; /* Varying product dummy*dummy covariate */
1603: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1604: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1605: #define APFD 16; /* Age product * fixed dummy covariate */
1606: #define APFQ 17; /* Age product * fixed quantitative covariate */
1607: #define APVD 18; /* Age product * varying dummy covariate */
1608: #define APVQ 19; /* Age product * varying quantitative covariate */
1609:
1610: #define FTYPE 1; /* Fixed covariate */
1611: #define VTYPE 2; /* Varying covariate (loop in wave) */
1612: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1613:
1614: struct kmodel{
1615: int maintype; /* main type */
1616: int subtype; /* subtype */
1617: };
1618: struct kmodel modell[NCOVMAX];
1619:
1.143 brouard 1620: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1621: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1622:
1623: /**************** split *************************/
1624: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1625: {
1626: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1627: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1628: */
1629: char *ss; /* pointer */
1.186 brouard 1630: int l1=0, l2=0; /* length counters */
1.126 brouard 1631:
1632: l1 = strlen(path ); /* length of path */
1633: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1634: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1635: if ( ss == NULL ) { /* no directory, so determine current directory */
1636: strcpy( name, path ); /* we got the fullname name because no directory */
1637: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1638: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1639: /* get current working directory */
1640: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1641: #ifdef WIN32
1642: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1643: #else
1644: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1645: #endif
1.126 brouard 1646: return( GLOCK_ERROR_GETCWD );
1647: }
1648: /* got dirc from getcwd*/
1649: printf(" DIRC = %s \n",dirc);
1.205 brouard 1650: } else { /* strip directory from path */
1.126 brouard 1651: ss++; /* after this, the filename */
1652: l2 = strlen( ss ); /* length of filename */
1653: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1654: strcpy( name, ss ); /* save file name */
1655: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1656: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1657: printf(" DIRC2 = %s \n",dirc);
1658: }
1659: /* We add a separator at the end of dirc if not exists */
1660: l1 = strlen( dirc ); /* length of directory */
1661: if( dirc[l1-1] != DIRSEPARATOR ){
1662: dirc[l1] = DIRSEPARATOR;
1663: dirc[l1+1] = 0;
1664: printf(" DIRC3 = %s \n",dirc);
1665: }
1666: ss = strrchr( name, '.' ); /* find last / */
1667: if (ss >0){
1668: ss++;
1669: strcpy(ext,ss); /* save extension */
1670: l1= strlen( name);
1671: l2= strlen(ss)+1;
1672: strncpy( finame, name, l1-l2);
1673: finame[l1-l2]= 0;
1674: }
1675:
1676: return( 0 ); /* we're done */
1677: }
1678:
1679:
1680: /******************************************/
1681:
1682: void replace_back_to_slash(char *s, char*t)
1683: {
1684: int i;
1685: int lg=0;
1686: i=0;
1687: lg=strlen(t);
1688: for(i=0; i<= lg; i++) {
1689: (s[i] = t[i]);
1690: if (t[i]== '\\') s[i]='/';
1691: }
1692: }
1693:
1.132 brouard 1694: char *trimbb(char *out, char *in)
1.137 brouard 1695: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1696: char *s;
1697: s=out;
1698: while (*in != '\0'){
1.137 brouard 1699: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1700: in++;
1701: }
1702: *out++ = *in++;
1703: }
1704: *out='\0';
1705: return s;
1706: }
1707:
1.187 brouard 1708: /* char *substrchaine(char *out, char *in, char *chain) */
1709: /* { */
1710: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1711: /* char *s, *t; */
1712: /* t=in;s=out; */
1713: /* while ((*in != *chain) && (*in != '\0')){ */
1714: /* *out++ = *in++; */
1715: /* } */
1716:
1717: /* /\* *in matches *chain *\/ */
1718: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1719: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1720: /* } */
1721: /* in--; chain--; */
1722: /* while ( (*in != '\0')){ */
1723: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1724: /* *out++ = *in++; */
1725: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1726: /* } */
1727: /* *out='\0'; */
1728: /* out=s; */
1729: /* return out; */
1730: /* } */
1731: char *substrchaine(char *out, char *in, char *chain)
1732: {
1733: /* Substract chain 'chain' from 'in', return and output 'out' */
1734: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1735:
1736: char *strloc;
1737:
1738: strcpy (out, in);
1739: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1740: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1741: if(strloc != NULL){
1742: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1743: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1744: /* strcpy (strloc, strloc +strlen(chain));*/
1745: }
1746: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1747: return out;
1748: }
1749:
1750:
1.145 brouard 1751: char *cutl(char *blocc, char *alocc, char *in, char occ)
1752: {
1.187 brouard 1753: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1754: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1755: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1756: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1757: */
1.160 brouard 1758: char *s, *t;
1.145 brouard 1759: t=in;s=in;
1760: while ((*in != occ) && (*in != '\0')){
1761: *alocc++ = *in++;
1762: }
1763: if( *in == occ){
1764: *(alocc)='\0';
1765: s=++in;
1766: }
1767:
1768: if (s == t) {/* occ not found */
1769: *(alocc-(in-s))='\0';
1770: in=s;
1771: }
1772: while ( *in != '\0'){
1773: *blocc++ = *in++;
1774: }
1775:
1776: *blocc='\0';
1777: return t;
1778: }
1.137 brouard 1779: char *cutv(char *blocc, char *alocc, char *in, char occ)
1780: {
1.187 brouard 1781: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1782: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1783: gives blocc="abcdef2ghi" and alocc="j".
1784: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1785: */
1786: char *s, *t;
1787: t=in;s=in;
1788: while (*in != '\0'){
1789: while( *in == occ){
1790: *blocc++ = *in++;
1791: s=in;
1792: }
1793: *blocc++ = *in++;
1794: }
1795: if (s == t) /* occ not found */
1796: *(blocc-(in-s))='\0';
1797: else
1798: *(blocc-(in-s)-1)='\0';
1799: in=s;
1800: while ( *in != '\0'){
1801: *alocc++ = *in++;
1802: }
1803:
1804: *alocc='\0';
1805: return s;
1806: }
1807:
1.126 brouard 1808: int nbocc(char *s, char occ)
1809: {
1810: int i,j=0;
1811: int lg=20;
1812: i=0;
1813: lg=strlen(s);
1814: for(i=0; i<= lg; i++) {
1.234 brouard 1815: if (s[i] == occ ) j++;
1.126 brouard 1816: }
1817: return j;
1818: }
1819:
1.137 brouard 1820: /* void cutv(char *u,char *v, char*t, char occ) */
1821: /* { */
1822: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1823: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1824: /* gives u="abcdef2ghi" and v="j" *\/ */
1825: /* int i,lg,j,p=0; */
1826: /* i=0; */
1827: /* lg=strlen(t); */
1828: /* for(j=0; j<=lg-1; j++) { */
1829: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1830: /* } */
1.126 brouard 1831:
1.137 brouard 1832: /* for(j=0; j<p; j++) { */
1833: /* (u[j] = t[j]); */
1834: /* } */
1835: /* u[p]='\0'; */
1.126 brouard 1836:
1.137 brouard 1837: /* for(j=0; j<= lg; j++) { */
1838: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1839: /* } */
1840: /* } */
1.126 brouard 1841:
1.160 brouard 1842: #ifdef _WIN32
1843: char * strsep(char **pp, const char *delim)
1844: {
1845: char *p, *q;
1846:
1847: if ((p = *pp) == NULL)
1848: return 0;
1849: if ((q = strpbrk (p, delim)) != NULL)
1850: {
1851: *pp = q + 1;
1852: *q = '\0';
1853: }
1854: else
1855: *pp = 0;
1856: return p;
1857: }
1858: #endif
1859:
1.126 brouard 1860: /********************** nrerror ********************/
1861:
1862: void nrerror(char error_text[])
1863: {
1864: fprintf(stderr,"ERREUR ...\n");
1865: fprintf(stderr,"%s\n",error_text);
1866: exit(EXIT_FAILURE);
1867: }
1868: /*********************** vector *******************/
1869: double *vector(int nl, int nh)
1870: {
1871: double *v;
1872: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1873: if (!v) nrerror("allocation failure in vector");
1874: return v-nl+NR_END;
1875: }
1876:
1877: /************************ free vector ******************/
1878: void free_vector(double*v, int nl, int nh)
1879: {
1880: free((FREE_ARG)(v+nl-NR_END));
1881: }
1882:
1883: /************************ivector *******************************/
1884: int *ivector(long nl,long nh)
1885: {
1886: int *v;
1887: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1888: if (!v) nrerror("allocation failure in ivector");
1889: return v-nl+NR_END;
1890: }
1891:
1892: /******************free ivector **************************/
1893: void free_ivector(int *v, long nl, long nh)
1894: {
1895: free((FREE_ARG)(v+nl-NR_END));
1896: }
1897:
1898: /************************lvector *******************************/
1899: long *lvector(long nl,long nh)
1900: {
1901: long *v;
1902: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1903: if (!v) nrerror("allocation failure in ivector");
1904: return v-nl+NR_END;
1905: }
1906:
1907: /******************free lvector **************************/
1908: void free_lvector(long *v, long nl, long nh)
1909: {
1910: free((FREE_ARG)(v+nl-NR_END));
1911: }
1912:
1913: /******************* imatrix *******************************/
1914: int **imatrix(long nrl, long nrh, long ncl, long nch)
1915: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1916: {
1917: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1918: int **m;
1919:
1920: /* allocate pointers to rows */
1921: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1922: if (!m) nrerror("allocation failure 1 in matrix()");
1923: m += NR_END;
1924: m -= nrl;
1925:
1926:
1927: /* allocate rows and set pointers to them */
1928: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1929: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1930: m[nrl] += NR_END;
1931: m[nrl] -= ncl;
1932:
1933: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1934:
1935: /* return pointer to array of pointers to rows */
1936: return m;
1937: }
1938:
1939: /****************** free_imatrix *************************/
1940: void free_imatrix(m,nrl,nrh,ncl,nch)
1941: int **m;
1942: long nch,ncl,nrh,nrl;
1943: /* free an int matrix allocated by imatrix() */
1944: {
1945: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1946: free((FREE_ARG) (m+nrl-NR_END));
1947: }
1948:
1949: /******************* matrix *******************************/
1950: double **matrix(long nrl, long nrh, long ncl, long nch)
1951: {
1952: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1953: double **m;
1954:
1955: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1956: if (!m) nrerror("allocation failure 1 in matrix()");
1957: m += NR_END;
1958: m -= nrl;
1959:
1960: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1961: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1962: m[nrl] += NR_END;
1963: m[nrl] -= ncl;
1964:
1965: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1966: return m;
1.145 brouard 1967: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1968: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1969: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1970: */
1971: }
1972:
1973: /*************************free matrix ************************/
1974: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1975: {
1976: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1977: free((FREE_ARG)(m+nrl-NR_END));
1978: }
1979:
1980: /******************* ma3x *******************************/
1981: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1982: {
1983: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1984: double ***m;
1985:
1986: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1987: if (!m) nrerror("allocation failure 1 in matrix()");
1988: m += NR_END;
1989: m -= nrl;
1990:
1991: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1992: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1993: m[nrl] += NR_END;
1994: m[nrl] -= ncl;
1995:
1996: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1997:
1998: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1999: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
2000: m[nrl][ncl] += NR_END;
2001: m[nrl][ncl] -= nll;
2002: for (j=ncl+1; j<=nch; j++)
2003: m[nrl][j]=m[nrl][j-1]+nlay;
2004:
2005: for (i=nrl+1; i<=nrh; i++) {
2006: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2007: for (j=ncl+1; j<=nch; j++)
2008: m[i][j]=m[i][j-1]+nlay;
2009: }
2010: return m;
2011: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2012: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2013: */
2014: }
2015:
2016: /*************************free ma3x ************************/
2017: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2018: {
2019: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2020: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2021: free((FREE_ARG)(m+nrl-NR_END));
2022: }
2023:
2024: /*************** function subdirf ***********/
2025: char *subdirf(char fileres[])
2026: {
2027: /* Caution optionfilefiname is hidden */
2028: strcpy(tmpout,optionfilefiname);
2029: strcat(tmpout,"/"); /* Add to the right */
2030: strcat(tmpout,fileres);
2031: return tmpout;
2032: }
2033:
2034: /*************** function subdirf2 ***********/
2035: char *subdirf2(char fileres[], char *preop)
2036: {
1.314 brouard 2037: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2038: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2039: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2040: /* Caution optionfilefiname is hidden */
2041: strcpy(tmpout,optionfilefiname);
2042: strcat(tmpout,"/");
2043: strcat(tmpout,preop);
2044: strcat(tmpout,fileres);
2045: return tmpout;
2046: }
2047:
2048: /*************** function subdirf3 ***********/
2049: char *subdirf3(char fileres[], char *preop, char *preop2)
2050: {
2051:
2052: /* Caution optionfilefiname is hidden */
2053: strcpy(tmpout,optionfilefiname);
2054: strcat(tmpout,"/");
2055: strcat(tmpout,preop);
2056: strcat(tmpout,preop2);
2057: strcat(tmpout,fileres);
2058: return tmpout;
2059: }
1.213 brouard 2060:
2061: /*************** function subdirfext ***********/
2062: char *subdirfext(char fileres[], char *preop, char *postop)
2063: {
2064:
2065: strcpy(tmpout,preop);
2066: strcat(tmpout,fileres);
2067: strcat(tmpout,postop);
2068: return tmpout;
2069: }
1.126 brouard 2070:
1.213 brouard 2071: /*************** function subdirfext3 ***********/
2072: char *subdirfext3(char fileres[], char *preop, char *postop)
2073: {
2074:
2075: /* Caution optionfilefiname is hidden */
2076: strcpy(tmpout,optionfilefiname);
2077: strcat(tmpout,"/");
2078: strcat(tmpout,preop);
2079: strcat(tmpout,fileres);
2080: strcat(tmpout,postop);
2081: return tmpout;
2082: }
2083:
1.162 brouard 2084: char *asc_diff_time(long time_sec, char ascdiff[])
2085: {
2086: long sec_left, days, hours, minutes;
2087: days = (time_sec) / (60*60*24);
2088: sec_left = (time_sec) % (60*60*24);
2089: hours = (sec_left) / (60*60) ;
2090: sec_left = (sec_left) %(60*60);
2091: minutes = (sec_left) /60;
2092: sec_left = (sec_left) % (60);
2093: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2094: return ascdiff;
2095: }
2096:
1.126 brouard 2097: /***************** f1dim *************************/
2098: extern int ncom;
2099: extern double *pcom,*xicom;
2100: extern double (*nrfunc)(double []);
2101:
2102: double f1dim(double x)
2103: {
2104: int j;
2105: double f;
2106: double *xt;
2107:
2108: xt=vector(1,ncom);
2109: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2110: f=(*nrfunc)(xt);
2111: free_vector(xt,1,ncom);
2112: return f;
2113: }
2114:
2115: /*****************brent *************************/
2116: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2117: {
2118: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2119: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2120: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2121: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2122: * returned function value.
2123: */
1.126 brouard 2124: int iter;
2125: double a,b,d,etemp;
1.159 brouard 2126: double fu=0,fv,fw,fx;
1.164 brouard 2127: double ftemp=0.;
1.126 brouard 2128: double p,q,r,tol1,tol2,u,v,w,x,xm;
2129: double e=0.0;
2130:
2131: a=(ax < cx ? ax : cx);
2132: b=(ax > cx ? ax : cx);
2133: x=w=v=bx;
2134: fw=fv=fx=(*f)(x);
2135: for (iter=1;iter<=ITMAX;iter++) {
2136: xm=0.5*(a+b);
2137: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2138: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2139: printf(".");fflush(stdout);
2140: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2141: #ifdef DEBUGBRENT
1.126 brouard 2142: 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);
2143: 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);
2144: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2145: #endif
2146: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2147: *xmin=x;
2148: return fx;
2149: }
2150: ftemp=fu;
2151: if (fabs(e) > tol1) {
2152: r=(x-w)*(fx-fv);
2153: q=(x-v)*(fx-fw);
2154: p=(x-v)*q-(x-w)*r;
2155: q=2.0*(q-r);
2156: if (q > 0.0) p = -p;
2157: q=fabs(q);
2158: etemp=e;
2159: e=d;
2160: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2161: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2162: else {
1.224 brouard 2163: d=p/q;
2164: u=x+d;
2165: if (u-a < tol2 || b-u < tol2)
2166: d=SIGN(tol1,xm-x);
1.126 brouard 2167: }
2168: } else {
2169: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2170: }
2171: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2172: fu=(*f)(u);
2173: if (fu <= fx) {
2174: if (u >= x) a=x; else b=x;
2175: SHFT(v,w,x,u)
1.183 brouard 2176: SHFT(fv,fw,fx,fu)
2177: } else {
2178: if (u < x) a=u; else b=u;
2179: if (fu <= fw || w == x) {
1.224 brouard 2180: v=w;
2181: w=u;
2182: fv=fw;
2183: fw=fu;
1.183 brouard 2184: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2185: v=u;
2186: fv=fu;
1.183 brouard 2187: }
2188: }
1.126 brouard 2189: }
2190: nrerror("Too many iterations in brent");
2191: *xmin=x;
2192: return fx;
2193: }
2194:
2195: /****************** mnbrak ***********************/
2196:
2197: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2198: double (*func)(double))
1.183 brouard 2199: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2200: the downhill direction (defined by the function as evaluated at the initial points) and returns
2201: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2202: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2203: */
1.126 brouard 2204: double ulim,u,r,q, dum;
2205: double fu;
1.187 brouard 2206:
2207: double scale=10.;
2208: int iterscale=0;
2209:
2210: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2211: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2212:
2213:
2214: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2215: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2216: /* *bx = *ax - (*ax - *bx)/scale; */
2217: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2218: /* } */
2219:
1.126 brouard 2220: if (*fb > *fa) {
2221: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2222: SHFT(dum,*fb,*fa,dum)
2223: }
1.126 brouard 2224: *cx=(*bx)+GOLD*(*bx-*ax);
2225: *fc=(*func)(*cx);
1.183 brouard 2226: #ifdef DEBUG
1.224 brouard 2227: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2228: 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 2229: #endif
1.224 brouard 2230: 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 2231: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2232: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2233: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2234: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2235: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2236: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2237: fu=(*func)(u);
1.163 brouard 2238: #ifdef DEBUG
2239: /* f(x)=A(x-u)**2+f(u) */
2240: double A, fparabu;
2241: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2242: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2243: 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);
2244: 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 2245: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2246: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2247: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2248: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2249: #endif
1.184 brouard 2250: #ifdef MNBRAKORIGINAL
1.183 brouard 2251: #else
1.191 brouard 2252: /* if (fu > *fc) { */
2253: /* #ifdef DEBUG */
2254: /* printf("mnbrak4 fu > fc \n"); */
2255: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2256: /* #endif */
2257: /* /\* 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 *\\/ *\/ */
2258: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2259: /* dum=u; /\* Shifting c and u *\/ */
2260: /* u = *cx; */
2261: /* *cx = dum; */
2262: /* dum = fu; */
2263: /* fu = *fc; */
2264: /* *fc =dum; */
2265: /* } else { /\* end *\/ */
2266: /* #ifdef DEBUG */
2267: /* printf("mnbrak3 fu < fc \n"); */
2268: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2269: /* #endif */
2270: /* dum=u; /\* Shifting c and u *\/ */
2271: /* u = *cx; */
2272: /* *cx = dum; */
2273: /* dum = fu; */
2274: /* fu = *fc; */
2275: /* *fc =dum; */
2276: /* } */
1.224 brouard 2277: #ifdef DEBUGMNBRAK
2278: double A, fparabu;
2279: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2280: fparabu= *fa - A*(*ax-u)*(*ax-u);
2281: 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);
2282: 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 2283: #endif
1.191 brouard 2284: dum=u; /* Shifting c and u */
2285: u = *cx;
2286: *cx = dum;
2287: dum = fu;
2288: fu = *fc;
2289: *fc =dum;
1.183 brouard 2290: #endif
1.162 brouard 2291: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2292: #ifdef DEBUG
1.224 brouard 2293: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2294: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2295: #endif
1.126 brouard 2296: fu=(*func)(u);
2297: if (fu < *fc) {
1.183 brouard 2298: #ifdef DEBUG
1.224 brouard 2299: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2300: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2301: #endif
2302: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2303: SHFT(*fb,*fc,fu,(*func)(u))
2304: #ifdef DEBUG
2305: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2306: #endif
2307: }
1.162 brouard 2308: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2309: #ifdef DEBUG
1.224 brouard 2310: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2311: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2312: #endif
1.126 brouard 2313: u=ulim;
2314: fu=(*func)(u);
1.183 brouard 2315: } else { /* u could be left to b (if r > q parabola has a maximum) */
2316: #ifdef DEBUG
1.224 brouard 2317: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2318: 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 2319: #endif
1.126 brouard 2320: u=(*cx)+GOLD*(*cx-*bx);
2321: fu=(*func)(u);
1.224 brouard 2322: #ifdef DEBUG
2323: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2324: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2325: #endif
1.183 brouard 2326: } /* end tests */
1.126 brouard 2327: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2328: SHFT(*fa,*fb,*fc,fu)
2329: #ifdef DEBUG
1.224 brouard 2330: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2331: 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 2332: #endif
2333: } /* 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 2334: }
2335:
2336: /*************** linmin ************************/
1.162 brouard 2337: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2338: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2339: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2340: the value of func at the returned location p . This is actually all accomplished by calling the
2341: routines mnbrak and brent .*/
1.126 brouard 2342: int ncom;
2343: double *pcom,*xicom;
2344: double (*nrfunc)(double []);
2345:
1.224 brouard 2346: #ifdef LINMINORIGINAL
1.126 brouard 2347: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2348: #else
2349: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2350: #endif
1.126 brouard 2351: {
2352: double brent(double ax, double bx, double cx,
2353: double (*f)(double), double tol, double *xmin);
2354: double f1dim(double x);
2355: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2356: double *fc, double (*func)(double));
2357: int j;
2358: double xx,xmin,bx,ax;
2359: double fx,fb,fa;
1.187 brouard 2360:
1.203 brouard 2361: #ifdef LINMINORIGINAL
2362: #else
2363: double scale=10., axs, xxs; /* Scale added for infinity */
2364: #endif
2365:
1.126 brouard 2366: ncom=n;
2367: pcom=vector(1,n);
2368: xicom=vector(1,n);
2369: nrfunc=func;
2370: for (j=1;j<=n;j++) {
2371: pcom[j]=p[j];
1.202 brouard 2372: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2373: }
1.187 brouard 2374:
1.203 brouard 2375: #ifdef LINMINORIGINAL
2376: xx=1.;
2377: #else
2378: axs=0.0;
2379: xxs=1.;
2380: do{
2381: xx= xxs;
2382: #endif
1.187 brouard 2383: ax=0.;
2384: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2385: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2386: /* 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)) */
2387: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2388: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2389: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2390: /* 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 2391: #ifdef LINMINORIGINAL
2392: #else
2393: if (fx != fx){
1.224 brouard 2394: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2395: printf("|");
2396: fprintf(ficlog,"|");
1.203 brouard 2397: #ifdef DEBUGLINMIN
1.224 brouard 2398: 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 2399: #endif
2400: }
1.224 brouard 2401: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2402: #endif
2403:
1.191 brouard 2404: #ifdef DEBUGLINMIN
2405: 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 2406: 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 2407: #endif
1.224 brouard 2408: #ifdef LINMINORIGINAL
2409: #else
1.317 brouard 2410: if(fb == fx){ /* Flat function in the direction */
2411: xmin=xx;
1.224 brouard 2412: *flat=1;
1.317 brouard 2413: }else{
1.224 brouard 2414: *flat=0;
2415: #endif
2416: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2417: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2418: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2419: /* fmin = f(p[j] + xmin * xi[j]) */
2420: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2421: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2422: #ifdef DEBUG
1.224 brouard 2423: 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);
2424: 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);
2425: #endif
2426: #ifdef LINMINORIGINAL
2427: #else
2428: }
1.126 brouard 2429: #endif
1.191 brouard 2430: #ifdef DEBUGLINMIN
2431: printf("linmin end ");
1.202 brouard 2432: fprintf(ficlog,"linmin end ");
1.191 brouard 2433: #endif
1.126 brouard 2434: for (j=1;j<=n;j++) {
1.203 brouard 2435: #ifdef LINMINORIGINAL
2436: xi[j] *= xmin;
2437: #else
2438: #ifdef DEBUGLINMIN
2439: if(xxs <1.0)
2440: printf(" before xi[%d]=%12.8f", j,xi[j]);
2441: #endif
2442: 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) */
2443: #ifdef DEBUGLINMIN
2444: if(xxs <1.0)
2445: 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 );
2446: #endif
2447: #endif
1.187 brouard 2448: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2449: }
1.191 brouard 2450: #ifdef DEBUGLINMIN
1.203 brouard 2451: printf("\n");
1.191 brouard 2452: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2453: 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 2454: for (j=1;j<=n;j++) {
1.202 brouard 2455: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2456: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2457: if(j % ncovmodel == 0){
1.191 brouard 2458: printf("\n");
1.202 brouard 2459: fprintf(ficlog,"\n");
2460: }
1.191 brouard 2461: }
1.203 brouard 2462: #else
1.191 brouard 2463: #endif
1.126 brouard 2464: free_vector(xicom,1,n);
2465: free_vector(pcom,1,n);
2466: }
2467:
2468:
2469: /*************** powell ************************/
1.162 brouard 2470: /*
1.317 brouard 2471: Minimization of a function func of n variables. Input consists in an initial starting point
2472: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2473: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2474: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2475: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2476: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2477: */
1.224 brouard 2478: #ifdef LINMINORIGINAL
2479: #else
2480: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2481: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2482: #endif
1.126 brouard 2483: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2484: double (*func)(double []))
2485: {
1.224 brouard 2486: #ifdef LINMINORIGINAL
2487: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2488: double (*func)(double []));
1.224 brouard 2489: #else
1.241 brouard 2490: void linmin(double p[], double xi[], int n, double *fret,
2491: double (*func)(double []),int *flat);
1.224 brouard 2492: #endif
1.239 brouard 2493: int i,ibig,j,jk,k;
1.126 brouard 2494: double del,t,*pt,*ptt,*xit;
1.181 brouard 2495: double directest;
1.126 brouard 2496: double fp,fptt;
2497: double *xits;
2498: int niterf, itmp;
2499:
2500: pt=vector(1,n);
2501: ptt=vector(1,n);
2502: xit=vector(1,n);
2503: xits=vector(1,n);
2504: *fret=(*func)(p);
2505: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2506: rcurr_time = time(NULL);
1.126 brouard 2507: for (*iter=1;;++(*iter)) {
2508: ibig=0;
2509: del=0.0;
1.157 brouard 2510: rlast_time=rcurr_time;
2511: /* (void) gettimeofday(&curr_time,&tzp); */
2512: rcurr_time = time(NULL);
2513: curr_time = *localtime(&rcurr_time);
1.337 ! brouard 2514: /* 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); */
! 2515: /* 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); */
! 2516: printf("\nPowell iter=%d -2*LL=%.12f gain=%.3lg %ld sec. %ld sec.",*iter,*fret,fp-*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
! 2517: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f gain=%.3lg %ld sec. %ld sec.",*iter,*fret,fp-*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
1.157 brouard 2518: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2519: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2520: for (i=1;i<=n;i++) {
1.126 brouard 2521: fprintf(ficrespow," %.12lf", p[i]);
2522: }
1.239 brouard 2523: fprintf(ficrespow,"\n");fflush(ficrespow);
2524: printf("\n#model= 1 + age ");
2525: fprintf(ficlog,"\n#model= 1 + age ");
2526: if(nagesqr==1){
1.241 brouard 2527: printf(" + age*age ");
2528: fprintf(ficlog," + age*age ");
1.239 brouard 2529: }
2530: for(j=1;j <=ncovmodel-2;j++){
2531: if(Typevar[j]==0) {
2532: printf(" + V%d ",Tvar[j]);
2533: fprintf(ficlog," + V%d ",Tvar[j]);
2534: }else if(Typevar[j]==1) {
2535: printf(" + V%d*age ",Tvar[j]);
2536: fprintf(ficlog," + V%d*age ",Tvar[j]);
2537: }else if(Typevar[j]==2) {
2538: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2539: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2540: }
2541: }
1.126 brouard 2542: printf("\n");
1.239 brouard 2543: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2544: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2545: fprintf(ficlog,"\n");
1.239 brouard 2546: for(i=1,jk=1; i <=nlstate; i++){
2547: for(k=1; k <=(nlstate+ndeath); k++){
2548: if (k != i) {
2549: printf("%d%d ",i,k);
2550: fprintf(ficlog,"%d%d ",i,k);
2551: for(j=1; j <=ncovmodel; j++){
2552: printf("%12.7f ",p[jk]);
2553: fprintf(ficlog,"%12.7f ",p[jk]);
2554: jk++;
2555: }
2556: printf("\n");
2557: fprintf(ficlog,"\n");
2558: }
2559: }
2560: }
1.241 brouard 2561: if(*iter <=3 && *iter >1){
1.157 brouard 2562: tml = *localtime(&rcurr_time);
2563: strcpy(strcurr,asctime(&tml));
2564: rforecast_time=rcurr_time;
1.126 brouard 2565: itmp = strlen(strcurr);
2566: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2567: strcurr[itmp-1]='\0';
1.162 brouard 2568: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2569: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2570: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2571: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2572: forecast_time = *localtime(&rforecast_time);
2573: strcpy(strfor,asctime(&forecast_time));
2574: itmp = strlen(strfor);
2575: if(strfor[itmp-1]=='\n')
2576: strfor[itmp-1]='\0';
2577: 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);
2578: 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 2579: }
2580: }
1.187 brouard 2581: for (i=1;i<=n;i++) { /* For each direction i */
2582: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2583: fptt=(*fret);
2584: #ifdef DEBUG
1.203 brouard 2585: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2586: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2587: #endif
1.203 brouard 2588: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2589: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2590: #ifdef LINMINORIGINAL
1.188 brouard 2591: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2592: #else
2593: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2594: flatdir[i]=flat; /* Function is vanishing in that direction i */
2595: #endif
2596: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2597: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2598: /* because that direction will be replaced unless the gain del is small */
2599: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2600: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2601: /* with the new direction. */
2602: del=fabs(fptt-(*fret));
2603: ibig=i;
1.126 brouard 2604: }
2605: #ifdef DEBUG
2606: printf("%d %.12e",i,(*fret));
2607: fprintf(ficlog,"%d %.12e",i,(*fret));
2608: for (j=1;j<=n;j++) {
1.224 brouard 2609: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2610: printf(" x(%d)=%.12e",j,xit[j]);
2611: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2612: }
2613: for(j=1;j<=n;j++) {
1.225 brouard 2614: printf(" p(%d)=%.12e",j,p[j]);
2615: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2616: }
2617: printf("\n");
2618: fprintf(ficlog,"\n");
2619: #endif
1.187 brouard 2620: } /* end loop on each direction i */
2621: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2622: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2623: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2624: for(j=1;j<=n;j++) {
2625: if(flatdir[j] >0){
2626: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2627: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2628: }
1.319 brouard 2629: /* printf("\n"); */
2630: /* fprintf(ficlog,"\n"); */
2631: }
1.243 brouard 2632: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2633: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2634: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2635: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2636: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2637: /* decreased of more than 3.84 */
2638: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2639: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2640: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2641:
1.188 brouard 2642: /* Starting the program with initial values given by a former maximization will simply change */
2643: /* the scales of the directions and the directions, because the are reset to canonical directions */
2644: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2645: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2646: #ifdef DEBUG
2647: int k[2],l;
2648: k[0]=1;
2649: k[1]=-1;
2650: printf("Max: %.12e",(*func)(p));
2651: fprintf(ficlog,"Max: %.12e",(*func)(p));
2652: for (j=1;j<=n;j++) {
2653: printf(" %.12e",p[j]);
2654: fprintf(ficlog," %.12e",p[j]);
2655: }
2656: printf("\n");
2657: fprintf(ficlog,"\n");
2658: for(l=0;l<=1;l++) {
2659: for (j=1;j<=n;j++) {
2660: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2661: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2662: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2663: }
2664: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2665: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2666: }
2667: #endif
2668:
2669: free_vector(xit,1,n);
2670: free_vector(xits,1,n);
2671: free_vector(ptt,1,n);
2672: free_vector(pt,1,n);
2673: return;
1.192 brouard 2674: } /* enough precision */
1.240 brouard 2675: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2676: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2677: ptt[j]=2.0*p[j]-pt[j];
2678: xit[j]=p[j]-pt[j];
2679: pt[j]=p[j];
2680: }
1.181 brouard 2681: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2682: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2683: if (*iter <=4) {
1.225 brouard 2684: #else
2685: #endif
1.224 brouard 2686: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2687: #else
1.161 brouard 2688: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2689: #endif
1.162 brouard 2690: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2691: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2692: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2693: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2694: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2695: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2696: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2697: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2698: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2699: /* Even if f3 <f1, directest can be negative and t >0 */
2700: /* mu² and del² are equal when f3=f1 */
2701: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2702: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2703: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2704: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2705: #ifdef NRCORIGINAL
2706: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2707: #else
2708: 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 2709: t= t- del*SQR(fp-fptt);
1.183 brouard 2710: #endif
1.202 brouard 2711: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2712: #ifdef DEBUG
1.181 brouard 2713: 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);
2714: 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 2715: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2716: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2717: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2718: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2719: 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);
2720: 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);
2721: #endif
1.183 brouard 2722: #ifdef POWELLORIGINAL
2723: if (t < 0.0) { /* Then we use it for new direction */
2724: #else
1.182 brouard 2725: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2726: 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 2727: 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 2728: 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 2729: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2730: }
1.181 brouard 2731: if (directest < 0.0) { /* Then we use it for new direction */
2732: #endif
1.191 brouard 2733: #ifdef DEBUGLINMIN
1.234 brouard 2734: printf("Before linmin in direction P%d-P0\n",n);
2735: for (j=1;j<=n;j++) {
2736: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2737: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2738: if(j % ncovmodel == 0){
2739: printf("\n");
2740: fprintf(ficlog,"\n");
2741: }
2742: }
1.224 brouard 2743: #endif
2744: #ifdef LINMINORIGINAL
1.234 brouard 2745: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2746: #else
1.234 brouard 2747: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2748: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2749: #endif
1.234 brouard 2750:
1.191 brouard 2751: #ifdef DEBUGLINMIN
1.234 brouard 2752: for (j=1;j<=n;j++) {
2753: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2754: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2755: if(j % ncovmodel == 0){
2756: printf("\n");
2757: fprintf(ficlog,"\n");
2758: }
2759: }
1.224 brouard 2760: #endif
1.234 brouard 2761: for (j=1;j<=n;j++) {
2762: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2763: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2764: }
1.224 brouard 2765: #ifdef LINMINORIGINAL
2766: #else
1.234 brouard 2767: for (j=1, flatd=0;j<=n;j++) {
2768: if(flatdir[j]>0)
2769: flatd++;
2770: }
2771: if(flatd >0){
1.255 brouard 2772: printf("%d flat directions: ",flatd);
2773: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2774: for (j=1;j<=n;j++) {
2775: if(flatdir[j]>0){
2776: printf("%d ",j);
2777: fprintf(ficlog,"%d ",j);
2778: }
2779: }
2780: printf("\n");
2781: fprintf(ficlog,"\n");
1.319 brouard 2782: #ifdef FLATSUP
2783: free_vector(xit,1,n);
2784: free_vector(xits,1,n);
2785: free_vector(ptt,1,n);
2786: free_vector(pt,1,n);
2787: return;
2788: #endif
1.234 brouard 2789: }
1.191 brouard 2790: #endif
1.234 brouard 2791: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2792: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2793:
1.126 brouard 2794: #ifdef DEBUG
1.234 brouard 2795: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2796: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2797: for(j=1;j<=n;j++){
2798: printf(" %lf",xit[j]);
2799: fprintf(ficlog," %lf",xit[j]);
2800: }
2801: printf("\n");
2802: fprintf(ficlog,"\n");
1.126 brouard 2803: #endif
1.192 brouard 2804: } /* end of t or directest negative */
1.224 brouard 2805: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2806: #else
1.234 brouard 2807: } /* end if (fptt < fp) */
1.192 brouard 2808: #endif
1.225 brouard 2809: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2810: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2811: #else
1.224 brouard 2812: #endif
1.234 brouard 2813: } /* loop iteration */
1.126 brouard 2814: }
1.234 brouard 2815:
1.126 brouard 2816: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2817:
1.235 brouard 2818: 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 2819: {
1.279 brouard 2820: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2821: * (and selected quantitative values in nres)
2822: * by left multiplying the unit
2823: * matrix by transitions matrix until convergence is reached with precision ftolpl
2824: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2825: * Wx is row vector: population in state 1, population in state 2, population dead
2826: * or prevalence in state 1, prevalence in state 2, 0
2827: * newm is the matrix after multiplications, its rows are identical at a factor.
2828: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2829: * Output is prlim.
2830: * Initial matrix pimij
2831: */
1.206 brouard 2832: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2833: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2834: /* 0, 0 , 1} */
2835: /*
2836: * and after some iteration: */
2837: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2838: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2839: /* 0, 0 , 1} */
2840: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2841: /* {0.51571254859325999, 0.4842874514067399, */
2842: /* 0.51326036147820708, 0.48673963852179264} */
2843: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2844:
1.332 brouard 2845: int i, ii,j,k, k1;
1.209 brouard 2846: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2847: /* double **matprod2(); */ /* test */
1.218 brouard 2848: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2849: double **newm;
1.209 brouard 2850: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2851: int ncvloop=0;
1.288 brouard 2852: int first=0;
1.169 brouard 2853:
1.209 brouard 2854: min=vector(1,nlstate);
2855: max=vector(1,nlstate);
2856: meandiff=vector(1,nlstate);
2857:
1.218 brouard 2858: /* Starting with matrix unity */
1.126 brouard 2859: for (ii=1;ii<=nlstate+ndeath;ii++)
2860: for (j=1;j<=nlstate+ndeath;j++){
2861: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2862: }
1.169 brouard 2863:
2864: cov[1]=1.;
2865:
2866: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2867: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2868: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2869: ncvloop++;
1.126 brouard 2870: newm=savm;
2871: /* Covariates have to be included here again */
1.138 brouard 2872: cov[2]=agefin;
1.319 brouard 2873: if(nagesqr==1){
2874: cov[3]= agefin*agefin;
2875: }
1.332 brouard 2876: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
2877: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
2878: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
2879: if(Typevar[k1]==1){ /* A product with age */
2880: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
2881: }else{
2882: cov[2+nagesqr+k1]=precov[nres][k1];
2883: }
2884: }/* End of loop on model equation */
2885:
2886: /* Start of old code (replaced by a loop on position in the model equation */
2887: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
2888: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
2889: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
2890: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
2891: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
2892: /* * k 1 2 3 4 5 6 7 8 */
2893: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
2894: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
2895: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
2896: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
2897: /* *nsd=3 (1) (2) (3) */
2898: /* *TvarsD[nsd] [1]=2 1 3 */
2899: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
2900: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
2901: /* *Tage[] [1]=1 [2]=2 [3]=3 */
2902: /* *Tvard[] [1][1]=1 [2][1]=1 */
2903: /* * [1][2]=3 [2][2]=2 */
2904: /* *Tprod[](=k) [1]=1 [2]=8 */
2905: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
2906: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
2907: /* *TvarsDpType */
2908: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
2909: /* * nsd=1 (1) (2) */
2910: /* *TvarsD[nsd] 3 2 */
2911: /* *TnsdVar (3)=1 (2)=2 */
2912: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
2913: /* *Tage[] [1]=2 [2]= 3 */
2914: /* *\/ */
2915: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
2916: /* /\* 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)); *\/ */
2917: /* } */
2918: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
2919: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
2920: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
2921: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
2922: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
2923: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2924: /* /\* 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]); *\/ */
2925: /* } */
2926: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
2927: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
2928: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
2929: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
2930: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
2931: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
2932: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2933: /* } */
2934: /* /\* 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]); *\/ */
2935: /* } */
2936: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
2937: /* /\* 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]); *\/ */
2938: /* if(Dummy[Tvard[k][1]]==0){ */
2939: /* if(Dummy[Tvard[k][2]]==0){ */
2940: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
2941: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
2942: /* }else{ */
2943: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
2944: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
2945: /* } */
2946: /* }else{ */
2947: /* if(Dummy[Tvard[k][2]]==0){ */
2948: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
2949: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
2950: /* }else{ */
2951: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
2952: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
2953: /* } */
2954: /* } */
2955: /* } /\* End product without age *\/ */
2956: /* ENd of old code */
1.138 brouard 2957: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2958: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2959: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2960: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2961: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 2962: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2963: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2964:
1.126 brouard 2965: savm=oldm;
2966: oldm=newm;
1.209 brouard 2967:
2968: for(j=1; j<=nlstate; j++){
2969: max[j]=0.;
2970: min[j]=1.;
2971: }
2972: for(i=1;i<=nlstate;i++){
2973: sumnew=0;
2974: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2975: for(j=1; j<=nlstate; j++){
2976: prlim[i][j]= newm[i][j]/(1-sumnew);
2977: max[j]=FMAX(max[j],prlim[i][j]);
2978: min[j]=FMIN(min[j],prlim[i][j]);
2979: }
2980: }
2981:
1.126 brouard 2982: maxmax=0.;
1.209 brouard 2983: for(j=1; j<=nlstate; j++){
2984: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2985: maxmax=FMAX(maxmax,meandiff[j]);
2986: /* 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 2987: } /* j loop */
1.203 brouard 2988: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2989: /* 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 2990: if(maxmax < ftolpl){
1.209 brouard 2991: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2992: free_vector(min,1,nlstate);
2993: free_vector(max,1,nlstate);
2994: free_vector(meandiff,1,nlstate);
1.126 brouard 2995: return prlim;
2996: }
1.288 brouard 2997: } /* agefin loop */
1.208 brouard 2998: /* After some age loop it doesn't converge */
1.288 brouard 2999: if(!first){
3000: first=1;
3001: 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 3002: 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);
3003: }else if (first >=1 && first <10){
3004: 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);
3005: first++;
3006: }else if (first ==10){
3007: 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);
3008: 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");
3009: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
3010: first++;
1.288 brouard 3011: }
3012:
1.209 brouard 3013: /* 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); */
3014: free_vector(min,1,nlstate);
3015: free_vector(max,1,nlstate);
3016: free_vector(meandiff,1,nlstate);
1.208 brouard 3017:
1.169 brouard 3018: return prlim; /* should not reach here */
1.126 brouard 3019: }
3020:
1.217 brouard 3021:
3022: /**** Back Prevalence limit (stable or period prevalence) ****************/
3023:
1.218 brouard 3024: /* 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) */
3025: /* 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 3026: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 3027: {
1.264 brouard 3028: /* 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 3029: matrix by transitions matrix until convergence is reached with precision ftolpl */
3030: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
3031: /* Wx is row vector: population in state 1, population in state 2, population dead */
3032: /* or prevalence in state 1, prevalence in state 2, 0 */
3033: /* newm is the matrix after multiplications, its rows are identical at a factor */
3034: /* Initial matrix pimij */
3035: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
3036: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
3037: /* 0, 0 , 1} */
3038: /*
3039: * and after some iteration: */
3040: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
3041: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
3042: /* 0, 0 , 1} */
3043: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
3044: /* {0.51571254859325999, 0.4842874514067399, */
3045: /* 0.51326036147820708, 0.48673963852179264} */
3046: /* If we start from prlim again, prlim tends to a constant matrix */
3047:
1.332 brouard 3048: int i, ii,j,k, k1;
1.247 brouard 3049: int first=0;
1.217 brouard 3050: double *min, *max, *meandiff, maxmax,sumnew=0.;
3051: /* double **matprod2(); */ /* test */
3052: double **out, cov[NCOVMAX+1], **bmij();
3053: double **newm;
1.218 brouard 3054: double **dnewm, **doldm, **dsavm; /* for use */
3055: double **oldm, **savm; /* for use */
3056:
1.217 brouard 3057: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
3058: int ncvloop=0;
3059:
3060: min=vector(1,nlstate);
3061: max=vector(1,nlstate);
3062: meandiff=vector(1,nlstate);
3063:
1.266 brouard 3064: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
3065: oldm=oldms; savm=savms;
3066:
3067: /* Starting with matrix unity */
3068: for (ii=1;ii<=nlstate+ndeath;ii++)
3069: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 3070: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3071: }
3072:
3073: cov[1]=1.;
3074:
3075: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3076: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 3077: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 3078: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
3079: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 3080: ncvloop++;
1.218 brouard 3081: newm=savm; /* oldm should be kept from previous iteration or unity at start */
3082: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 3083: /* Covariates have to be included here again */
3084: cov[2]=agefin;
1.319 brouard 3085: if(nagesqr==1){
1.217 brouard 3086: cov[3]= agefin*agefin;;
1.319 brouard 3087: }
1.332 brouard 3088: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3089: if(Typevar[k1]==1){ /* A product with age */
3090: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 3091: }else{
1.332 brouard 3092: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 3093: }
1.332 brouard 3094: }/* End of loop on model equation */
3095:
3096: /* Old code */
3097:
3098: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
3099: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
3100: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
3101: /* /\* 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)); *\/ */
3102: /* } */
3103: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
3104: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3105: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3106: /* /\* /\\* 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])]); *\\/ *\/ */
3107: /* /\* } *\/ */
3108: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3109: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3110: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3111: /* /\* 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]); *\/ */
3112: /* } */
3113: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
3114: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
3115: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
3116: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3117: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3118: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
3119: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3120: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3121: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3122: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3123: /* } */
3124: /* /\* 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]); *\/ */
3125: /* } */
3126: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3127: /* /\* 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]); *\/ */
3128: /* if(Dummy[Tvard[k][1]]==0){ */
3129: /* if(Dummy[Tvard[k][2]]==0){ */
3130: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3131: /* }else{ */
3132: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3133: /* } */
3134: /* }else{ */
3135: /* if(Dummy[Tvard[k][2]]==0){ */
3136: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3137: /* }else{ */
3138: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3139: /* } */
3140: /* } */
3141: /* } */
1.217 brouard 3142:
3143: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3144: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3145: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3146: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3147: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3148: /* ij should be linked to the correct index of cov */
3149: /* age and covariate values ij are in 'cov', but we need to pass
3150: * ij for the observed prevalence at age and status and covariate
3151: * number: prevacurrent[(int)agefin][ii][ij]
3152: */
3153: /* 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 *\/ */
3154: /* 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 *\/ */
3155: 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 3156: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3157: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3158: /* for(i=1; i<=nlstate+ndeath; i++) { */
3159: /* printf("%d newm= ",i); */
3160: /* for(j=1;j<=nlstate+ndeath;j++) { */
3161: /* printf("%f ",newm[i][j]); */
3162: /* } */
3163: /* printf("oldm * "); */
3164: /* for(j=1;j<=nlstate+ndeath;j++) { */
3165: /* printf("%f ",oldm[i][j]); */
3166: /* } */
1.268 brouard 3167: /* printf(" bmmij "); */
1.266 brouard 3168: /* for(j=1;j<=nlstate+ndeath;j++) { */
3169: /* printf("%f ",pmmij[i][j]); */
3170: /* } */
3171: /* printf("\n"); */
3172: /* } */
3173: /* } */
1.217 brouard 3174: savm=oldm;
3175: oldm=newm;
1.266 brouard 3176:
1.217 brouard 3177: for(j=1; j<=nlstate; j++){
3178: max[j]=0.;
3179: min[j]=1.;
3180: }
3181: for(j=1; j<=nlstate; j++){
3182: for(i=1;i<=nlstate;i++){
1.234 brouard 3183: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3184: bprlim[i][j]= newm[i][j];
3185: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3186: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3187: }
3188: }
1.218 brouard 3189:
1.217 brouard 3190: maxmax=0.;
3191: for(i=1; i<=nlstate; i++){
1.318 brouard 3192: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3193: maxmax=FMAX(maxmax,meandiff[i]);
3194: /* 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 3195: } /* i loop */
1.217 brouard 3196: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3197: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3198: if(maxmax < ftolpl){
1.220 brouard 3199: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3200: free_vector(min,1,nlstate);
3201: free_vector(max,1,nlstate);
3202: free_vector(meandiff,1,nlstate);
3203: return bprlim;
3204: }
1.288 brouard 3205: } /* agefin loop */
1.217 brouard 3206: /* After some age loop it doesn't converge */
1.288 brouard 3207: if(!first){
1.247 brouard 3208: first=1;
3209: 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\
3210: 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);
3211: }
3212: 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 3213: 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);
3214: /* 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); */
3215: free_vector(min,1,nlstate);
3216: free_vector(max,1,nlstate);
3217: free_vector(meandiff,1,nlstate);
3218:
3219: return bprlim; /* should not reach here */
3220: }
3221:
1.126 brouard 3222: /*************** transition probabilities ***************/
3223:
3224: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3225: {
1.138 brouard 3226: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3227: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3228: model to the ncovmodel covariates (including constant and age).
3229: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3230: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3231: ncth covariate in the global vector x is given by the formula:
3232: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3233: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3234: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3235: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3236: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3237: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3238: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3239: */
3240: double s1, lnpijopii;
1.126 brouard 3241: /*double t34;*/
1.164 brouard 3242: int i,j, nc, ii, jj;
1.126 brouard 3243:
1.223 brouard 3244: for(i=1; i<= nlstate; i++){
3245: for(j=1; j<i;j++){
3246: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3247: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3248: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3249: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3250: }
3251: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3252: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3253: }
3254: for(j=i+1; j<=nlstate+ndeath;j++){
3255: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3256: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3257: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3258: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3259: }
3260: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3261: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3262: }
3263: }
1.218 brouard 3264:
1.223 brouard 3265: for(i=1; i<= nlstate; i++){
3266: s1=0;
3267: for(j=1; j<i; j++){
3268: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 brouard 3269: /* 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 3270: }
3271: for(j=i+1; j<=nlstate+ndeath; j++){
3272: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 brouard 3273: /* 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 3274: }
3275: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3276: ps[i][i]=1./(s1+1.);
3277: /* Computing other pijs */
3278: for(j=1; j<i; j++)
1.325 brouard 3279: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3280: for(j=i+1; j<=nlstate+ndeath; j++)
3281: ps[i][j]= exp(ps[i][j])*ps[i][i];
3282: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3283: } /* end i */
1.218 brouard 3284:
1.223 brouard 3285: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3286: for(jj=1; jj<= nlstate+ndeath; jj++){
3287: ps[ii][jj]=0;
3288: ps[ii][ii]=1;
3289: }
3290: }
1.294 brouard 3291:
3292:
1.223 brouard 3293: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3294: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3295: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3296: /* } */
3297: /* printf("\n "); */
3298: /* } */
3299: /* printf("\n ");printf("%lf ",cov[2]);*/
3300: /*
3301: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3302: goto end;*/
1.266 brouard 3303: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3304: }
3305:
1.218 brouard 3306: /*************** backward transition probabilities ***************/
3307:
3308: /* 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 ) */
3309: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3310: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3311: {
1.302 brouard 3312: /* 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 3313: * 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 3314: */
1.218 brouard 3315: int i, ii, j,k;
1.222 brouard 3316:
3317: double **out, **pmij();
3318: double sumnew=0.;
1.218 brouard 3319: double agefin;
1.292 brouard 3320: 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 3321: double **dnewm, **dsavm, **doldm;
3322: double **bbmij;
3323:
1.218 brouard 3324: doldm=ddoldms; /* global pointers */
1.222 brouard 3325: dnewm=ddnewms;
3326: dsavm=ddsavms;
1.318 brouard 3327:
3328: /* Debug */
3329: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3330: agefin=cov[2];
1.268 brouard 3331: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3332: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3333: the observed prevalence (with this covariate ij) at beginning of transition */
3334: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3335:
3336: /* P_x */
1.325 brouard 3337: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3338: /* outputs pmmij which is a stochastic matrix in row */
3339:
3340: /* Diag(w_x) */
1.292 brouard 3341: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3342: sumnew=0.;
1.269 brouard 3343: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3344: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3345: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3346: sumnew+=prevacurrent[(int)agefin][ii][ij];
3347: }
3348: if(sumnew >0.01){ /* At least some value in the prevalence */
3349: for (ii=1;ii<=nlstate+ndeath;ii++){
3350: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3351: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3352: }
3353: }else{
3354: for (ii=1;ii<=nlstate+ndeath;ii++){
3355: for (j=1;j<=nlstate+ndeath;j++)
3356: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3357: }
3358: /* if(sumnew <0.9){ */
3359: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3360: /* } */
3361: }
3362: k3=0.0; /* We put the last diagonal to 0 */
3363: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3364: doldm[ii][ii]= k3;
3365: }
3366: /* End doldm, At the end doldm is diag[(w_i)] */
3367:
1.292 brouard 3368: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3369: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3370:
1.292 brouard 3371: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3372: /* 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 3373: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3374: sumnew=0.;
1.222 brouard 3375: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3376: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3377: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3378: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3379: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3380: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3381: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3382: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3383: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3384: /* }else */
1.268 brouard 3385: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3386: } /*End ii */
3387: } /* 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 */
3388:
1.292 brouard 3389: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3390: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3391: /* end bmij */
1.266 brouard 3392: return ps; /*pointer is unchanged */
1.218 brouard 3393: }
1.217 brouard 3394: /*************** transition probabilities ***************/
3395:
1.218 brouard 3396: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3397: {
3398: /* According to parameters values stored in x and the covariate's values stored in cov,
3399: computes the probability to be observed in state j being in state i by appying the
3400: model to the ncovmodel covariates (including constant and age).
3401: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3402: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3403: ncth covariate in the global vector x is given by the formula:
3404: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3405: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3406: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3407: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3408: Outputs ps[i][j] the probability to be observed in j being in j according to
3409: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3410: */
3411: double s1, lnpijopii;
3412: /*double t34;*/
3413: int i,j, nc, ii, jj;
3414:
1.234 brouard 3415: for(i=1; i<= nlstate; i++){
3416: for(j=1; j<i;j++){
3417: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3418: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3419: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3420: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3421: }
3422: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3423: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3424: }
3425: for(j=i+1; j<=nlstate+ndeath;j++){
3426: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3427: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3428: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3429: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3430: }
3431: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3432: }
3433: }
3434:
3435: for(i=1; i<= nlstate; i++){
3436: s1=0;
3437: for(j=1; j<i; j++){
3438: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3439: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3440: }
3441: for(j=i+1; j<=nlstate+ndeath; j++){
3442: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3443: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3444: }
3445: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3446: ps[i][i]=1./(s1+1.);
3447: /* Computing other pijs */
3448: for(j=1; j<i; j++)
3449: ps[i][j]= exp(ps[i][j])*ps[i][i];
3450: for(j=i+1; j<=nlstate+ndeath; j++)
3451: ps[i][j]= exp(ps[i][j])*ps[i][i];
3452: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3453: } /* end i */
3454:
3455: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3456: for(jj=1; jj<= nlstate+ndeath; jj++){
3457: ps[ii][jj]=0;
3458: ps[ii][ii]=1;
3459: }
3460: }
1.296 brouard 3461: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3462: for(jj=1; jj<= nlstate+ndeath; jj++){
3463: s1=0.;
3464: for(ii=1; ii<= nlstate+ndeath; ii++){
3465: s1+=ps[ii][jj];
3466: }
3467: for(ii=1; ii<= nlstate; ii++){
3468: ps[ii][jj]=ps[ii][jj]/s1;
3469: }
3470: }
3471: /* Transposition */
3472: for(jj=1; jj<= nlstate+ndeath; jj++){
3473: for(ii=jj; ii<= nlstate+ndeath; ii++){
3474: s1=ps[ii][jj];
3475: ps[ii][jj]=ps[jj][ii];
3476: ps[jj][ii]=s1;
3477: }
3478: }
3479: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3480: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3481: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3482: /* } */
3483: /* printf("\n "); */
3484: /* } */
3485: /* printf("\n ");printf("%lf ",cov[2]);*/
3486: /*
3487: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3488: goto end;*/
3489: return ps;
1.217 brouard 3490: }
3491:
3492:
1.126 brouard 3493: /**************** Product of 2 matrices ******************/
3494:
1.145 brouard 3495: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3496: {
3497: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3498: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3499: /* in, b, out are matrice of pointers which should have been initialized
3500: before: only the contents of out is modified. The function returns
3501: a pointer to pointers identical to out */
1.145 brouard 3502: int i, j, k;
1.126 brouard 3503: for(i=nrl; i<= nrh; i++)
1.145 brouard 3504: for(k=ncolol; k<=ncoloh; k++){
3505: out[i][k]=0.;
3506: for(j=ncl; j<=nch; j++)
3507: out[i][k] +=in[i][j]*b[j][k];
3508: }
1.126 brouard 3509: return out;
3510: }
3511:
3512:
3513: /************* Higher Matrix Product ***************/
3514:
1.235 brouard 3515: 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 3516: {
1.336 brouard 3517: /* Already optimized with precov.
3518: 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 3519: 'nhstepm*hstepm*stepm' months (i.e. until
3520: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3521: nhstepm*hstepm matrices.
3522: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3523: (typically every 2 years instead of every month which is too big
3524: for the memory).
3525: Model is determined by parameters x and covariates have to be
3526: included manually here.
3527:
3528: */
3529:
1.330 brouard 3530: int i, j, d, h, k, k1;
1.131 brouard 3531: double **out, cov[NCOVMAX+1];
1.126 brouard 3532: double **newm;
1.187 brouard 3533: double agexact;
1.214 brouard 3534: double agebegin, ageend;
1.126 brouard 3535:
3536: /* Hstepm could be zero and should return the unit matrix */
3537: for (i=1;i<=nlstate+ndeath;i++)
3538: for (j=1;j<=nlstate+ndeath;j++){
3539: oldm[i][j]=(i==j ? 1.0 : 0.0);
3540: po[i][j][0]=(i==j ? 1.0 : 0.0);
3541: }
3542: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3543: for(h=1; h <=nhstepm; h++){
3544: for(d=1; d <=hstepm; d++){
3545: newm=savm;
3546: /* Covariates have to be included here again */
3547: cov[1]=1.;
1.214 brouard 3548: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3549: cov[2]=agexact;
1.319 brouard 3550: if(nagesqr==1){
1.227 brouard 3551: cov[3]= agexact*agexact;
1.319 brouard 3552: }
1.330 brouard 3553: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
3554: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
3555: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.332 brouard 3556: if(Typevar[k1]==1){ /* A product with age */
3557: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
3558: }else{
3559: cov[2+nagesqr+k1]=precov[nres][k1];
3560: }
3561: }/* End of loop on model equation */
3562: /* Old code */
3563: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
3564: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
3565: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
3566: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
3567: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
3568: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3569: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3570: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
3571: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
3572: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
3573: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
3574: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
3575: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3576: /* /\* 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]])); *\/ */
3577: /* 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); */
3578: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3579: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
3580: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
3581: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
3582: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
3583: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
3584: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3585: /* 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]]); */
3586: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3587: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
3588: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
3589: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
3590: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
3591: /* 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]); */
3592: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3593:
3594: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
3595: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
3596: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
3597: /* /\* *\/ */
1.330 brouard 3598: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3599: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3600: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 3601: /* /\*cptcovage=2 1 2 *\/ */
3602: /* /\*Tage[k]= 5 8 *\/ */
3603: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
3604: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3605: /* 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]]); */
3606: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3607: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
3608: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
3609: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
3610: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
3611: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
3612: /* /\* 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); *\/ */
3613: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
3614: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
3615: /* /\* } *\/ */
3616: /* /\* 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]); *\/ */
3617: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
3618: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
3619: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
3620: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
3621: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
3622: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
3623: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
3624: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
3625: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 3626:
1.332 brouard 3627: /* /\* 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])]); *\/ */
3628: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3629: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
3630: /* 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]]); */
3631: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3632:
3633: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
3634: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
3635: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3636: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
3637: /* /\* 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]])]; *\/ */
3638: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
3639: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
3640: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
3641: /* /\* } *\/ */
3642: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
3643: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
3644: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
3645: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3646: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
3647: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
3648: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3649: /* /\* } *\/ */
3650: /* /\* }/\\*end of products quantitative *\\/ *\/ */
3651: /* }/\*end of products *\/ */
3652: /* } /\* End of loop on model equation *\/ */
1.235 brouard 3653: /* for (k=1; k<=cptcovn;k++) */
3654: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3655: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3656: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3657: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3658: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3659:
3660:
1.126 brouard 3661: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3662: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3663: /* right multiplication of oldm by the current matrix */
1.126 brouard 3664: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3665: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3666: /* if((int)age == 70){ */
3667: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3668: /* for(i=1; i<=nlstate+ndeath; i++) { */
3669: /* printf("%d pmmij ",i); */
3670: /* for(j=1;j<=nlstate+ndeath;j++) { */
3671: /* printf("%f ",pmmij[i][j]); */
3672: /* } */
3673: /* printf(" oldm "); */
3674: /* for(j=1;j<=nlstate+ndeath;j++) { */
3675: /* printf("%f ",oldm[i][j]); */
3676: /* } */
3677: /* printf("\n"); */
3678: /* } */
3679: /* } */
1.126 brouard 3680: savm=oldm;
3681: oldm=newm;
3682: }
3683: for(i=1; i<=nlstate+ndeath; i++)
3684: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3685: po[i][j][h]=newm[i][j];
3686: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3687: }
1.128 brouard 3688: /*printf("h=%d ",h);*/
1.126 brouard 3689: } /* end h */
1.267 brouard 3690: /* printf("\n H=%d \n",h); */
1.126 brouard 3691: return po;
3692: }
3693:
1.217 brouard 3694: /************* Higher Back Matrix Product ***************/
1.218 brouard 3695: /* 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 3696: 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 3697: {
1.332 brouard 3698: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
3699: computes the transition matrix starting at age 'age' over
1.217 brouard 3700: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3701: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3702: nhstepm*hstepm matrices.
3703: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3704: (typically every 2 years instead of every month which is too big
1.217 brouard 3705: for the memory).
1.218 brouard 3706: Model is determined by parameters x and covariates have to be
1.266 brouard 3707: included manually here. Then we use a call to bmij(x and cov)
3708: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3709: */
1.217 brouard 3710:
1.332 brouard 3711: int i, j, d, h, k, k1;
1.266 brouard 3712: double **out, cov[NCOVMAX+1], **bmij();
3713: double **newm, ***newmm;
1.217 brouard 3714: double agexact;
3715: double agebegin, ageend;
1.222 brouard 3716: double **oldm, **savm;
1.217 brouard 3717:
1.266 brouard 3718: newmm=po; /* To be saved */
3719: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3720: /* Hstepm could be zero and should return the unit matrix */
3721: for (i=1;i<=nlstate+ndeath;i++)
3722: for (j=1;j<=nlstate+ndeath;j++){
3723: oldm[i][j]=(i==j ? 1.0 : 0.0);
3724: po[i][j][0]=(i==j ? 1.0 : 0.0);
3725: }
3726: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3727: for(h=1; h <=nhstepm; h++){
3728: for(d=1; d <=hstepm; d++){
3729: newm=savm;
3730: /* Covariates have to be included here again */
3731: cov[1]=1.;
1.271 brouard 3732: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3733: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3734: /* Debug */
3735: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3736: cov[2]=agexact;
1.332 brouard 3737: if(nagesqr==1){
1.222 brouard 3738: cov[3]= agexact*agexact;
1.332 brouard 3739: }
3740: /** New code */
3741: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3742: if(Typevar[k1]==1){ /* A product with age */
3743: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 3744: }else{
1.332 brouard 3745: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 3746: }
1.332 brouard 3747: }/* End of loop on model equation */
3748: /** End of new code */
3749: /** This was old code */
3750: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
3751: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3752: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3753: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
3754: /* /\* 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)); *\/ */
3755: /* } */
3756: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3757: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3758: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3759: /* /\* 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]); *\/ */
3760: /* } */
3761: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
3762: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
3763: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3764: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3765: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3766: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3767: /* } */
3768: /* /\* 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]); *\/ */
3769: /* } */
3770: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
3771: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3772: /* if(Dummy[Tvard[k][1]]==0){ */
3773: /* if(Dummy[Tvard[k][2]]==0){ */
3774: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
3775: /* }else{ */
3776: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3777: /* } */
3778: /* }else{ */
3779: /* if(Dummy[Tvard[k][2]]==0){ */
3780: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3781: /* }else{ */
3782: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3783: /* } */
3784: /* } */
3785: /* } */
3786: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
3787: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
3788: /** End of old code */
3789:
1.218 brouard 3790: /* Careful transposed matrix */
1.266 brouard 3791: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3792: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3793: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3794: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3795: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3796: /* if((int)age == 70){ */
3797: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3798: /* for(i=1; i<=nlstate+ndeath; i++) { */
3799: /* printf("%d pmmij ",i); */
3800: /* for(j=1;j<=nlstate+ndeath;j++) { */
3801: /* printf("%f ",pmmij[i][j]); */
3802: /* } */
3803: /* printf(" oldm "); */
3804: /* for(j=1;j<=nlstate+ndeath;j++) { */
3805: /* printf("%f ",oldm[i][j]); */
3806: /* } */
3807: /* printf("\n"); */
3808: /* } */
3809: /* } */
3810: savm=oldm;
3811: oldm=newm;
3812: }
3813: for(i=1; i<=nlstate+ndeath; i++)
3814: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3815: po[i][j][h]=newm[i][j];
1.268 brouard 3816: /* if(h==nhstepm) */
3817: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3818: }
1.268 brouard 3819: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3820: } /* end h */
1.268 brouard 3821: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3822: return po;
3823: }
3824:
3825:
1.162 brouard 3826: #ifdef NLOPT
3827: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3828: double fret;
3829: double *xt;
3830: int j;
3831: myfunc_data *d2 = (myfunc_data *) pd;
3832: /* xt = (p1-1); */
3833: xt=vector(1,n);
3834: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3835:
3836: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3837: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3838: printf("Function = %.12lf ",fret);
3839: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3840: printf("\n");
3841: free_vector(xt,1,n);
3842: return fret;
3843: }
3844: #endif
1.126 brouard 3845:
3846: /*************** log-likelihood *************/
3847: double func( double *x)
3848: {
1.336 brouard 3849: int i, ii, j, k, mi, d, kk, kf=0;
1.226 brouard 3850: int ioffset=0;
3851: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3852: double **out;
3853: double lli; /* Individual log likelihood */
3854: int s1, s2;
1.228 brouard 3855: 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 3856:
1.226 brouard 3857: double bbh, survp;
3858: double agexact;
1.336 brouard 3859: double agebegin, ageend;
1.226 brouard 3860: /*extern weight */
3861: /* We are differentiating ll according to initial status */
3862: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3863: /*for(i=1;i<imx;i++)
3864: printf(" %d\n",s[4][i]);
3865: */
1.162 brouard 3866:
1.226 brouard 3867: ++countcallfunc;
1.162 brouard 3868:
1.226 brouard 3869: cov[1]=1.;
1.126 brouard 3870:
1.226 brouard 3871: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3872: ioffset=0;
1.226 brouard 3873: if(mle==1){
3874: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3875: /* Computes the values of the ncovmodel covariates of the model
3876: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3877: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3878: to be observed in j being in i according to the model.
3879: */
1.243 brouard 3880: ioffset=2+nagesqr ;
1.233 brouard 3881: /* Fixed */
1.336 brouard 3882: for (kf=1; kf<=ncovf;kf++){ /* For each fixed covariate dummu or quant or prod */
1.319 brouard 3883: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3884: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3885: /* 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 3886: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.336 brouard 3887: 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 3888: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3889: }
1.226 brouard 3890: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3891: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3892: has been calculated etc */
3893: /* For an individual i, wav[i] gives the number of effective waves */
3894: /* We compute the contribution to Likelihood of each effective transition
3895: mw[mi][i] is real wave of the mi th effectve wave */
3896: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3897: s2=s[mw[mi+1][i]][i];
3898: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3899: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3900: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3901: */
1.336 brouard 3902: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
3903: /* Wave varying (but not age varying) */
1.319 brouard 3904: 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*/
3905: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? */
1.242 brouard 3906: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3907: }
3908: for (ii=1;ii<=nlstate+ndeath;ii++)
3909: for (j=1;j<=nlstate+ndeath;j++){
3910: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3911: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3912: }
1.336 brouard 3913:
3914: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3915: 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 3916: for(d=0; d<dh[mi][i]; d++){
3917: newm=savm;
3918: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3919: cov[2]=agexact;
3920: if(nagesqr==1)
3921: cov[3]= agexact*agexact; /* Should be changed here */
3922: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3923: if(!FixedV[Tvar[Tage[kk]]])
3924: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3925: else
3926: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3927: }
3928: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3929: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3930: savm=oldm;
3931: oldm=newm;
3932: } /* end mult */
3933:
3934: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3935: /* But now since version 0.9 we anticipate for bias at large stepm.
3936: * If stepm is larger than one month (smallest stepm) and if the exact delay
3937: * (in months) between two waves is not a multiple of stepm, we rounded to
3938: * the nearest (and in case of equal distance, to the lowest) interval but now
3939: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3940: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3941: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3942: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3943: * -stepm/2 to stepm/2 .
3944: * For stepm=1 the results are the same as for previous versions of Imach.
3945: * For stepm > 1 the results are less biased than in previous versions.
3946: */
1.234 brouard 3947: s1=s[mw[mi][i]][i];
3948: s2=s[mw[mi+1][i]][i];
3949: bbh=(double)bh[mi][i]/(double)stepm;
3950: /* bias bh is positive if real duration
3951: * is higher than the multiple of stepm and negative otherwise.
3952: */
3953: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3954: if( s2 > nlstate){
3955: /* i.e. if s2 is a death state and if the date of death is known
3956: then the contribution to the likelihood is the probability to
3957: die between last step unit time and current step unit time,
3958: which is also equal to probability to die before dh
3959: minus probability to die before dh-stepm .
3960: In version up to 0.92 likelihood was computed
3961: as if date of death was unknown. Death was treated as any other
3962: health state: the date of the interview describes the actual state
3963: and not the date of a change in health state. The former idea was
3964: to consider that at each interview the state was recorded
3965: (healthy, disable or death) and IMaCh was corrected; but when we
3966: introduced the exact date of death then we should have modified
3967: the contribution of an exact death to the likelihood. This new
3968: contribution is smaller and very dependent of the step unit
3969: stepm. It is no more the probability to die between last interview
3970: and month of death but the probability to survive from last
3971: interview up to one month before death multiplied by the
3972: probability to die within a month. Thanks to Chris
3973: Jackson for correcting this bug. Former versions increased
3974: mortality artificially. The bad side is that we add another loop
3975: which slows down the processing. The difference can be up to 10%
3976: lower mortality.
3977: */
3978: /* If, at the beginning of the maximization mostly, the
3979: cumulative probability or probability to be dead is
3980: constant (ie = 1) over time d, the difference is equal to
3981: 0. out[s1][3] = savm[s1][3]: probability, being at state
3982: s1 at precedent wave, to be dead a month before current
3983: wave is equal to probability, being at state s1 at
3984: precedent wave, to be dead at mont of the current
3985: wave. Then the observed probability (that this person died)
3986: is null according to current estimated parameter. In fact,
3987: it should be very low but not zero otherwise the log go to
3988: infinity.
3989: */
1.183 brouard 3990: /* #ifdef INFINITYORIGINAL */
3991: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3992: /* #else */
3993: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3994: /* lli=log(mytinydouble); */
3995: /* else */
3996: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3997: /* #endif */
1.226 brouard 3998: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3999:
1.226 brouard 4000: } else if ( s2==-1 ) { /* alive */
4001: for (j=1,survp=0. ; j<=nlstate; j++)
4002: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4003: /*survp += out[s1][j]; */
4004: lli= log(survp);
4005: }
1.336 brouard 4006: /* else if (s2==-4) { */
4007: /* for (j=3,survp=0. ; j<=nlstate; j++) */
4008: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4009: /* lli= log(survp); */
4010: /* } */
4011: /* else if (s2==-5) { */
4012: /* for (j=1,survp=0. ; j<=2; j++) */
4013: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4014: /* lli= log(survp); */
4015: /* } */
1.226 brouard 4016: else{
4017: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4018: /* 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 */
4019: }
4020: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
4021: /*if(lli ==000.0)*/
4022: /*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); */
4023: ipmx +=1;
4024: sw += weight[i];
4025: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4026: /* if (lli < log(mytinydouble)){ */
4027: /* 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); */
4028: /* 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]); */
4029: /* } */
4030: } /* end of wave */
4031: } /* end of individual */
4032: } else if(mle==2){
4033: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 4034: ioffset=2+nagesqr ;
4035: for (k=1; k<=ncovf;k++)
4036: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 4037: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 4038: for(k=1; k <= ncovv ; k++){
4039: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4040: }
1.226 brouard 4041: for (ii=1;ii<=nlstate+ndeath;ii++)
4042: for (j=1;j<=nlstate+ndeath;j++){
4043: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4044: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4045: }
4046: for(d=0; d<=dh[mi][i]; d++){
4047: newm=savm;
4048: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4049: cov[2]=agexact;
4050: if(nagesqr==1)
4051: cov[3]= agexact*agexact;
4052: for (kk=1; kk<=cptcovage;kk++) {
4053: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4054: }
4055: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4056: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4057: savm=oldm;
4058: oldm=newm;
4059: } /* end mult */
4060:
4061: s1=s[mw[mi][i]][i];
4062: s2=s[mw[mi+1][i]][i];
4063: bbh=(double)bh[mi][i]/(double)stepm;
4064: 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 */
4065: ipmx +=1;
4066: sw += weight[i];
4067: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4068: } /* end of wave */
4069: } /* end of individual */
4070: } else if(mle==3){ /* exponential inter-extrapolation */
4071: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4072: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4073: for(mi=1; mi<= wav[i]-1; mi++){
4074: for (ii=1;ii<=nlstate+ndeath;ii++)
4075: for (j=1;j<=nlstate+ndeath;j++){
4076: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4077: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4078: }
4079: for(d=0; d<dh[mi][i]; d++){
4080: newm=savm;
4081: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4082: cov[2]=agexact;
4083: if(nagesqr==1)
4084: cov[3]= agexact*agexact;
4085: for (kk=1; kk<=cptcovage;kk++) {
4086: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4087: }
4088: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4089: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4090: savm=oldm;
4091: oldm=newm;
4092: } /* end mult */
4093:
4094: s1=s[mw[mi][i]][i];
4095: s2=s[mw[mi+1][i]][i];
4096: bbh=(double)bh[mi][i]/(double)stepm;
4097: 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 */
4098: ipmx +=1;
4099: sw += weight[i];
4100: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4101: } /* end of wave */
4102: } /* end of individual */
4103: }else if (mle==4){ /* ml=4 no inter-extrapolation */
4104: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4105: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4106: for(mi=1; mi<= wav[i]-1; mi++){
4107: for (ii=1;ii<=nlstate+ndeath;ii++)
4108: for (j=1;j<=nlstate+ndeath;j++){
4109: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4110: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4111: }
4112: for(d=0; d<dh[mi][i]; d++){
4113: newm=savm;
4114: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4115: cov[2]=agexact;
4116: if(nagesqr==1)
4117: cov[3]= agexact*agexact;
4118: for (kk=1; kk<=cptcovage;kk++) {
4119: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4120: }
1.126 brouard 4121:
1.226 brouard 4122: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4123: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4124: savm=oldm;
4125: oldm=newm;
4126: } /* end mult */
4127:
4128: s1=s[mw[mi][i]][i];
4129: s2=s[mw[mi+1][i]][i];
4130: if( s2 > nlstate){
4131: lli=log(out[s1][s2] - savm[s1][s2]);
4132: } else if ( s2==-1 ) { /* alive */
4133: for (j=1,survp=0. ; j<=nlstate; j++)
4134: survp += out[s1][j];
4135: lli= log(survp);
4136: }else{
4137: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4138: }
4139: ipmx +=1;
4140: sw += weight[i];
4141: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 4142: /* 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 4143: } /* end of wave */
4144: } /* end of individual */
4145: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4146: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4147: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4148: for(mi=1; mi<= wav[i]-1; mi++){
4149: for (ii=1;ii<=nlstate+ndeath;ii++)
4150: for (j=1;j<=nlstate+ndeath;j++){
4151: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4152: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4153: }
4154: for(d=0; d<dh[mi][i]; d++){
4155: newm=savm;
4156: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4157: cov[2]=agexact;
4158: if(nagesqr==1)
4159: cov[3]= agexact*agexact;
4160: for (kk=1; kk<=cptcovage;kk++) {
4161: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4162: }
1.126 brouard 4163:
1.226 brouard 4164: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4165: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4166: savm=oldm;
4167: oldm=newm;
4168: } /* end mult */
4169:
4170: s1=s[mw[mi][i]][i];
4171: s2=s[mw[mi+1][i]][i];
4172: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4173: ipmx +=1;
4174: sw += weight[i];
4175: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4176: /*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]);*/
4177: } /* end of wave */
4178: } /* end of individual */
4179: } /* End of if */
4180: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4181: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4182: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4183: return -l;
1.126 brouard 4184: }
4185:
4186: /*************** log-likelihood *************/
4187: double funcone( double *x)
4188: {
1.228 brouard 4189: /* Same as func but slower because of a lot of printf and if */
1.335 brouard 4190: int i, ii, j, k, mi, d, kk, kf=0;
1.228 brouard 4191: int ioffset=0;
1.131 brouard 4192: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4193: double **out;
4194: double lli; /* Individual log likelihood */
4195: double llt;
4196: int s1, s2;
1.228 brouard 4197: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4198:
1.126 brouard 4199: double bbh, survp;
1.187 brouard 4200: double agexact;
1.214 brouard 4201: double agebegin, ageend;
1.126 brouard 4202: /*extern weight */
4203: /* We are differentiating ll according to initial status */
4204: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4205: /*for(i=1;i<imx;i++)
4206: printf(" %d\n",s[4][i]);
4207: */
4208: cov[1]=1.;
4209:
4210: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4211: ioffset=0;
4212: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.336 brouard 4213: /* Computes the values of the ncovmodel covariates of the model
4214: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
4215: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
4216: to be observed in j being in i according to the model.
4217: */
1.243 brouard 4218: /* ioffset=2+nagesqr+cptcovage; */
4219: ioffset=2+nagesqr;
1.232 brouard 4220: /* Fixed */
1.224 brouard 4221: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4222: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.335 brouard 4223: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
4224: 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 4225: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4226: /* cov[2+6]=covar[Tvar[6]][i]; */
4227: /* cov[2+6]=covar[2][i]; V2 */
4228: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4229: /* cov[2+7]=covar[Tvar[7]][i]; */
4230: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4231: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4232: /* cov[2+9]=covar[Tvar[9]][i]; */
4233: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4234: }
1.336 brouard 4235: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
4236: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
4237: has been calculated etc */
4238: /* For an individual i, wav[i] gives the number of effective waves */
4239: /* We compute the contribution to Likelihood of each effective transition
4240: mw[mi][i] is real wave of the mi th effectve wave */
4241: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
4242: s2=s[mw[mi+1][i]][i];
4243: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
4244: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
4245: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
4246: */
4247: /* This part may be useless now because everythin should be in covar */
1.232 brouard 4248: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4249: /* 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?)*\/ */
4250: /* } */
1.231 brouard 4251: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4252: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4253: /* } */
1.225 brouard 4254:
1.233 brouard 4255:
4256: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 4257: /* Wave varying (but not age varying) */
4258: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 4259: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
4260: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4261: }
1.232 brouard 4262: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 4263: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4264: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4265: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4266: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4267: /* 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 4268: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4269: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4270: /* /\* 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]); *\/ */
4271: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4272: /* } */
1.126 brouard 4273: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4274: for (j=1;j<=nlstate+ndeath;j++){
4275: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4276: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4277: }
1.214 brouard 4278:
4279: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4280: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4281: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4282: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4283: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4284: and mw[mi+1][i]. dh depends on stepm.*/
4285: newm=savm;
1.247 brouard 4286: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4287: cov[2]=agexact;
4288: if(nagesqr==1)
4289: cov[3]= agexact*agexact;
4290: for (kk=1; kk<=cptcovage;kk++) {
4291: if(!FixedV[Tvar[Tage[kk]]])
4292: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4293: else
4294: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
4295: }
4296: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4297: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4298: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4299: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4300: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4301: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4302: savm=oldm;
4303: oldm=newm;
1.126 brouard 4304: } /* end mult */
1.336 brouard 4305: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4306: /* But now since version 0.9 we anticipate for bias at large stepm.
4307: * If stepm is larger than one month (smallest stepm) and if the exact delay
4308: * (in months) between two waves is not a multiple of stepm, we rounded to
4309: * the nearest (and in case of equal distance, to the lowest) interval but now
4310: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4311: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4312: * probability in order to take into account the bias as a fraction of the way
4313: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4314: * -stepm/2 to stepm/2 .
4315: * For stepm=1 the results are the same as for previous versions of Imach.
4316: * For stepm > 1 the results are less biased than in previous versions.
4317: */
1.126 brouard 4318: s1=s[mw[mi][i]][i];
4319: s2=s[mw[mi+1][i]][i];
1.217 brouard 4320: /* if(s2==-1){ */
1.268 brouard 4321: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4322: /* /\* exit(1); *\/ */
4323: /* } */
1.126 brouard 4324: bbh=(double)bh[mi][i]/(double)stepm;
4325: /* bias is positive if real duration
4326: * is higher than the multiple of stepm and negative otherwise.
4327: */
4328: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4329: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4330: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4331: for (j=1,survp=0. ; j<=nlstate; j++)
4332: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4333: lli= log(survp);
1.126 brouard 4334: }else if (mle==1){
1.242 brouard 4335: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4336: } else if(mle==2){
1.242 brouard 4337: 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 4338: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4339: 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 4340: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4341: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4342: } else{ /* mle=0 back to 1 */
1.242 brouard 4343: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4344: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4345: } /* End of if */
4346: ipmx +=1;
4347: sw += weight[i];
4348: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.335 brouard 4349: /* 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 4350: if(globpr){
1.246 brouard 4351: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4352: %11.6f %11.6f %11.6f ", \
1.242 brouard 4353: 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 4354: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.335 brouard 4355: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
4356: /* %11.6f %11.6f %11.6f ", \ */
4357: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
4358: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 4359: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4360: llt +=ll[k]*gipmx/gsw;
4361: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 brouard 4362: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 4363: }
4364: fprintf(ficresilk," %10.6f\n", -llt);
1.335 brouard 4365: /* printf(" %10.6f\n", -llt); */
1.126 brouard 4366: }
1.335 brouard 4367: } /* end of wave */
4368: } /* end of individual */
4369: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 4370: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 brouard 4371: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4372: if(globpr==0){ /* First time we count the contributions and weights */
4373: gipmx=ipmx;
4374: gsw=sw;
4375: }
1.232 brouard 4376: return -l;
1.126 brouard 4377: }
4378:
4379:
4380: /*************** function likelione ***********/
1.292 brouard 4381: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4382: {
4383: /* This routine should help understanding what is done with
4384: the selection of individuals/waves and
4385: to check the exact contribution to the likelihood.
4386: Plotting could be done.
4387: */
4388: int k;
4389:
4390: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4391: strcpy(fileresilk,"ILK_");
1.202 brouard 4392: strcat(fileresilk,fileresu);
1.126 brouard 4393: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4394: printf("Problem with resultfile: %s\n", fileresilk);
4395: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4396: }
1.214 brouard 4397: 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");
4398: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4399: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4400: for(k=1; k<=nlstate; k++)
4401: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4402: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4403: }
4404:
1.292 brouard 4405: *fretone=(*func)(p);
1.126 brouard 4406: if(*globpri !=0){
4407: fclose(ficresilk);
1.205 brouard 4408: if (mle ==0)
4409: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4410: else if(mle >=1)
4411: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4412: 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 4413: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4414:
4415: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4416: 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 4417: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4418: }
1.207 brouard 4419: 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 4420: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4421: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4422: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4423: fflush(fichtm);
1.205 brouard 4424: }
1.126 brouard 4425: return;
4426: }
4427:
4428:
4429: /*********** Maximum Likelihood Estimation ***************/
4430:
4431: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4432: {
1.319 brouard 4433: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4434: double **xi;
4435: double fret;
4436: double fretone; /* Only one call to likelihood */
4437: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4438:
4439: #ifdef NLOPT
4440: int creturn;
4441: nlopt_opt opt;
4442: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4443: double *lb;
4444: double minf; /* the minimum objective value, upon return */
4445: double * p1; /* Shifted parameters from 0 instead of 1 */
4446: myfunc_data dinst, *d = &dinst;
4447: #endif
4448:
4449:
1.126 brouard 4450: xi=matrix(1,npar,1,npar);
4451: for (i=1;i<=npar;i++)
4452: for (j=1;j<=npar;j++)
4453: xi[i][j]=(i==j ? 1.0 : 0.0);
4454: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4455: strcpy(filerespow,"POW_");
1.126 brouard 4456: strcat(filerespow,fileres);
4457: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4458: printf("Problem with resultfile: %s\n", filerespow);
4459: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4460: }
4461: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4462: for (i=1;i<=nlstate;i++)
4463: for(j=1;j<=nlstate+ndeath;j++)
4464: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4465: fprintf(ficrespow,"\n");
1.162 brouard 4466: #ifdef POWELL
1.319 brouard 4467: #ifdef LINMINORIGINAL
4468: #else /* LINMINORIGINAL */
4469:
4470: flatdir=ivector(1,npar);
4471: for (j=1;j<=npar;j++) flatdir[j]=0;
4472: #endif /*LINMINORIGINAL */
4473:
4474: #ifdef FLATSUP
4475: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4476: /* reorganizing p by suppressing flat directions */
4477: for(i=1, jk=1; i <=nlstate; i++){
4478: for(k=1; k <=(nlstate+ndeath); k++){
4479: if (k != i) {
4480: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4481: if(flatdir[jk]==1){
4482: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4483: }
4484: for(j=1; j <=ncovmodel; j++){
4485: printf("%12.7f ",p[jk]);
4486: jk++;
4487: }
4488: printf("\n");
4489: }
4490: }
4491: }
4492: /* skipping */
4493: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4494: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4495: for(k=1; k <=(nlstate+ndeath); k++){
4496: if (k != i) {
4497: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4498: if(flatdir[jk]==1){
4499: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4500: for(j=1; j <=ncovmodel; jk++,j++){
4501: printf(" p[%d]=%12.7f",jk, p[jk]);
4502: /*q[jjk]=p[jk];*/
4503: }
4504: }else{
4505: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4506: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4507: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4508: /*q[jjk]=p[jk];*/
4509: }
4510: }
4511: printf("\n");
4512: }
4513: fflush(stdout);
4514: }
4515: }
4516: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4517: #else /* FLATSUP */
1.126 brouard 4518: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4519: #endif /* FLATSUP */
4520:
4521: #ifdef LINMINORIGINAL
4522: #else
4523: free_ivector(flatdir,1,npar);
4524: #endif /* LINMINORIGINAL*/
4525: #endif /* POWELL */
1.126 brouard 4526:
1.162 brouard 4527: #ifdef NLOPT
4528: #ifdef NEWUOA
4529: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4530: #else
4531: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4532: #endif
4533: lb=vector(0,npar-1);
4534: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4535: nlopt_set_lower_bounds(opt, lb);
4536: nlopt_set_initial_step1(opt, 0.1);
4537:
4538: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4539: d->function = func;
4540: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4541: nlopt_set_min_objective(opt, myfunc, d);
4542: nlopt_set_xtol_rel(opt, ftol);
4543: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4544: printf("nlopt failed! %d\n",creturn);
4545: }
4546: else {
4547: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4548: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4549: iter=1; /* not equal */
4550: }
4551: nlopt_destroy(opt);
4552: #endif
1.319 brouard 4553: #ifdef FLATSUP
4554: /* npared = npar -flatd/ncovmodel; */
4555: /* xired= matrix(1,npared,1,npared); */
4556: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4557: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4558: /* free_matrix(xire,1,npared,1,npared); */
4559: #else /* FLATSUP */
4560: #endif /* FLATSUP */
1.126 brouard 4561: free_matrix(xi,1,npar,1,npar);
4562: fclose(ficrespow);
1.203 brouard 4563: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4564: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4565: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4566:
4567: }
4568:
4569: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4570: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4571: {
4572: double **a,**y,*x,pd;
1.203 brouard 4573: /* double **hess; */
1.164 brouard 4574: int i, j;
1.126 brouard 4575: int *indx;
4576:
4577: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4578: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4579: void lubksb(double **a, int npar, int *indx, double b[]) ;
4580: void ludcmp(double **a, int npar, int *indx, double *d) ;
4581: double gompertz(double p[]);
1.203 brouard 4582: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4583:
4584: printf("\nCalculation of the hessian matrix. Wait...\n");
4585: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4586: for (i=1;i<=npar;i++){
1.203 brouard 4587: printf("%d-",i);fflush(stdout);
4588: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4589:
4590: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4591:
4592: /* printf(" %f ",p[i]);
4593: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4594: }
4595:
4596: for (i=1;i<=npar;i++) {
4597: for (j=1;j<=npar;j++) {
4598: if (j>i) {
1.203 brouard 4599: printf(".%d-%d",i,j);fflush(stdout);
4600: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4601: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4602:
4603: hess[j][i]=hess[i][j];
4604: /*printf(" %lf ",hess[i][j]);*/
4605: }
4606: }
4607: }
4608: printf("\n");
4609: fprintf(ficlog,"\n");
4610:
4611: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4612: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4613:
4614: a=matrix(1,npar,1,npar);
4615: y=matrix(1,npar,1,npar);
4616: x=vector(1,npar);
4617: indx=ivector(1,npar);
4618: for (i=1;i<=npar;i++)
4619: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4620: ludcmp(a,npar,indx,&pd);
4621:
4622: for (j=1;j<=npar;j++) {
4623: for (i=1;i<=npar;i++) x[i]=0;
4624: x[j]=1;
4625: lubksb(a,npar,indx,x);
4626: for (i=1;i<=npar;i++){
4627: matcov[i][j]=x[i];
4628: }
4629: }
4630:
4631: printf("\n#Hessian matrix#\n");
4632: fprintf(ficlog,"\n#Hessian matrix#\n");
4633: for (i=1;i<=npar;i++) {
4634: for (j=1;j<=npar;j++) {
1.203 brouard 4635: printf("%.6e ",hess[i][j]);
4636: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4637: }
4638: printf("\n");
4639: fprintf(ficlog,"\n");
4640: }
4641:
1.203 brouard 4642: /* printf("\n#Covariance matrix#\n"); */
4643: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4644: /* for (i=1;i<=npar;i++) { */
4645: /* for (j=1;j<=npar;j++) { */
4646: /* printf("%.6e ",matcov[i][j]); */
4647: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4648: /* } */
4649: /* printf("\n"); */
4650: /* fprintf(ficlog,"\n"); */
4651: /* } */
4652:
1.126 brouard 4653: /* Recompute Inverse */
1.203 brouard 4654: /* for (i=1;i<=npar;i++) */
4655: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4656: /* ludcmp(a,npar,indx,&pd); */
4657:
4658: /* printf("\n#Hessian matrix recomputed#\n"); */
4659:
4660: /* for (j=1;j<=npar;j++) { */
4661: /* for (i=1;i<=npar;i++) x[i]=0; */
4662: /* x[j]=1; */
4663: /* lubksb(a,npar,indx,x); */
4664: /* for (i=1;i<=npar;i++){ */
4665: /* y[i][j]=x[i]; */
4666: /* printf("%.3e ",y[i][j]); */
4667: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4668: /* } */
4669: /* printf("\n"); */
4670: /* fprintf(ficlog,"\n"); */
4671: /* } */
4672:
4673: /* Verifying the inverse matrix */
4674: #ifdef DEBUGHESS
4675: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4676:
1.203 brouard 4677: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4678: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4679:
4680: for (j=1;j<=npar;j++) {
4681: for (i=1;i<=npar;i++){
1.203 brouard 4682: printf("%.2f ",y[i][j]);
4683: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4684: }
4685: printf("\n");
4686: fprintf(ficlog,"\n");
4687: }
1.203 brouard 4688: #endif
1.126 brouard 4689:
4690: free_matrix(a,1,npar,1,npar);
4691: free_matrix(y,1,npar,1,npar);
4692: free_vector(x,1,npar);
4693: free_ivector(indx,1,npar);
1.203 brouard 4694: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4695:
4696:
4697: }
4698:
4699: /*************** hessian matrix ****************/
4700: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4701: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4702: int i;
4703: int l=1, lmax=20;
1.203 brouard 4704: double k1,k2, res, fx;
1.132 brouard 4705: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4706: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4707: int k=0,kmax=10;
4708: double l1;
4709:
4710: fx=func(x);
4711: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4712: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4713: l1=pow(10,l);
4714: delts=delt;
4715: for(k=1 ; k <kmax; k=k+1){
4716: delt = delta*(l1*k);
4717: p2[theta]=x[theta] +delt;
1.145 brouard 4718: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4719: p2[theta]=x[theta]-delt;
4720: k2=func(p2)-fx;
4721: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4722: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4723:
1.203 brouard 4724: #ifdef DEBUGHESSII
1.126 brouard 4725: 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);
4726: 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);
4727: #endif
4728: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4729: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4730: k=kmax;
4731: }
4732: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4733: k=kmax; l=lmax*10;
1.126 brouard 4734: }
4735: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4736: delts=delt;
4737: }
1.203 brouard 4738: } /* End loop k */
1.126 brouard 4739: }
4740: delti[theta]=delts;
4741: return res;
4742:
4743: }
4744:
1.203 brouard 4745: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4746: {
4747: int i;
1.164 brouard 4748: int l=1, lmax=20;
1.126 brouard 4749: double k1,k2,k3,k4,res,fx;
1.132 brouard 4750: double p2[MAXPARM+1];
1.203 brouard 4751: int k, kmax=1;
4752: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4753:
4754: int firstime=0;
1.203 brouard 4755:
1.126 brouard 4756: fx=func(x);
1.203 brouard 4757: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4758: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4759: p2[thetai]=x[thetai]+delti[thetai]*k;
4760: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4761: k1=func(p2)-fx;
4762:
1.203 brouard 4763: p2[thetai]=x[thetai]+delti[thetai]*k;
4764: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4765: k2=func(p2)-fx;
4766:
1.203 brouard 4767: p2[thetai]=x[thetai]-delti[thetai]*k;
4768: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4769: k3=func(p2)-fx;
4770:
1.203 brouard 4771: p2[thetai]=x[thetai]-delti[thetai]*k;
4772: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4773: k4=func(p2)-fx;
1.203 brouard 4774: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4775: if(k1*k2*k3*k4 <0.){
1.208 brouard 4776: firstime=1;
1.203 brouard 4777: kmax=kmax+10;
1.208 brouard 4778: }
4779: if(kmax >=10 || firstime ==1){
1.246 brouard 4780: 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);
4781: 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 4782: 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);
4783: 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);
4784: }
4785: #ifdef DEBUGHESSIJ
4786: v1=hess[thetai][thetai];
4787: v2=hess[thetaj][thetaj];
4788: cv12=res;
4789: /* Computing eigen value of Hessian matrix */
4790: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4791: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4792: if ((lc2 <0) || (lc1 <0) ){
4793: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4794: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4795: 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);
4796: 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);
4797: }
1.126 brouard 4798: #endif
4799: }
4800: return res;
4801: }
4802:
1.203 brouard 4803: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4804: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4805: /* { */
4806: /* int i; */
4807: /* int l=1, lmax=20; */
4808: /* double k1,k2,k3,k4,res,fx; */
4809: /* double p2[MAXPARM+1]; */
4810: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4811: /* int k=0,kmax=10; */
4812: /* double l1; */
4813:
4814: /* fx=func(x); */
4815: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4816: /* l1=pow(10,l); */
4817: /* delts=delt; */
4818: /* for(k=1 ; k <kmax; k=k+1){ */
4819: /* delt = delti*(l1*k); */
4820: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4821: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4822: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4823: /* k1=func(p2)-fx; */
4824:
4825: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4826: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4827: /* k2=func(p2)-fx; */
4828:
4829: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4830: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4831: /* k3=func(p2)-fx; */
4832:
4833: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4834: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4835: /* k4=func(p2)-fx; */
4836: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4837: /* #ifdef DEBUGHESSIJ */
4838: /* 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); */
4839: /* 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); */
4840: /* #endif */
4841: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4842: /* k=kmax; */
4843: /* } */
4844: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4845: /* k=kmax; l=lmax*10; */
4846: /* } */
4847: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4848: /* delts=delt; */
4849: /* } */
4850: /* } /\* End loop k *\/ */
4851: /* } */
4852: /* delti[theta]=delts; */
4853: /* return res; */
4854: /* } */
4855:
4856:
1.126 brouard 4857: /************** Inverse of matrix **************/
4858: void ludcmp(double **a, int n, int *indx, double *d)
4859: {
4860: int i,imax,j,k;
4861: double big,dum,sum,temp;
4862: double *vv;
4863:
4864: vv=vector(1,n);
4865: *d=1.0;
4866: for (i=1;i<=n;i++) {
4867: big=0.0;
4868: for (j=1;j<=n;j++)
4869: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4870: if (big == 0.0){
4871: printf(" Singular Hessian matrix at row %d:\n",i);
4872: for (j=1;j<=n;j++) {
4873: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4874: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4875: }
4876: fflush(ficlog);
4877: fclose(ficlog);
4878: nrerror("Singular matrix in routine ludcmp");
4879: }
1.126 brouard 4880: vv[i]=1.0/big;
4881: }
4882: for (j=1;j<=n;j++) {
4883: for (i=1;i<j;i++) {
4884: sum=a[i][j];
4885: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4886: a[i][j]=sum;
4887: }
4888: big=0.0;
4889: for (i=j;i<=n;i++) {
4890: sum=a[i][j];
4891: for (k=1;k<j;k++)
4892: sum -= a[i][k]*a[k][j];
4893: a[i][j]=sum;
4894: if ( (dum=vv[i]*fabs(sum)) >= big) {
4895: big=dum;
4896: imax=i;
4897: }
4898: }
4899: if (j != imax) {
4900: for (k=1;k<=n;k++) {
4901: dum=a[imax][k];
4902: a[imax][k]=a[j][k];
4903: a[j][k]=dum;
4904: }
4905: *d = -(*d);
4906: vv[imax]=vv[j];
4907: }
4908: indx[j]=imax;
4909: if (a[j][j] == 0.0) a[j][j]=TINY;
4910: if (j != n) {
4911: dum=1.0/(a[j][j]);
4912: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4913: }
4914: }
4915: free_vector(vv,1,n); /* Doesn't work */
4916: ;
4917: }
4918:
4919: void lubksb(double **a, int n, int *indx, double b[])
4920: {
4921: int i,ii=0,ip,j;
4922: double sum;
4923:
4924: for (i=1;i<=n;i++) {
4925: ip=indx[i];
4926: sum=b[ip];
4927: b[ip]=b[i];
4928: if (ii)
4929: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4930: else if (sum) ii=i;
4931: b[i]=sum;
4932: }
4933: for (i=n;i>=1;i--) {
4934: sum=b[i];
4935: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4936: b[i]=sum/a[i][i];
4937: }
4938: }
4939:
4940: void pstamp(FILE *fichier)
4941: {
1.196 brouard 4942: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4943: }
4944:
1.297 brouard 4945: void date2dmy(double date,double *day, double *month, double *year){
4946: double yp=0., yp1=0., yp2=0.;
4947:
4948: yp1=modf(date,&yp);/* extracts integral of date in yp and
4949: fractional in yp1 */
4950: *year=yp;
4951: yp2=modf((yp1*12),&yp);
4952: *month=yp;
4953: yp1=modf((yp2*30.5),&yp);
4954: *day=yp;
4955: if(*day==0) *day=1;
4956: if(*month==0) *month=1;
4957: }
4958:
1.253 brouard 4959:
4960:
1.126 brouard 4961: /************ Frequencies ********************/
1.251 brouard 4962: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4963: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4964: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4965: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 4966: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 4967: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4968: int iind=0, iage=0;
4969: int mi; /* Effective wave */
4970: int first;
4971: double ***freq; /* Frequencies */
1.268 brouard 4972: 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 */
4973: 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 4974: double *meanq, *stdq, *idq;
1.226 brouard 4975: double **meanqt;
4976: double *pp, **prop, *posprop, *pospropt;
4977: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4978: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4979: double agebegin, ageend;
4980:
4981: pp=vector(1,nlstate);
1.251 brouard 4982: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4983: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4984: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4985: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4986: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4987: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4988: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4989: meanqt=matrix(1,lastpass,1,nqtveff);
4990: strcpy(fileresp,"P_");
4991: strcat(fileresp,fileresu);
4992: /*strcat(fileresphtm,fileresu);*/
4993: if((ficresp=fopen(fileresp,"w"))==NULL) {
4994: printf("Problem with prevalence resultfile: %s\n", fileresp);
4995: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4996: exit(0);
4997: }
1.240 brouard 4998:
1.226 brouard 4999: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
5000: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
5001: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5002: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5003: fflush(ficlog);
5004: exit(70);
5005: }
5006: else{
5007: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 5008: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5009: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5010: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5011: }
1.319 brouard 5012: 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 5013:
1.226 brouard 5014: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
5015: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
5016: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5017: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5018: fflush(ficlog);
5019: exit(70);
1.240 brouard 5020: } else{
1.226 brouard 5021: 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 5022: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5023: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5024: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5025: }
1.319 brouard 5026: 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 5027:
1.253 brouard 5028: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
5029: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 5030: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5031: j1=0;
1.126 brouard 5032:
1.227 brouard 5033: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 brouard 5034: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 5035: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 5036: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 5037:
5038:
1.226 brouard 5039: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
5040: reference=low_education V1=0,V2=0
5041: med_educ V1=1 V2=0,
5042: high_educ V1=0 V2=1
1.330 brouard 5043: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 5044: */
1.249 brouard 5045: dateintsum=0;
5046: k2cpt=0;
5047:
1.253 brouard 5048: if(cptcoveff == 0 )
1.265 brouard 5049: nl=1; /* Constant and age model only */
1.253 brouard 5050: else
5051: nl=2;
1.265 brouard 5052:
5053: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
5054: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 brouard 5055: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 5056: * freq[s1][s2][iage] =0.
5057: * Loop on iind
5058: * ++freq[s1][s2][iage] weighted
5059: * end iind
5060: * if covariate and j!0
5061: * headers Variable on one line
5062: * endif cov j!=0
5063: * header of frequency table by age
5064: * Loop on age
5065: * pp[s1]+=freq[s1][s2][iage] weighted
5066: * pos+=freq[s1][s2][iage] weighted
5067: * Loop on s1 initial state
5068: * fprintf(ficresp
5069: * end s1
5070: * end age
5071: * if j!=0 computes starting values
5072: * end compute starting values
5073: * end j1
5074: * end nl
5075: */
1.253 brouard 5076: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
5077: if(nj==1)
5078: j=0; /* First pass for the constant */
1.265 brouard 5079: else{
1.335 brouard 5080: 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 5081: }
1.251 brouard 5082: first=1;
1.332 brouard 5083: 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 5084: posproptt=0.;
1.330 brouard 5085: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 5086: scanf("%d", i);*/
5087: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 5088: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 5089: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 5090: freq[i][s2][m]=0;
1.251 brouard 5091:
5092: for (i=1; i<=nlstate; i++) {
1.240 brouard 5093: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 5094: prop[i][m]=0;
5095: posprop[i]=0;
5096: pospropt[i]=0;
5097: }
1.283 brouard 5098: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 5099: idq[z1]=0.;
5100: meanq[z1]=0.;
5101: stdq[z1]=0.;
1.283 brouard 5102: }
5103: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 5104: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 5105: /* meanqt[m][z1]=0.; */
5106: /* } */
5107: /* } */
1.251 brouard 5108: /* dateintsum=0; */
5109: /* k2cpt=0; */
5110:
1.265 brouard 5111: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 5112: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
5113: bool=1;
5114: if(j !=0){
5115: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 brouard 5116: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
5117: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 5118: /* if(Tvaraff[z1] ==-20){ */
5119: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
5120: /* }else if(Tvaraff[z1] ==-10){ */
5121: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 5122: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 brouard 5123: /* 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); */
5124: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
5125: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 5126: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 5127: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 5128: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 5129: /* 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", */
5130: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
5131: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 5132: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
5133: } /* Onlyf fixed */
5134: } /* end z1 */
1.335 brouard 5135: } /* cptcoveff > 0 */
1.251 brouard 5136: } /* end any */
5137: }/* end j==0 */
1.265 brouard 5138: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 5139: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 5140: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 5141: m=mw[mi][iind];
5142: if(j!=0){
5143: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 brouard 5144: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 5145: if( Fixed[Tmodelind[z1]]==1){
5146: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.332 brouard 5147: 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 5148: value is -1, we don't select. It differs from the
5149: constant and age model which counts them. */
5150: bool=0; /* not selected */
5151: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 5152: /* i1=Tvaraff[z1]; */
5153: /* i2=TnsdVar[i1]; */
5154: /* i3=nbcode[i1][i2]; */
5155: /* i4=covar[i1][iind]; */
5156: /* if(i4 != i3){ */
5157: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 5158: bool=0;
5159: }
5160: }
5161: }
5162: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
5163: } /* end j==0 */
5164: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 5165: if(bool==1){ /*Selected */
1.251 brouard 5166: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
5167: and mw[mi+1][iind]. dh depends on stepm. */
5168: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
5169: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
5170: if(m >=firstpass && m <=lastpass){
5171: k2=anint[m][iind]+(mint[m][iind]/12.);
5172: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
5173: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
5174: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
5175: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
5176: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
5177: if (m<lastpass) {
5178: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
5179: /* 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]); */
5180: if(s[m][iind]==-1)
5181: 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.));
5182: 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 5183: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
5184: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 5185: idq[z1]=idq[z1]+weight[iind];
5186: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
5187: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
5188: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 5189: }
1.284 brouard 5190: }
1.251 brouard 5191: /* if((int)agev[m][iind] == 55) */
5192: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5193: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5194: 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 5195: }
1.251 brouard 5196: } /* end if between passes */
5197: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5198: dateintsum=dateintsum+k2; /* on all covariates ?*/
5199: k2cpt++;
5200: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5201: }
1.251 brouard 5202: }else{
5203: bool=1;
5204: }/* end bool 2 */
5205: } /* end m */
1.284 brouard 5206: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5207: /* idq[z1]=idq[z1]+weight[iind]; */
5208: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5209: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5210: /* } */
1.251 brouard 5211: } /* end bool */
5212: } /* end iind = 1 to imx */
1.319 brouard 5213: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5214: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5215:
5216:
5217: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 brouard 5218: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5219: pstamp(ficresp);
1.335 brouard 5220: if (cptcoveff>0 && j!=0){
1.265 brouard 5221: pstamp(ficresp);
1.251 brouard 5222: printf( "\n#********** Variable ");
5223: fprintf(ficresp, "\n#********** Variable ");
5224: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5225: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5226: fprintf(ficlog, "\n#********** Variable ");
1.330 brouard 5227: for (z1=1; z1<=cptcovs; z1++){
1.251 brouard 5228: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 5229: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5230: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5231: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5232: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5233: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5234: }else{
1.330 brouard 5235: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5236: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5237: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5238: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5239: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5240: }
5241: }
5242: printf( "**********\n#");
5243: fprintf(ficresp, "**********\n#");
5244: fprintf(ficresphtm, "**********</h3>\n");
5245: fprintf(ficresphtmfr, "**********</h3>\n");
5246: fprintf(ficlog, "**********\n");
5247: }
1.284 brouard 5248: /*
5249: Printing means of quantitative variables if any
5250: */
5251: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5252: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5253: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5254: if(weightopt==1){
5255: printf(" Weighted mean and standard deviation of");
5256: fprintf(ficlog," Weighted mean and standard deviation of");
5257: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5258: }
1.311 brouard 5259: /* mu = \frac{w x}{\sum w}
5260: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5261: */
5262: 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]));
5263: 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]));
5264: 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 5265: }
5266: /* for (z1=1; z1<= nqtveff; z1++) { */
5267: /* for(m=1;m<=lastpass;m++){ */
5268: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5269: /* } */
5270: /* } */
1.283 brouard 5271:
1.251 brouard 5272: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 brouard 5273: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5274: fprintf(ficresp, " Age");
1.335 brouard 5275: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
5276: 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]]);
5277: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5278: }
1.251 brouard 5279: for(i=1; i<=nlstate;i++) {
1.335 brouard 5280: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5281: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5282: }
1.335 brouard 5283: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5284: fprintf(ficresphtm, "\n");
5285:
5286: /* Header of frequency table by age */
5287: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5288: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5289: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5290: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5291: if(s2!=0 && m!=0)
5292: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5293: }
1.226 brouard 5294: }
1.251 brouard 5295: fprintf(ficresphtmfr, "\n");
5296:
5297: /* For each age */
5298: for(iage=iagemin; iage <= iagemax+3; iage++){
5299: fprintf(ficresphtm,"<tr>");
5300: if(iage==iagemax+1){
5301: fprintf(ficlog,"1");
5302: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5303: }else if(iage==iagemax+2){
5304: fprintf(ficlog,"0");
5305: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5306: }else if(iage==iagemax+3){
5307: fprintf(ficlog,"Total");
5308: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5309: }else{
1.240 brouard 5310: if(first==1){
1.251 brouard 5311: first=0;
5312: printf("See log file for details...\n");
5313: }
5314: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5315: fprintf(ficlog,"Age %d", iage);
5316: }
1.265 brouard 5317: for(s1=1; s1 <=nlstate ; s1++){
5318: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5319: pp[s1] += freq[s1][m][iage];
1.251 brouard 5320: }
1.265 brouard 5321: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5322: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5323: pos += freq[s1][m][iage];
5324: if(pp[s1]>=1.e-10){
1.251 brouard 5325: if(first==1){
1.265 brouard 5326: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5327: }
1.265 brouard 5328: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5329: }else{
5330: if(first==1)
1.265 brouard 5331: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5332: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5333: }
5334: }
5335:
1.265 brouard 5336: for(s1=1; s1 <=nlstate ; s1++){
5337: /* posprop[s1]=0; */
5338: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5339: pp[s1] += freq[s1][m][iage];
5340: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5341:
5342: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5343: pos += pp[s1]; /* pos is the total number of transitions until this age */
5344: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5345: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5346: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5347: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5348: }
5349:
5350: /* Writing ficresp */
1.335 brouard 5351: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5352: if( iage <= iagemax){
5353: fprintf(ficresp," %d",iage);
5354: }
5355: }else if( nj==2){
5356: if( iage <= iagemax){
5357: fprintf(ficresp," %d",iage);
1.335 brouard 5358: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 5359: }
1.240 brouard 5360: }
1.265 brouard 5361: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5362: if(pos>=1.e-5){
1.251 brouard 5363: if(first==1)
1.265 brouard 5364: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5365: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5366: }else{
5367: if(first==1)
1.265 brouard 5368: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5369: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5370: }
5371: if( iage <= iagemax){
5372: if(pos>=1.e-5){
1.335 brouard 5373: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5374: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5375: }else if( nj==2){
5376: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5377: }
5378: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5379: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5380: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5381: } else{
1.335 brouard 5382: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5383: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5384: }
1.240 brouard 5385: }
1.265 brouard 5386: pospropt[s1] +=posprop[s1];
5387: } /* end loop s1 */
1.251 brouard 5388: /* pospropt=0.; */
1.265 brouard 5389: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5390: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5391: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5392: if(first==1){
1.265 brouard 5393: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5394: }
1.265 brouard 5395: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5396: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5397: }
1.265 brouard 5398: if(s1!=0 && m!=0)
5399: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5400: }
1.265 brouard 5401: } /* end loop s1 */
1.251 brouard 5402: posproptt=0.;
1.265 brouard 5403: for(s1=1; s1 <=nlstate; s1++){
5404: posproptt += pospropt[s1];
1.251 brouard 5405: }
5406: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5407: fprintf(ficresphtm,"</tr>\n");
1.335 brouard 5408: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 5409: if(iage <= iagemax)
5410: fprintf(ficresp,"\n");
1.240 brouard 5411: }
1.251 brouard 5412: if(first==1)
5413: printf("Others in log...\n");
5414: fprintf(ficlog,"\n");
5415: } /* end loop age iage */
1.265 brouard 5416:
1.251 brouard 5417: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5418: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5419: if(posproptt < 1.e-5){
1.265 brouard 5420: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5421: }else{
1.265 brouard 5422: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5423: }
1.226 brouard 5424: }
1.251 brouard 5425: fprintf(ficresphtm,"</tr>\n");
5426: fprintf(ficresphtm,"</table>\n");
5427: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5428: if(posproptt < 1.e-5){
1.251 brouard 5429: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5430: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5431: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5432: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5433: invalidvarcomb[j1]=1;
1.226 brouard 5434: }else{
1.251 brouard 5435: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
5436: invalidvarcomb[j1]=0;
1.226 brouard 5437: }
1.251 brouard 5438: fprintf(ficresphtmfr,"</table>\n");
5439: fprintf(ficlog,"\n");
5440: if(j!=0){
5441: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5442: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5443: for(k=1; k <=(nlstate+ndeath); k++){
5444: if (k != i) {
1.265 brouard 5445: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5446: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5447: if(j1==1){ /* All dummy covariates to zero */
5448: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5449: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5450: printf("%d%d ",i,k);
5451: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5452: 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]));
5453: 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]));
5454: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5455: }
1.253 brouard 5456: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5457: for(iage=iagemin; iage <= iagemax+3; iage++){
5458: x[iage]= (double)iage;
5459: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5460: /* 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 5461: }
1.268 brouard 5462: /* Some are not finite, but linreg will ignore these ages */
5463: no=0;
1.253 brouard 5464: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5465: pstart[s1]=b;
5466: pstart[s1-1]=a;
1.252 brouard 5467: }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 */
5468: 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]);
5469: 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 5470: 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 5471: printf("%d%d ",i,k);
5472: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5473: 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 5474: }else{ /* Other cases, like quantitative fixed or varying covariates */
5475: ;
5476: }
5477: /* printf("%12.7f )", param[i][jj][k]); */
5478: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5479: s1++;
1.251 brouard 5480: } /* end jj */
5481: } /* end k!= i */
5482: } /* end k */
1.265 brouard 5483: } /* end i, s1 */
1.251 brouard 5484: } /* end j !=0 */
5485: } /* end selected combination of covariate j1 */
5486: if(j==0){ /* We can estimate starting values from the occurences in each case */
5487: printf("#Freqsummary: Starting values for the constants:\n");
5488: fprintf(ficlog,"\n");
1.265 brouard 5489: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5490: for(k=1; k <=(nlstate+ndeath); k++){
5491: if (k != i) {
5492: printf("%d%d ",i,k);
5493: fprintf(ficlog,"%d%d ",i,k);
5494: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5495: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5496: if(jj==1){ /* Age has to be done */
1.265 brouard 5497: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5498: 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]));
5499: 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 5500: }
5501: /* printf("%12.7f )", param[i][jj][k]); */
5502: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5503: s1++;
1.250 brouard 5504: }
1.251 brouard 5505: printf("\n");
5506: fprintf(ficlog,"\n");
1.250 brouard 5507: }
5508: }
1.284 brouard 5509: } /* end of state i */
1.251 brouard 5510: printf("#Freqsummary\n");
5511: fprintf(ficlog,"\n");
1.265 brouard 5512: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5513: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5514: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5515: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5516: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5517: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5518: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5519: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5520: /* } */
5521: }
1.265 brouard 5522: } /* end loop s1 */
1.251 brouard 5523:
5524: printf("\n");
5525: fprintf(ficlog,"\n");
5526: } /* end j=0 */
1.249 brouard 5527: } /* end j */
1.252 brouard 5528:
1.253 brouard 5529: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5530: for(i=1, jk=1; i <=nlstate; i++){
5531: for(j=1; j <=nlstate+ndeath; j++){
5532: if(j!=i){
5533: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5534: printf("%1d%1d",i,j);
5535: fprintf(ficparo,"%1d%1d",i,j);
5536: for(k=1; k<=ncovmodel;k++){
5537: /* printf(" %lf",param[i][j][k]); */
5538: /* fprintf(ficparo," %lf",param[i][j][k]); */
5539: p[jk]=pstart[jk];
5540: printf(" %f ",pstart[jk]);
5541: fprintf(ficparo," %f ",pstart[jk]);
5542: jk++;
5543: }
5544: printf("\n");
5545: fprintf(ficparo,"\n");
5546: }
5547: }
5548: }
5549: } /* end mle=-2 */
1.226 brouard 5550: dateintmean=dateintsum/k2cpt;
1.296 brouard 5551: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5552:
1.226 brouard 5553: fclose(ficresp);
5554: fclose(ficresphtm);
5555: fclose(ficresphtmfr);
1.283 brouard 5556: free_vector(idq,1,nqfveff);
1.226 brouard 5557: free_vector(meanq,1,nqfveff);
1.284 brouard 5558: free_vector(stdq,1,nqfveff);
1.226 brouard 5559: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5560: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5561: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5562: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5563: free_vector(pospropt,1,nlstate);
5564: free_vector(posprop,1,nlstate);
1.251 brouard 5565: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5566: free_vector(pp,1,nlstate);
5567: /* End of freqsummary */
5568: }
1.126 brouard 5569:
1.268 brouard 5570: /* Simple linear regression */
5571: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5572:
5573: /* y=a+bx regression */
5574: double sumx = 0.0; /* sum of x */
5575: double sumx2 = 0.0; /* sum of x**2 */
5576: double sumxy = 0.0; /* sum of x * y */
5577: double sumy = 0.0; /* sum of y */
5578: double sumy2 = 0.0; /* sum of y**2 */
5579: double sume2 = 0.0; /* sum of square or residuals */
5580: double yhat;
5581:
5582: double denom=0;
5583: int i;
5584: int ne=*no;
5585:
5586: for ( i=ifi, ne=0;i<=ila;i++) {
5587: if(!isfinite(x[i]) || !isfinite(y[i])){
5588: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5589: continue;
5590: }
5591: ne=ne+1;
5592: sumx += x[i];
5593: sumx2 += x[i]*x[i];
5594: sumxy += x[i] * y[i];
5595: sumy += y[i];
5596: sumy2 += y[i]*y[i];
5597: denom = (ne * sumx2 - sumx*sumx);
5598: /* 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); */
5599: }
5600:
5601: denom = (ne * sumx2 - sumx*sumx);
5602: if (denom == 0) {
5603: // vertical, slope m is infinity
5604: *b = INFINITY;
5605: *a = 0;
5606: if (r) *r = 0;
5607: return 1;
5608: }
5609:
5610: *b = (ne * sumxy - sumx * sumy) / denom;
5611: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5612: if (r!=NULL) {
5613: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5614: sqrt((sumx2 - sumx*sumx/ne) *
5615: (sumy2 - sumy*sumy/ne));
5616: }
5617: *no=ne;
5618: for ( i=ifi, ne=0;i<=ila;i++) {
5619: if(!isfinite(x[i]) || !isfinite(y[i])){
5620: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5621: continue;
5622: }
5623: ne=ne+1;
5624: yhat = y[i] - *a -*b* x[i];
5625: sume2 += yhat * yhat ;
5626:
5627: denom = (ne * sumx2 - sumx*sumx);
5628: /* 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); */
5629: }
5630: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5631: *sa= *sb * sqrt(sumx2/ne);
5632:
5633: return 0;
5634: }
5635:
1.126 brouard 5636: /************ Prevalence ********************/
1.227 brouard 5637: 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)
5638: {
5639: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5640: in each health status at the date of interview (if between dateprev1 and dateprev2).
5641: We still use firstpass and lastpass as another selection.
5642: */
1.126 brouard 5643:
1.227 brouard 5644: int i, m, jk, j1, bool, z1,j, iv;
5645: int mi; /* Effective wave */
5646: int iage;
5647: double agebegin, ageend;
5648:
5649: double **prop;
5650: double posprop;
5651: double y2; /* in fractional years */
5652: int iagemin, iagemax;
5653: int first; /** to stop verbosity which is redirected to log file */
5654:
5655: iagemin= (int) agemin;
5656: iagemax= (int) agemax;
5657: /*pp=vector(1,nlstate);*/
1.251 brouard 5658: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5659: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5660: j1=0;
1.222 brouard 5661:
1.227 brouard 5662: /*j=cptcoveff;*/
5663: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5664:
1.288 brouard 5665: first=0;
1.335 brouard 5666: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 5667: for (i=1; i<=nlstate; i++)
1.251 brouard 5668: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5669: prop[i][iage]=0.0;
5670: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5671: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5672: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5673:
5674: for (i=1; i<=imx; i++) { /* Each individual */
5675: bool=1;
5676: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5677: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5678: m=mw[mi][i];
5679: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5680: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5681: for (z1=1; z1<=cptcoveff; z1++){
5682: if( Fixed[Tmodelind[z1]]==1){
5683: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.332 brouard 5684: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 5685: bool=0;
5686: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 5687: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 5688: bool=0;
5689: }
5690: }
5691: if(bool==1){ /* Otherwise we skip that wave/person */
5692: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5693: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5694: if(m >=firstpass && m <=lastpass){
5695: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5696: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5697: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5698: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5699: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5700: 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);
5701: exit(1);
5702: }
5703: if (s[m][i]>0 && s[m][i]<=nlstate) {
5704: /*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]]);*/
5705: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5706: prop[s[m][i]][iagemax+3] += weight[i];
5707: } /* end valid statuses */
5708: } /* end selection of dates */
5709: } /* end selection of waves */
5710: } /* end bool */
5711: } /* end wave */
5712: } /* end individual */
5713: for(i=iagemin; i <= iagemax+3; i++){
5714: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5715: posprop += prop[jk][i];
5716: }
5717:
5718: for(jk=1; jk <=nlstate ; jk++){
5719: if( i <= iagemax){
5720: if(posprop>=1.e-5){
5721: probs[i][jk][j1]= prop[jk][i]/posprop;
5722: } else{
1.288 brouard 5723: if(!first){
5724: first=1;
1.266 brouard 5725: 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]);
5726: }else{
1.288 brouard 5727: 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 5728: }
5729: }
5730: }
5731: }/* end jk */
5732: }/* end i */
1.222 brouard 5733: /*} *//* end i1 */
1.227 brouard 5734: } /* end j1 */
1.222 brouard 5735:
1.227 brouard 5736: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5737: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5738: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5739: } /* End of prevalence */
1.126 brouard 5740:
5741: /************* Waves Concatenation ***************/
5742:
5743: 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)
5744: {
1.298 brouard 5745: /* 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 5746: Death is a valid wave (if date is known).
5747: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5748: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5749: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5750: */
1.126 brouard 5751:
1.224 brouard 5752: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5753: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5754: double sum=0., jmean=0.;*/
1.224 brouard 5755: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5756: int j, k=0,jk, ju, jl;
5757: double sum=0.;
5758: first=0;
1.214 brouard 5759: firstwo=0;
1.217 brouard 5760: firsthree=0;
1.218 brouard 5761: firstfour=0;
1.164 brouard 5762: jmin=100000;
1.126 brouard 5763: jmax=-1;
5764: jmean=0.;
1.224 brouard 5765:
5766: /* Treating live states */
1.214 brouard 5767: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5768: mi=0; /* First valid wave */
1.227 brouard 5769: mli=0; /* Last valid wave */
1.309 brouard 5770: m=firstpass; /* Loop on waves */
5771: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5772: 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 */
5773: mli=m-1;/* mw[++mi][i]=m-1; */
5774: }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 5775: 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 5776: mli=m;
1.224 brouard 5777: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5778: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5779: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5780: }
1.309 brouard 5781: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5782: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5783: break;
1.224 brouard 5784: #else
1.317 brouard 5785: 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 5786: if(firsthree == 0){
1.302 brouard 5787: 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 5788: firsthree=1;
1.317 brouard 5789: }else if(firsthree >=1 && firsthree < 10){
5790: 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);
5791: firsthree++;
5792: }else if(firsthree == 10){
5793: printf("Information, too many Information flags: no more reported to log either\n");
5794: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5795: firsthree++;
5796: }else{
5797: firsthree++;
1.227 brouard 5798: }
1.309 brouard 5799: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5800: mli=m;
5801: }
5802: if(s[m][i]==-2){ /* Vital status is really unknown */
5803: nbwarn++;
1.309 brouard 5804: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5805: 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);
5806: 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);
5807: }
5808: break;
5809: }
5810: break;
1.224 brouard 5811: #endif
1.227 brouard 5812: }/* End m >= lastpass */
1.126 brouard 5813: }/* end while */
1.224 brouard 5814:
1.227 brouard 5815: /* 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 5816: /* After last pass */
1.224 brouard 5817: /* Treating death states */
1.214 brouard 5818: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5819: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5820: /* } */
1.126 brouard 5821: mi++; /* Death is another wave */
5822: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5823: /* Only death is a correct wave */
1.126 brouard 5824: mw[mi][i]=m;
1.257 brouard 5825: } /* else not in a death state */
1.224 brouard 5826: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5827: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5828: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5829: 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 5830: nbwarn++;
5831: if(firstfiv==0){
1.309 brouard 5832: 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 5833: firstfiv=1;
5834: }else{
1.309 brouard 5835: 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 5836: }
1.309 brouard 5837: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5838: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5839: nberr++;
5840: if(firstwo==0){
1.309 brouard 5841: 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 5842: firstwo=1;
5843: }
1.309 brouard 5844: 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 5845: }
1.257 brouard 5846: }else{ /* if date of interview is unknown */
1.227 brouard 5847: /* death is known but not confirmed by death status at any wave */
5848: if(firstfour==0){
1.309 brouard 5849: 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 5850: firstfour=1;
5851: }
1.309 brouard 5852: 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 5853: }
1.224 brouard 5854: } /* end if date of death is known */
5855: #endif
1.309 brouard 5856: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5857: /* wav[i]=mw[mi][i]; */
1.126 brouard 5858: if(mi==0){
5859: nbwarn++;
5860: if(first==0){
1.227 brouard 5861: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5862: first=1;
1.126 brouard 5863: }
5864: if(first==1){
1.227 brouard 5865: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5866: }
5867: } /* end mi==0 */
5868: } /* End individuals */
1.214 brouard 5869: /* wav and mw are no more changed */
1.223 brouard 5870:
1.317 brouard 5871: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
5872: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
5873:
5874:
1.126 brouard 5875: for(i=1; i<=imx; i++){
5876: for(mi=1; mi<wav[i];mi++){
5877: if (stepm <=0)
1.227 brouard 5878: dh[mi][i]=1;
1.126 brouard 5879: else{
1.260 brouard 5880: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5881: if (agedc[i] < 2*AGESUP) {
5882: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5883: if(j==0) j=1; /* Survives at least one month after exam */
5884: else if(j<0){
5885: nberr++;
5886: 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]);
5887: j=1; /* Temporary Dangerous patch */
5888: 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);
5889: 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]);
5890: 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);
5891: }
5892: k=k+1;
5893: if (j >= jmax){
5894: jmax=j;
5895: ijmax=i;
5896: }
5897: if (j <= jmin){
5898: jmin=j;
5899: ijmin=i;
5900: }
5901: sum=sum+j;
5902: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5903: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5904: }
5905: }
5906: else{
5907: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5908: /* 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 5909:
1.227 brouard 5910: k=k+1;
5911: if (j >= jmax) {
5912: jmax=j;
5913: ijmax=i;
5914: }
5915: else if (j <= jmin){
5916: jmin=j;
5917: ijmin=i;
5918: }
5919: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5920: /*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]);*/
5921: if(j<0){
5922: nberr++;
5923: 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]);
5924: 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]);
5925: }
5926: sum=sum+j;
5927: }
5928: jk= j/stepm;
5929: jl= j -jk*stepm;
5930: ju= j -(jk+1)*stepm;
5931: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5932: if(jl==0){
5933: dh[mi][i]=jk;
5934: bh[mi][i]=0;
5935: }else{ /* We want a negative bias in order to only have interpolation ie
5936: * to avoid the price of an extra matrix product in likelihood */
5937: dh[mi][i]=jk+1;
5938: bh[mi][i]=ju;
5939: }
5940: }else{
5941: if(jl <= -ju){
5942: dh[mi][i]=jk;
5943: bh[mi][i]=jl; /* bias is positive if real duration
5944: * is higher than the multiple of stepm and negative otherwise.
5945: */
5946: }
5947: else{
5948: dh[mi][i]=jk+1;
5949: bh[mi][i]=ju;
5950: }
5951: if(dh[mi][i]==0){
5952: dh[mi][i]=1; /* At least one step */
5953: bh[mi][i]=ju; /* At least one step */
5954: /* 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);*/
5955: }
5956: } /* end if mle */
1.126 brouard 5957: }
5958: } /* end wave */
5959: }
5960: jmean=sum/k;
5961: 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 5962: 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 5963: }
1.126 brouard 5964:
5965: /*********** Tricode ****************************/
1.220 brouard 5966: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5967: {
5968: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5969: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5970: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5971: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5972: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5973: */
1.130 brouard 5974:
1.242 brouard 5975: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5976: int modmaxcovj=0; /* Modality max of covariates j */
5977: int cptcode=0; /* Modality max of covariates j */
5978: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5979:
5980:
1.242 brouard 5981: /* cptcoveff=0; */
5982: /* *cptcov=0; */
1.126 brouard 5983:
1.242 brouard 5984: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5985: for (k=1; k <= maxncov; k++)
5986: for(j=1; j<=2; j++)
5987: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5988:
1.242 brouard 5989: /* Loop on covariates without age and products and no quantitative variable */
1.335 brouard 5990: 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 5991: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5992: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5993: switch(Fixed[k]) {
5994: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5995: modmaxcovj=0;
5996: modmincovj=0;
1.242 brouard 5997: 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*/
5998: ij=(int)(covar[Tvar[k]][i]);
5999: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
6000: * If product of Vn*Vm, still boolean *:
6001: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
6002: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
6003: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
6004: modality of the nth covariate of individual i. */
6005: if (ij > modmaxcovj)
6006: modmaxcovj=ij;
6007: else if (ij < modmincovj)
6008: modmincovj=ij;
1.287 brouard 6009: if (ij <0 || ij >1 ){
1.311 brouard 6010: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6011: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6012: fflush(ficlog);
6013: exit(1);
1.287 brouard 6014: }
6015: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 6016: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
6017: exit(1);
6018: }else
6019: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
6020: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
6021: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
6022: /* getting the maximum value of the modality of the covariate
6023: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
6024: female ies 1, then modmaxcovj=1.
6025: */
6026: } /* end for loop on individuals i */
6027: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6028: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6029: cptcode=modmaxcovj;
6030: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
6031: /*for (i=0; i<=cptcode; i++) {*/
6032: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
6033: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6034: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6035: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
6036: if( j != -1){
6037: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
6038: covariate for which somebody answered excluding
6039: undefined. Usually 2: 0 and 1. */
6040: }
6041: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
6042: covariate for which somebody answered including
6043: undefined. Usually 3: -1, 0 and 1. */
6044: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
6045: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
6046: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 6047:
1.242 brouard 6048: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
6049: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
6050: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
6051: /* modmincovj=3; modmaxcovj = 7; */
6052: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
6053: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
6054: /* defining two dummy variables: variables V1_1 and V1_2.*/
6055: /* nbcode[Tvar[j]][ij]=k; */
6056: /* nbcode[Tvar[j]][1]=0; */
6057: /* nbcode[Tvar[j]][2]=1; */
6058: /* nbcode[Tvar[j]][3]=2; */
6059: /* To be continued (not working yet). */
6060: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 6061:
6062: /* 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*/
6063: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
6064: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
6065: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
6066: /*, could be restored in the future */
6067: 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 6068: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
6069: break;
6070: }
6071: ij++;
1.287 brouard 6072: 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 6073: cptcode = ij; /* New max modality for covar j */
6074: } /* end of loop on modality i=-1 to 1 or more */
6075: break;
6076: case 1: /* Testing on varying covariate, could be simple and
6077: * should look at waves or product of fixed *
6078: * varying. No time to test -1, assuming 0 and 1 only */
6079: ij=0;
6080: for(i=0; i<=1;i++){
6081: nbcode[Tvar[k]][++ij]=i;
6082: }
6083: break;
6084: default:
6085: break;
6086: } /* end switch */
6087: } /* end dummy test */
1.334 brouard 6088: if(Dummy[k]==1 && Typevar[k] !=1){ /* Quantitative covariate and not age product */
1.311 brouard 6089: 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 6090: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
6091: printf("Error k=%d \n",k);
6092: exit(1);
6093: }
1.311 brouard 6094: if(isnan(covar[Tvar[k]][i])){
6095: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6096: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6097: fflush(ficlog);
6098: exit(1);
6099: }
6100: }
1.335 brouard 6101: } /* end Quanti */
1.287 brouard 6102: } /* 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 6103:
6104: for (k=-1; k< maxncov; k++) Ndum[k]=0;
6105: /* Look at fixed dummy (single or product) covariates to check empty modalities */
6106: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
6107: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
6108: 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 */
6109: 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 */
6110: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
6111: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
6112:
6113: ij=0;
6114: /* 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 6115: 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 */
6116: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 6117: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
6118: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 brouard 6119: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
6120: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
6121: /* 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 6122: /* If product not in single variable we don't print results */
6123: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 brouard 6124: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
6125: /* k= 1 2 3 4 5 6 7 8 9 */
6126: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
6127: /* ij 1 2 3 */
6128: /* Tvaraff[ij]= 4 3 1 */
6129: /* Tmodelind[ij]=2 3 9 */
6130: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 6131: 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*/
6132: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
6133: 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 */
6134: if(Fixed[k]!=0)
6135: anyvaryingduminmodel=1;
6136: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
6137: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
6138: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
6139: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
6140: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
6141: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
6142: }
6143: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
6144: /* ij--; */
6145: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 brouard 6146: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 6147: * because they can be excluded from the model and real
6148: * if in the model but excluded because missing values, but how to get k from ij?*/
6149: for(j=ij+1; j<= cptcovt; j++){
6150: Tvaraff[j]=0;
6151: Tmodelind[j]=0;
6152: }
6153: for(j=ntveff+1; j<= cptcovt; j++){
6154: TmodelInvind[j]=0;
6155: }
6156: /* To be sorted */
6157: ;
6158: }
1.126 brouard 6159:
1.145 brouard 6160:
1.126 brouard 6161: /*********** Health Expectancies ****************/
6162:
1.235 brouard 6163: 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 6164:
6165: {
6166: /* Health expectancies, no variances */
1.329 brouard 6167: /* cij is the combination in the list of combination of dummy covariates */
6168: /* strstart is a string of time at start of computing */
1.164 brouard 6169: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 6170: int nhstepma, nstepma; /* Decreasing with age */
6171: double age, agelim, hf;
6172: double ***p3mat;
6173: double eip;
6174:
1.238 brouard 6175: /* pstamp(ficreseij); */
1.126 brouard 6176: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
6177: fprintf(ficreseij,"# Age");
6178: for(i=1; i<=nlstate;i++){
6179: for(j=1; j<=nlstate;j++){
6180: fprintf(ficreseij," e%1d%1d ",i,j);
6181: }
6182: fprintf(ficreseij," e%1d. ",i);
6183: }
6184: fprintf(ficreseij,"\n");
6185:
6186:
6187: if(estepm < stepm){
6188: printf ("Problem %d lower than %d\n",estepm, stepm);
6189: }
6190: else hstepm=estepm;
6191: /* We compute the life expectancy from trapezoids spaced every estepm months
6192: * This is mainly to measure the difference between two models: for example
6193: * if stepm=24 months pijx are given only every 2 years and by summing them
6194: * we are calculating an estimate of the Life Expectancy assuming a linear
6195: * progression in between and thus overestimating or underestimating according
6196: * to the curvature of the survival function. If, for the same date, we
6197: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6198: * to compare the new estimate of Life expectancy with the same linear
6199: * hypothesis. A more precise result, taking into account a more precise
6200: * curvature will be obtained if estepm is as small as stepm. */
6201:
6202: /* For example we decided to compute the life expectancy with the smallest unit */
6203: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6204: nhstepm is the number of hstepm from age to agelim
6205: nstepm is the number of stepm from age to agelin.
1.270 brouard 6206: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6207: and note for a fixed period like estepm months */
6208: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6209: survival function given by stepm (the optimization length). Unfortunately it
6210: means that if the survival funtion is printed only each two years of age and if
6211: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6212: results. So we changed our mind and took the option of the best precision.
6213: */
6214: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6215:
6216: agelim=AGESUP;
6217: /* If stepm=6 months */
6218: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6219: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6220:
6221: /* nhstepm age range expressed in number of stepm */
6222: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6223: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6224: /* if (stepm >= YEARM) hstepm=1;*/
6225: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6226: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6227:
6228: for (age=bage; age<=fage; age ++){
6229: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6230: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6231: /* if (stepm >= YEARM) hstepm=1;*/
6232: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6233:
6234: /* If stepm=6 months */
6235: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6236: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 6237: /* 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 6238: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6239:
6240: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6241:
6242: printf("%d|",(int)age);fflush(stdout);
6243: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6244:
6245: /* Computing expectancies */
6246: for(i=1; i<=nlstate;i++)
6247: for(j=1; j<=nlstate;j++)
6248: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6249: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6250:
6251: /* 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]);*/
6252:
6253: }
6254:
6255: fprintf(ficreseij,"%3.0f",age );
6256: for(i=1; i<=nlstate;i++){
6257: eip=0;
6258: for(j=1; j<=nlstate;j++){
6259: eip +=eij[i][j][(int)age];
6260: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6261: }
6262: fprintf(ficreseij,"%9.4f", eip );
6263: }
6264: fprintf(ficreseij,"\n");
6265:
6266: }
6267: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6268: printf("\n");
6269: fprintf(ficlog,"\n");
6270:
6271: }
6272:
1.235 brouard 6273: 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 6274:
6275: {
6276: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6277: to initial status i, ei. .
1.126 brouard 6278: */
1.336 brouard 6279: /* Very time consuming function, but already optimized with precov */
1.126 brouard 6280: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6281: int nhstepma, nstepma; /* Decreasing with age */
6282: double age, agelim, hf;
6283: double ***p3matp, ***p3matm, ***varhe;
6284: double **dnewm,**doldm;
6285: double *xp, *xm;
6286: double **gp, **gm;
6287: double ***gradg, ***trgradg;
6288: int theta;
6289:
6290: double eip, vip;
6291:
6292: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6293: xp=vector(1,npar);
6294: xm=vector(1,npar);
6295: dnewm=matrix(1,nlstate*nlstate,1,npar);
6296: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6297:
6298: pstamp(ficresstdeij);
6299: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6300: fprintf(ficresstdeij,"# Age");
6301: for(i=1; i<=nlstate;i++){
6302: for(j=1; j<=nlstate;j++)
6303: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6304: fprintf(ficresstdeij," e%1d. ",i);
6305: }
6306: fprintf(ficresstdeij,"\n");
6307:
6308: pstamp(ficrescveij);
6309: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6310: fprintf(ficrescveij,"# Age");
6311: for(i=1; i<=nlstate;i++)
6312: for(j=1; j<=nlstate;j++){
6313: cptj= (j-1)*nlstate+i;
6314: for(i2=1; i2<=nlstate;i2++)
6315: for(j2=1; j2<=nlstate;j2++){
6316: cptj2= (j2-1)*nlstate+i2;
6317: if(cptj2 <= cptj)
6318: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6319: }
6320: }
6321: fprintf(ficrescveij,"\n");
6322:
6323: if(estepm < stepm){
6324: printf ("Problem %d lower than %d\n",estepm, stepm);
6325: }
6326: else hstepm=estepm;
6327: /* We compute the life expectancy from trapezoids spaced every estepm months
6328: * This is mainly to measure the difference between two models: for example
6329: * if stepm=24 months pijx are given only every 2 years and by summing them
6330: * we are calculating an estimate of the Life Expectancy assuming a linear
6331: * progression in between and thus overestimating or underestimating according
6332: * to the curvature of the survival function. If, for the same date, we
6333: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6334: * to compare the new estimate of Life expectancy with the same linear
6335: * hypothesis. A more precise result, taking into account a more precise
6336: * curvature will be obtained if estepm is as small as stepm. */
6337:
6338: /* For example we decided to compute the life expectancy with the smallest unit */
6339: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6340: nhstepm is the number of hstepm from age to agelim
6341: nstepm is the number of stepm from age to agelin.
6342: Look at hpijx to understand the reason of that which relies in memory size
6343: and note for a fixed period like estepm months */
6344: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6345: survival function given by stepm (the optimization length). Unfortunately it
6346: means that if the survival funtion is printed only each two years of age and if
6347: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6348: results. So we changed our mind and took the option of the best precision.
6349: */
6350: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6351:
6352: /* If stepm=6 months */
6353: /* nhstepm age range expressed in number of stepm */
6354: agelim=AGESUP;
6355: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6356: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6357: /* if (stepm >= YEARM) hstepm=1;*/
6358: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6359:
6360: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6361: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6362: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6363: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6364: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6365: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6366:
6367: for (age=bage; age<=fage; age ++){
6368: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6369: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6370: /* if (stepm >= YEARM) hstepm=1;*/
6371: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6372:
1.126 brouard 6373: /* If stepm=6 months */
6374: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6375: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6376:
6377: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6378:
1.126 brouard 6379: /* Computing Variances of health expectancies */
6380: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6381: decrease memory allocation */
6382: for(theta=1; theta <=npar; theta++){
6383: for(i=1; i<=npar; i++){
1.222 brouard 6384: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6385: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6386: }
1.235 brouard 6387: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6388: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6389:
1.126 brouard 6390: for(j=1; j<= nlstate; j++){
1.222 brouard 6391: for(i=1; i<=nlstate; i++){
6392: for(h=0; h<=nhstepm-1; h++){
6393: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6394: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6395: }
6396: }
1.126 brouard 6397: }
1.218 brouard 6398:
1.126 brouard 6399: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6400: for(h=0; h<=nhstepm-1; h++){
6401: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6402: }
1.126 brouard 6403: }/* End theta */
6404:
6405:
6406: for(h=0; h<=nhstepm-1; h++)
6407: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6408: for(theta=1; theta <=npar; theta++)
6409: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6410:
1.218 brouard 6411:
1.222 brouard 6412: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6413: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6414: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6415:
1.222 brouard 6416: printf("%d|",(int)age);fflush(stdout);
6417: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6418: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6419: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6420: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6421: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6422: for(ij=1;ij<=nlstate*nlstate;ij++)
6423: for(ji=1;ji<=nlstate*nlstate;ji++)
6424: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6425: }
6426: }
1.320 brouard 6427: /* if((int)age ==50){ */
6428: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6429: /* } */
1.126 brouard 6430: /* Computing expectancies */
1.235 brouard 6431: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6432: for(i=1; i<=nlstate;i++)
6433: for(j=1; j<=nlstate;j++)
1.222 brouard 6434: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6435: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6436:
1.222 brouard 6437: /* 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 6438:
1.222 brouard 6439: }
1.269 brouard 6440:
6441: /* Standard deviation of expectancies ij */
1.126 brouard 6442: fprintf(ficresstdeij,"%3.0f",age );
6443: for(i=1; i<=nlstate;i++){
6444: eip=0.;
6445: vip=0.;
6446: for(j=1; j<=nlstate;j++){
1.222 brouard 6447: eip += eij[i][j][(int)age];
6448: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6449: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6450: 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 6451: }
6452: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6453: }
6454: fprintf(ficresstdeij,"\n");
1.218 brouard 6455:
1.269 brouard 6456: /* Variance of expectancies ij */
1.126 brouard 6457: fprintf(ficrescveij,"%3.0f",age );
6458: for(i=1; i<=nlstate;i++)
6459: for(j=1; j<=nlstate;j++){
1.222 brouard 6460: cptj= (j-1)*nlstate+i;
6461: for(i2=1; i2<=nlstate;i2++)
6462: for(j2=1; j2<=nlstate;j2++){
6463: cptj2= (j2-1)*nlstate+i2;
6464: if(cptj2 <= cptj)
6465: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6466: }
1.126 brouard 6467: }
6468: fprintf(ficrescveij,"\n");
1.218 brouard 6469:
1.126 brouard 6470: }
6471: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6472: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6473: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6474: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6475: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6476: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6477: printf("\n");
6478: fprintf(ficlog,"\n");
1.218 brouard 6479:
1.126 brouard 6480: free_vector(xm,1,npar);
6481: free_vector(xp,1,npar);
6482: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6483: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6484: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6485: }
1.218 brouard 6486:
1.126 brouard 6487: /************ Variance ******************/
1.235 brouard 6488: 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 6489: {
1.279 brouard 6490: /** Variance of health expectancies
6491: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6492: * double **newm;
6493: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6494: */
1.218 brouard 6495:
6496: /* int movingaverage(); */
6497: double **dnewm,**doldm;
6498: double **dnewmp,**doldmp;
6499: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6500: int first=0;
1.218 brouard 6501: int k;
6502: double *xp;
1.279 brouard 6503: double **gp, **gm; /**< for var eij */
6504: double ***gradg, ***trgradg; /**< for var eij */
6505: double **gradgp, **trgradgp; /**< for var p point j */
6506: double *gpp, *gmp; /**< for var p point j */
6507: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6508: double ***p3mat;
6509: double age,agelim, hf;
6510: /* double ***mobaverage; */
6511: int theta;
6512: char digit[4];
6513: char digitp[25];
6514:
6515: char fileresprobmorprev[FILENAMELENGTH];
6516:
6517: if(popbased==1){
6518: if(mobilav!=0)
6519: strcpy(digitp,"-POPULBASED-MOBILAV_");
6520: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6521: }
6522: else
6523: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6524:
1.218 brouard 6525: /* if (mobilav!=0) { */
6526: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6527: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6528: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6529: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6530: /* } */
6531: /* } */
6532:
6533: strcpy(fileresprobmorprev,"PRMORPREV-");
6534: sprintf(digit,"%-d",ij);
6535: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6536: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6537: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6538: strcat(fileresprobmorprev,fileresu);
6539: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6540: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6541: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6542: }
6543: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6544: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6545: pstamp(ficresprobmorprev);
6546: 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 6547: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.337 ! brouard 6548:
! 6549: /* We use TinvDoQresult[nres][resultmodel[nres][j] we sort according to the equation model and the resultline: it is a choice */
! 6550: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ /\* To be done*\/ */
! 6551: /* fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
! 6552: /* } */
! 6553: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */ /* To be done*/
! 6554: fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 6555: }
1.337 ! brouard 6556: /* for(j=1;j<=cptcoveff;j++) */
! 6557: /* fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]); */
1.238 brouard 6558: fprintf(ficresprobmorprev,"\n");
6559:
1.218 brouard 6560: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6561: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6562: fprintf(ficresprobmorprev," p.%-d SE",j);
6563: for(i=1; i<=nlstate;i++)
6564: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6565: }
6566: fprintf(ficresprobmorprev,"\n");
6567:
6568: fprintf(ficgp,"\n# Routine varevsij");
6569: fprintf(ficgp,"\nunset title \n");
6570: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6571: 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");
6572: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6573:
1.218 brouard 6574: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6575: pstamp(ficresvij);
6576: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6577: if(popbased==1)
6578: 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);
6579: else
6580: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6581: fprintf(ficresvij,"# Age");
6582: for(i=1; i<=nlstate;i++)
6583: for(j=1; j<=nlstate;j++)
6584: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6585: fprintf(ficresvij,"\n");
6586:
6587: xp=vector(1,npar);
6588: dnewm=matrix(1,nlstate,1,npar);
6589: doldm=matrix(1,nlstate,1,nlstate);
6590: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6591: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6592:
6593: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6594: gpp=vector(nlstate+1,nlstate+ndeath);
6595: gmp=vector(nlstate+1,nlstate+ndeath);
6596: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6597:
1.218 brouard 6598: if(estepm < stepm){
6599: printf ("Problem %d lower than %d\n",estepm, stepm);
6600: }
6601: else hstepm=estepm;
6602: /* For example we decided to compute the life expectancy with the smallest unit */
6603: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6604: nhstepm is the number of hstepm from age to agelim
6605: nstepm is the number of stepm from age to agelim.
6606: Look at function hpijx to understand why because of memory size limitations,
6607: we decided (b) to get a life expectancy respecting the most precise curvature of the
6608: survival function given by stepm (the optimization length). Unfortunately it
6609: means that if the survival funtion is printed every two years of age and if
6610: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6611: results. So we changed our mind and took the option of the best precision.
6612: */
6613: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6614: agelim = AGESUP;
6615: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6616: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6617: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6618: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6619: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6620: gp=matrix(0,nhstepm,1,nlstate);
6621: gm=matrix(0,nhstepm,1,nlstate);
6622:
6623:
6624: for(theta=1; theta <=npar; theta++){
6625: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6626: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6627: }
1.279 brouard 6628: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6629: * returns into prlim .
1.288 brouard 6630: */
1.242 brouard 6631: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6632:
6633: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6634: if (popbased==1) {
6635: if(mobilav ==0){
6636: for(i=1; i<=nlstate;i++)
6637: prlim[i][i]=probs[(int)age][i][ij];
6638: }else{ /* mobilav */
6639: for(i=1; i<=nlstate;i++)
6640: prlim[i][i]=mobaverage[(int)age][i][ij];
6641: }
6642: }
1.295 brouard 6643: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6644: */
6645: 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 6646: /**< 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 6647: * at horizon h in state j including mortality.
6648: */
1.218 brouard 6649: for(j=1; j<= nlstate; j++){
6650: for(h=0; h<=nhstepm; h++){
6651: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6652: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6653: }
6654: }
1.279 brouard 6655: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6656: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6657: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6658: */
6659: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6660: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6661: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6662: }
6663:
6664: /* Again with minus shift */
1.218 brouard 6665:
6666: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6667: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6668:
1.242 brouard 6669: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6670:
6671: if (popbased==1) {
6672: if(mobilav ==0){
6673: for(i=1; i<=nlstate;i++)
6674: prlim[i][i]=probs[(int)age][i][ij];
6675: }else{ /* mobilav */
6676: for(i=1; i<=nlstate;i++)
6677: prlim[i][i]=mobaverage[(int)age][i][ij];
6678: }
6679: }
6680:
1.235 brouard 6681: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6682:
6683: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6684: for(h=0; h<=nhstepm; h++){
6685: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6686: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6687: }
6688: }
6689: /* This for computing probability of death (h=1 means
6690: computed over hstepm matrices product = hstepm*stepm months)
6691: as a weighted average of prlim.
6692: */
6693: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6694: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6695: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6696: }
1.279 brouard 6697: /* end shifting computations */
6698:
6699: /**< Computing gradient matrix at horizon h
6700: */
1.218 brouard 6701: for(j=1; j<= nlstate; j++) /* vareij */
6702: for(h=0; h<=nhstepm; h++){
6703: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6704: }
1.279 brouard 6705: /**< Gradient of overall mortality p.3 (or p.j)
6706: */
6707: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6708: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6709: }
6710:
6711: } /* End theta */
1.279 brouard 6712:
6713: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6714: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6715:
6716: for(h=0; h<=nhstepm; h++) /* veij */
6717: for(j=1; j<=nlstate;j++)
6718: for(theta=1; theta <=npar; theta++)
6719: trgradg[h][j][theta]=gradg[h][theta][j];
6720:
6721: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6722: for(theta=1; theta <=npar; theta++)
6723: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6724: /**< as well as its transposed matrix
6725: */
1.218 brouard 6726:
6727: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6728: for(i=1;i<=nlstate;i++)
6729: for(j=1;j<=nlstate;j++)
6730: vareij[i][j][(int)age] =0.;
1.279 brouard 6731:
6732: /* Computing trgradg by matcov by gradg at age and summing over h
6733: * and k (nhstepm) formula 15 of article
6734: * Lievre-Brouard-Heathcote
6735: */
6736:
1.218 brouard 6737: for(h=0;h<=nhstepm;h++){
6738: for(k=0;k<=nhstepm;k++){
6739: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6740: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6741: for(i=1;i<=nlstate;i++)
6742: for(j=1;j<=nlstate;j++)
6743: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6744: }
6745: }
6746:
1.279 brouard 6747: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6748: * p.j overall mortality formula 49 but computed directly because
6749: * we compute the grad (wix pijx) instead of grad (pijx),even if
6750: * wix is independent of theta.
6751: */
1.218 brouard 6752: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6753: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6754: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6755: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6756: varppt[j][i]=doldmp[j][i];
6757: /* end ppptj */
6758: /* x centered again */
6759:
1.242 brouard 6760: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6761:
6762: if (popbased==1) {
6763: if(mobilav ==0){
6764: for(i=1; i<=nlstate;i++)
6765: prlim[i][i]=probs[(int)age][i][ij];
6766: }else{ /* mobilav */
6767: for(i=1; i<=nlstate;i++)
6768: prlim[i][i]=mobaverage[(int)age][i][ij];
6769: }
6770: }
6771:
6772: /* This for computing probability of death (h=1 means
6773: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6774: as a weighted average of prlim.
6775: */
1.235 brouard 6776: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6777: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6778: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6779: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6780: }
6781: /* end probability of death */
6782:
6783: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6784: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6785: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6786: for(i=1; i<=nlstate;i++){
6787: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6788: }
6789: }
6790: fprintf(ficresprobmorprev,"\n");
6791:
6792: fprintf(ficresvij,"%.0f ",age );
6793: for(i=1; i<=nlstate;i++)
6794: for(j=1; j<=nlstate;j++){
6795: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6796: }
6797: fprintf(ficresvij,"\n");
6798: free_matrix(gp,0,nhstepm,1,nlstate);
6799: free_matrix(gm,0,nhstepm,1,nlstate);
6800: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6801: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6802: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6803: } /* End age */
6804: free_vector(gpp,nlstate+1,nlstate+ndeath);
6805: free_vector(gmp,nlstate+1,nlstate+ndeath);
6806: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6807: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6808: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6809: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6810: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6811: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6812: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6813: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6814: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6815: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6816: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6817: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6818: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6819: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6820: 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);
6821: /* 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 6822: */
1.218 brouard 6823: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6824: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6825:
1.218 brouard 6826: free_vector(xp,1,npar);
6827: free_matrix(doldm,1,nlstate,1,nlstate);
6828: free_matrix(dnewm,1,nlstate,1,npar);
6829: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6830: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6831: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6832: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6833: fclose(ficresprobmorprev);
6834: fflush(ficgp);
6835: fflush(fichtm);
6836: } /* end varevsij */
1.126 brouard 6837:
6838: /************ Variance of prevlim ******************/
1.269 brouard 6839: 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 6840: {
1.205 brouard 6841: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6842: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6843:
1.268 brouard 6844: double **dnewmpar,**doldm;
1.126 brouard 6845: int i, j, nhstepm, hstepm;
6846: double *xp;
6847: double *gp, *gm;
6848: double **gradg, **trgradg;
1.208 brouard 6849: double **mgm, **mgp;
1.126 brouard 6850: double age,agelim;
6851: int theta;
6852:
6853: pstamp(ficresvpl);
1.288 brouard 6854: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6855: fprintf(ficresvpl,"# Age ");
6856: if(nresult >=1)
6857: fprintf(ficresvpl," Result# ");
1.126 brouard 6858: for(i=1; i<=nlstate;i++)
6859: fprintf(ficresvpl," %1d-%1d",i,i);
6860: fprintf(ficresvpl,"\n");
6861:
6862: xp=vector(1,npar);
1.268 brouard 6863: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6864: doldm=matrix(1,nlstate,1,nlstate);
6865:
6866: hstepm=1*YEARM; /* Every year of age */
6867: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6868: agelim = AGESUP;
6869: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6870: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6871: if (stepm >= YEARM) hstepm=1;
6872: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6873: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6874: mgp=matrix(1,npar,1,nlstate);
6875: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6876: gp=vector(1,nlstate);
6877: gm=vector(1,nlstate);
6878:
6879: for(theta=1; theta <=npar; theta++){
6880: for(i=1; i<=npar; i++){ /* Computes gradient */
6881: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6882: }
1.288 brouard 6883: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6884: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6885: /* else */
6886: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6887: for(i=1;i<=nlstate;i++){
1.126 brouard 6888: gp[i] = prlim[i][i];
1.208 brouard 6889: mgp[theta][i] = prlim[i][i];
6890: }
1.126 brouard 6891: for(i=1; i<=npar; i++) /* Computes gradient */
6892: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6893: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6894: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6895: /* else */
6896: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6897: for(i=1;i<=nlstate;i++){
1.126 brouard 6898: gm[i] = prlim[i][i];
1.208 brouard 6899: mgm[theta][i] = prlim[i][i];
6900: }
1.126 brouard 6901: for(i=1;i<=nlstate;i++)
6902: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6903: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6904: } /* End theta */
6905:
6906: trgradg =matrix(1,nlstate,1,npar);
6907:
6908: for(j=1; j<=nlstate;j++)
6909: for(theta=1; theta <=npar; theta++)
6910: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6911: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6912: /* printf("\nmgm mgp %d ",(int)age); */
6913: /* for(j=1; j<=nlstate;j++){ */
6914: /* printf(" %d ",j); */
6915: /* for(theta=1; theta <=npar; theta++) */
6916: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6917: /* printf("\n "); */
6918: /* } */
6919: /* } */
6920: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6921: /* printf("\n gradg %d ",(int)age); */
6922: /* for(j=1; j<=nlstate;j++){ */
6923: /* printf("%d ",j); */
6924: /* for(theta=1; theta <=npar; theta++) */
6925: /* printf("%d %lf ",theta,gradg[theta][j]); */
6926: /* printf("\n "); */
6927: /* } */
6928: /* } */
1.126 brouard 6929:
6930: for(i=1;i<=nlstate;i++)
6931: varpl[i][(int)age] =0.;
1.209 brouard 6932: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6933: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6934: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6935: }else{
1.268 brouard 6936: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6937: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6938: }
1.126 brouard 6939: for(i=1;i<=nlstate;i++)
6940: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6941:
6942: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6943: if(nresult >=1)
6944: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6945: for(i=1; i<=nlstate;i++){
1.126 brouard 6946: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6947: /* for(j=1;j<=nlstate;j++) */
6948: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6949: }
1.126 brouard 6950: fprintf(ficresvpl,"\n");
6951: free_vector(gp,1,nlstate);
6952: free_vector(gm,1,nlstate);
1.208 brouard 6953: free_matrix(mgm,1,npar,1,nlstate);
6954: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6955: free_matrix(gradg,1,npar,1,nlstate);
6956: free_matrix(trgradg,1,nlstate,1,npar);
6957: } /* End age */
6958:
6959: free_vector(xp,1,npar);
6960: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6961: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6962:
6963: }
6964:
6965:
6966: /************ Variance of backprevalence limit ******************/
1.269 brouard 6967: 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 6968: {
6969: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6970: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6971:
6972: double **dnewmpar,**doldm;
6973: int i, j, nhstepm, hstepm;
6974: double *xp;
6975: double *gp, *gm;
6976: double **gradg, **trgradg;
6977: double **mgm, **mgp;
6978: double age,agelim;
6979: int theta;
6980:
6981: pstamp(ficresvbl);
6982: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6983: fprintf(ficresvbl,"# Age ");
6984: if(nresult >=1)
6985: fprintf(ficresvbl," Result# ");
6986: for(i=1; i<=nlstate;i++)
6987: fprintf(ficresvbl," %1d-%1d",i,i);
6988: fprintf(ficresvbl,"\n");
6989:
6990: xp=vector(1,npar);
6991: dnewmpar=matrix(1,nlstate,1,npar);
6992: doldm=matrix(1,nlstate,1,nlstate);
6993:
6994: hstepm=1*YEARM; /* Every year of age */
6995: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6996: agelim = AGEINF;
6997: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6998: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6999: if (stepm >= YEARM) hstepm=1;
7000: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7001: gradg=matrix(1,npar,1,nlstate);
7002: mgp=matrix(1,npar,1,nlstate);
7003: mgm=matrix(1,npar,1,nlstate);
7004: gp=vector(1,nlstate);
7005: gm=vector(1,nlstate);
7006:
7007: for(theta=1; theta <=npar; theta++){
7008: for(i=1; i<=npar; i++){ /* Computes gradient */
7009: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7010: }
7011: if(mobilavproj > 0 )
7012: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7013: else
7014: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7015: for(i=1;i<=nlstate;i++){
7016: gp[i] = bprlim[i][i];
7017: mgp[theta][i] = bprlim[i][i];
7018: }
7019: for(i=1; i<=npar; i++) /* Computes gradient */
7020: xp[i] = x[i] - (i==theta ?delti[theta]:0);
7021: if(mobilavproj > 0 )
7022: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7023: else
7024: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7025: for(i=1;i<=nlstate;i++){
7026: gm[i] = bprlim[i][i];
7027: mgm[theta][i] = bprlim[i][i];
7028: }
7029: for(i=1;i<=nlstate;i++)
7030: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
7031: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
7032: } /* End theta */
7033:
7034: trgradg =matrix(1,nlstate,1,npar);
7035:
7036: for(j=1; j<=nlstate;j++)
7037: for(theta=1; theta <=npar; theta++)
7038: trgradg[j][theta]=gradg[theta][j];
7039: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7040: /* printf("\nmgm mgp %d ",(int)age); */
7041: /* for(j=1; j<=nlstate;j++){ */
7042: /* printf(" %d ",j); */
7043: /* for(theta=1; theta <=npar; theta++) */
7044: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7045: /* printf("\n "); */
7046: /* } */
7047: /* } */
7048: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7049: /* printf("\n gradg %d ",(int)age); */
7050: /* for(j=1; j<=nlstate;j++){ */
7051: /* printf("%d ",j); */
7052: /* for(theta=1; theta <=npar; theta++) */
7053: /* printf("%d %lf ",theta,gradg[theta][j]); */
7054: /* printf("\n "); */
7055: /* } */
7056: /* } */
7057:
7058: for(i=1;i<=nlstate;i++)
7059: varbpl[i][(int)age] =0.;
7060: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
7061: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7062: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7063: }else{
7064: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7065: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7066: }
7067: for(i=1;i<=nlstate;i++)
7068: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7069:
7070: fprintf(ficresvbl,"%.0f ",age );
7071: if(nresult >=1)
7072: fprintf(ficresvbl,"%d ",nres );
7073: for(i=1; i<=nlstate;i++)
7074: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
7075: fprintf(ficresvbl,"\n");
7076: free_vector(gp,1,nlstate);
7077: free_vector(gm,1,nlstate);
7078: free_matrix(mgm,1,npar,1,nlstate);
7079: free_matrix(mgp,1,npar,1,nlstate);
7080: free_matrix(gradg,1,npar,1,nlstate);
7081: free_matrix(trgradg,1,nlstate,1,npar);
7082: } /* End age */
7083:
7084: free_vector(xp,1,npar);
7085: free_matrix(doldm,1,nlstate,1,npar);
7086: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 7087:
7088: }
7089:
7090: /************ Variance of one-step probabilities ******************/
7091: 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 7092: {
7093: int i, j=0, k1, l1, tj;
7094: int k2, l2, j1, z1;
7095: int k=0, l;
7096: int first=1, first1, first2;
1.326 brouard 7097: int nres=0; /* New */
1.222 brouard 7098: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
7099: double **dnewm,**doldm;
7100: double *xp;
7101: double *gp, *gm;
7102: double **gradg, **trgradg;
7103: double **mu;
7104: double age, cov[NCOVMAX+1];
7105: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
7106: int theta;
7107: char fileresprob[FILENAMELENGTH];
7108: char fileresprobcov[FILENAMELENGTH];
7109: char fileresprobcor[FILENAMELENGTH];
7110: double ***varpij;
7111:
7112: strcpy(fileresprob,"PROB_");
7113: strcat(fileresprob,fileres);
7114: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
7115: printf("Problem with resultfile: %s\n", fileresprob);
7116: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
7117: }
7118: strcpy(fileresprobcov,"PROBCOV_");
7119: strcat(fileresprobcov,fileresu);
7120: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
7121: printf("Problem with resultfile: %s\n", fileresprobcov);
7122: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
7123: }
7124: strcpy(fileresprobcor,"PROBCOR_");
7125: strcat(fileresprobcor,fileresu);
7126: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
7127: printf("Problem with resultfile: %s\n", fileresprobcor);
7128: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
7129: }
7130: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7131: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7132: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7133: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7134: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7135: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7136: pstamp(ficresprob);
7137: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
7138: fprintf(ficresprob,"# Age");
7139: pstamp(ficresprobcov);
7140: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
7141: fprintf(ficresprobcov,"# Age");
7142: pstamp(ficresprobcor);
7143: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
7144: fprintf(ficresprobcor,"# Age");
1.126 brouard 7145:
7146:
1.222 brouard 7147: for(i=1; i<=nlstate;i++)
7148: for(j=1; j<=(nlstate+ndeath);j++){
7149: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
7150: fprintf(ficresprobcov," p%1d-%1d ",i,j);
7151: fprintf(ficresprobcor," p%1d-%1d ",i,j);
7152: }
7153: /* fprintf(ficresprob,"\n");
7154: fprintf(ficresprobcov,"\n");
7155: fprintf(ficresprobcor,"\n");
7156: */
7157: xp=vector(1,npar);
7158: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7159: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7160: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
7161: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
7162: first=1;
7163: fprintf(ficgp,"\n# Routine varprob");
7164: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
7165: fprintf(fichtm,"\n");
7166:
1.288 brouard 7167: 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 7168: 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);
7169: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 7170: and drawn. It helps understanding how is the covariance between two incidences.\
7171: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 7172: 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 7173: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
7174: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
7175: standard deviations wide on each axis. <br>\
7176: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
7177: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
7178: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
7179:
1.222 brouard 7180: cov[1]=1;
7181: /* tj=cptcoveff; */
1.225 brouard 7182: tj = (int) pow(2,cptcoveff);
1.222 brouard 7183: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
7184: j1=0;
1.332 brouard 7185:
7186: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
7187: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.334 brouard 7188: 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 7189: if(tj != 1 && TKresult[nres]!= j1)
7190: continue;
7191:
7192: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
7193: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
7194: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 7195: if (cptcovn>0) {
1.334 brouard 7196: fprintf(ficresprob, "\n#********** Variable ");
7197: fprintf(ficresprobcov, "\n#********** Variable ");
7198: fprintf(ficgp, "\n#********** Variable ");
7199: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
7200: fprintf(ficresprobcor, "\n#********** Variable ");
7201:
7202: /* Including quantitative variables of the resultline to be done */
7203: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
7204: printf("Varprob modelresult[%d][%d]=%d model=%s \n",nres, z1, modelresult[nres][z1], model);
7205: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=%s \n",nres, z1, modelresult[nres][z1], model);
7206: /* fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=%s resultline[%d]=%s \n",nres, z1, modelresult[nres][z1], model, nres, resultline[nres]); */
7207: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
7208: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
7209: 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 */
7210: 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 */
7211: 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 */
7212: 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 */
7213: 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 */
7214: fprintf(ficresprob,"fixed ");
7215: fprintf(ficresprobcov,"fixed ");
7216: fprintf(ficgp,"fixed ");
7217: fprintf(fichtmcov,"fixed ");
7218: fprintf(ficresprobcor,"fixed ");
7219: }else{
7220: fprintf(ficresprob,"varyi ");
7221: fprintf(ficresprobcov,"varyi ");
7222: fprintf(ficgp,"varyi ");
7223: fprintf(fichtmcov,"varyi ");
7224: fprintf(ficresprobcor,"varyi ");
7225: }
7226: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
7227: /* For each selected (single) quantitative value */
1.337 ! brouard 7228: fprintf(ficresprob," V%d=%lg ",Tvqresult[nres][z1],Tqresult[nres][z1]);
1.334 brouard 7229: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
7230: fprintf(ficresprob,"fixed ");
7231: fprintf(ficresprobcov,"fixed ");
7232: fprintf(ficgp,"fixed ");
7233: fprintf(fichtmcov,"fixed ");
7234: fprintf(ficresprobcor,"fixed ");
7235: }else{
7236: fprintf(ficresprob,"varyi ");
7237: fprintf(ficresprobcov,"varyi ");
7238: fprintf(ficgp,"varyi ");
7239: fprintf(fichtmcov,"varyi ");
7240: fprintf(ficresprobcor,"varyi ");
7241: }
7242: }else{
7243: 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 */
7244: 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 */
7245: exit(1);
7246: }
7247: } /* End loop on variable of this resultline */
7248: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 7249: fprintf(ficresprob, "**********\n#\n");
7250: fprintf(ficresprobcov, "**********\n#\n");
7251: fprintf(ficgp, "**********\n#\n");
7252: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
7253: fprintf(ficresprobcor, "**********\n#");
7254: if(invalidvarcomb[j1]){
7255: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7256: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7257: continue;
7258: }
7259: }
7260: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7261: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7262: gp=vector(1,(nlstate)*(nlstate+ndeath));
7263: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 7264: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 7265: cov[2]=age;
7266: if(nagesqr==1)
7267: cov[3]= age*age;
1.334 brouard 7268: /* New code end of combination but for each resultline */
7269: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
7270: if(Typevar[k1]==1){ /* A product with age */
7271: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 7272: }else{
1.334 brouard 7273: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 7274: }
1.334 brouard 7275: }/* End of loop on model equation */
7276: /* Old code */
7277: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
7278: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
7279: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
7280: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
7281: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
7282: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
7283: /* * 1 1 1 1 1 */
7284: /* * 2 2 1 1 1 */
7285: /* * 3 1 2 1 1 */
7286: /* *\/ */
7287: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
7288: /* } */
7289: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
7290: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
7291: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
7292: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
7293: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
7294: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
7295: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7296: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
7297: /* 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]); */
7298: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
7299: /* /\* exit(1); *\/ */
7300: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
7301: /* } */
7302: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7303: /* } */
7304: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
7305: /* if(Dummy[Tvard[k][1]]==0){ */
7306: /* if(Dummy[Tvard[k][2]]==0){ */
7307: /* 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]])]; */
7308: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7309: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
7310: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
7311: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
7312: /* } */
7313: /* }else{ */
7314: /* if(Dummy[Tvard[k][2]]==0){ */
7315: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
7316: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
7317: /* }else{ */
7318: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
7319: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
7320: /* } */
7321: /* } */
7322: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7323: /* } */
1.326 brouard 7324: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7325: for(theta=1; theta <=npar; theta++){
7326: for(i=1; i<=npar; i++)
7327: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7328:
1.222 brouard 7329: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7330:
1.222 brouard 7331: k=0;
7332: for(i=1; i<= (nlstate); i++){
7333: for(j=1; j<=(nlstate+ndeath);j++){
7334: k=k+1;
7335: gp[k]=pmmij[i][j];
7336: }
7337: }
1.220 brouard 7338:
1.222 brouard 7339: for(i=1; i<=npar; i++)
7340: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7341:
1.222 brouard 7342: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7343: k=0;
7344: for(i=1; i<=(nlstate); i++){
7345: for(j=1; j<=(nlstate+ndeath);j++){
7346: k=k+1;
7347: gm[k]=pmmij[i][j];
7348: }
7349: }
1.220 brouard 7350:
1.222 brouard 7351: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7352: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7353: }
1.126 brouard 7354:
1.222 brouard 7355: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7356: for(theta=1; theta <=npar; theta++)
7357: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7358:
1.222 brouard 7359: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7360: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7361:
1.222 brouard 7362: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7363:
1.222 brouard 7364: k=0;
7365: for(i=1; i<=(nlstate); i++){
7366: for(j=1; j<=(nlstate+ndeath);j++){
7367: k=k+1;
7368: mu[k][(int) age]=pmmij[i][j];
7369: }
7370: }
7371: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7372: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7373: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7374:
1.222 brouard 7375: /*printf("\n%d ",(int)age);
7376: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7377: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7378: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7379: }*/
1.220 brouard 7380:
1.222 brouard 7381: fprintf(ficresprob,"\n%d ",(int)age);
7382: fprintf(ficresprobcov,"\n%d ",(int)age);
7383: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7384:
1.222 brouard 7385: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7386: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7387: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7388: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7389: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7390: }
7391: i=0;
7392: for (k=1; k<=(nlstate);k++){
7393: for (l=1; l<=(nlstate+ndeath);l++){
7394: i++;
7395: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7396: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7397: for (j=1; j<=i;j++){
7398: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7399: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7400: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7401: }
7402: }
7403: }/* end of loop for state */
7404: } /* end of loop for age */
7405: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7406: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7407: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7408: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7409:
7410: /* Confidence intervalle of pij */
7411: /*
7412: fprintf(ficgp,"\nunset parametric;unset label");
7413: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7414: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7415: 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);
7416: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7417: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7418: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7419: */
7420:
7421: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7422: first1=1;first2=2;
7423: for (k2=1; k2<=(nlstate);k2++){
7424: for (l2=1; l2<=(nlstate+ndeath);l2++){
7425: if(l2==k2) continue;
7426: j=(k2-1)*(nlstate+ndeath)+l2;
7427: for (k1=1; k1<=(nlstate);k1++){
7428: for (l1=1; l1<=(nlstate+ndeath);l1++){
7429: if(l1==k1) continue;
7430: i=(k1-1)*(nlstate+ndeath)+l1;
7431: if(i<=j) continue;
7432: for (age=bage; age<=fage; age ++){
7433: if ((int)age %5==0){
7434: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7435: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7436: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7437: mu1=mu[i][(int) age]/stepm*YEARM ;
7438: mu2=mu[j][(int) age]/stepm*YEARM;
7439: c12=cv12/sqrt(v1*v2);
7440: /* Computing eigen value of matrix of covariance */
7441: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7442: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7443: if ((lc2 <0) || (lc1 <0) ){
7444: if(first2==1){
7445: first1=0;
7446: 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);
7447: }
7448: 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);
7449: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7450: /* lc2=fabs(lc2); */
7451: }
1.220 brouard 7452:
1.222 brouard 7453: /* Eigen vectors */
1.280 brouard 7454: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7455: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7456: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7457: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7458: }else
7459: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7460: /*v21=sqrt(1.-v11*v11); *//* error */
7461: v21=(lc1-v1)/cv12*v11;
7462: v12=-v21;
7463: v22=v11;
7464: tnalp=v21/v11;
7465: if(first1==1){
7466: first1=0;
7467: 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);
7468: }
7469: 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);
7470: /*printf(fignu*/
7471: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7472: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7473: if(first==1){
7474: first=0;
7475: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7476: fprintf(ficgp,"\nset parametric;unset label");
7477: 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);
7478: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7479: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7480: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7481: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7482: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7483: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7484: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7485: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7486: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7487: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7488: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7489: 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 7490: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7491: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7492: }else{
7493: first=0;
7494: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7495: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7496: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7497: 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 7498: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7499: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7500: }/* if first */
7501: } /* age mod 5 */
7502: } /* end loop age */
7503: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7504: first=1;
7505: } /*l12 */
7506: } /* k12 */
7507: } /*l1 */
7508: }/* k1 */
1.332 brouard 7509: } /* loop on combination of covariates j1 */
1.326 brouard 7510: } /* loop on nres */
1.222 brouard 7511: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7512: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7513: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7514: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7515: free_vector(xp,1,npar);
7516: fclose(ficresprob);
7517: fclose(ficresprobcov);
7518: fclose(ficresprobcor);
7519: fflush(ficgp);
7520: fflush(fichtmcov);
7521: }
1.126 brouard 7522:
7523:
7524: /******************* Printing html file ***********/
1.201 brouard 7525: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7526: int lastpass, int stepm, int weightopt, char model[],\
7527: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7528: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7529: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7530: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7531: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7532: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7533: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7534: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7535: </ul>");
1.319 brouard 7536: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7537: /* </ul>", model); */
1.214 brouard 7538: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7539: 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",
7540: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 7541: 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 7542: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7543: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7544: fprintf(fichtm,"\
7545: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7546: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7547: fprintf(fichtm,"\
1.217 brouard 7548: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7549: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7550: fprintf(fichtm,"\
1.288 brouard 7551: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7552: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7553: fprintf(fichtm,"\
1.288 brouard 7554: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7555: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7556: fprintf(fichtm,"\
1.211 brouard 7557: - (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 7558: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7559: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7560: if(prevfcast==1){
7561: fprintf(fichtm,"\
7562: - Prevalence projections by age and states: \
1.201 brouard 7563: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7564: }
1.126 brouard 7565:
7566:
1.225 brouard 7567: m=pow(2,cptcoveff);
1.222 brouard 7568: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7569:
1.317 brouard 7570: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7571:
7572: jj1=0;
7573:
7574: fprintf(fichtm," \n<ul>");
1.337 ! brouard 7575: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
! 7576: /* k1=nres; */
! 7577: k1= TKresult[nres];
! 7578: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
! 7579: /* if(m != 1 && TKresult[nres]!= k1) */
! 7580: /* continue; */
1.264 brouard 7581: jj1++;
7582: if (cptcovn > 0) {
7583: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
1.337 ! brouard 7584: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
! 7585: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7586: }
1.337 ! brouard 7587: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
! 7588: /* fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
! 7589: /* } */
! 7590: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 7591: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 7592: /* } */
1.264 brouard 7593: fprintf(fichtm,"\">");
7594:
7595: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7596: fprintf(fichtm,"************ Results for covariates");
1.337 ! brouard 7597: for (cpt=1; cpt<=cptcovs;cpt++){
! 7598: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7599: }
1.337 ! brouard 7600: /* fprintf(fichtm,"************ Results for covariates"); */
! 7601: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
! 7602: /* fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
! 7603: /* } */
! 7604: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 7605: /* fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 7606: /* } */
1.264 brouard 7607: if(invalidvarcomb[k1]){
7608: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7609: continue;
7610: }
7611: fprintf(fichtm,"</a></li>");
7612: } /* cptcovn >0 */
7613: }
1.317 brouard 7614: fprintf(fichtm," \n</ul>");
1.264 brouard 7615:
1.222 brouard 7616: jj1=0;
1.237 brouard 7617:
1.337 ! brouard 7618: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
! 7619: /* k1=nres; */
! 7620: k1= TKresult[nres];
! 7621: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
! 7622: /* if(m != 1 && TKresult[nres]!= k1) */
! 7623: /* continue; */
1.220 brouard 7624:
1.222 brouard 7625: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7626: jj1++;
7627: if (cptcovn > 0) {
1.264 brouard 7628: fprintf(fichtm,"\n<p><a name=\"rescov");
1.337 ! brouard 7629: for (cpt=1; cpt<=cptcovs;cpt++){
! 7630: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7631: }
1.337 ! brouard 7632: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 7633: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 7634: /* } */
1.264 brouard 7635: fprintf(fichtm,"\"</a>");
7636:
1.222 brouard 7637: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 ! brouard 7638: for (cpt=1; cpt<=cptcovs;cpt++){
! 7639: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
! 7640: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 7641: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7642: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7643: }
1.230 brouard 7644: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.321 brouard 7645: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7646: if(invalidvarcomb[k1]){
7647: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7648: printf("\nCombination (%d) ignored because no cases \n",k1);
7649: continue;
7650: }
7651: }
7652: /* aij, bij */
1.259 brouard 7653: 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 7654: <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 7655: /* Pij */
1.241 brouard 7656: 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> \
7657: <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 7658: /* Quasi-incidences */
7659: 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 7660: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7661: 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 7662: 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> \
7663: <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 7664: /* Survival functions (period) in state j */
7665: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7666: 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);
7667: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7668: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7669: }
7670: /* State specific survival functions (period) */
7671: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7672: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7673: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7674: <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);
7675: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7676: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7677: }
1.288 brouard 7678: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7679: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7680: 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);
7681: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"P_"),subdirf2(optionfilefiname,"P_"));
7682: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7683: }
1.296 brouard 7684: if(prevbcast==1){
1.288 brouard 7685: /* Backward prevalence in each health state */
1.222 brouard 7686: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7687: 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 7688: <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 7689: }
1.217 brouard 7690: }
1.222 brouard 7691: if(prevfcast==1){
1.288 brouard 7692: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7693: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7694: 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);
7695: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7696: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7697: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7698: }
7699: }
1.296 brouard 7700: if(prevbcast==1){
1.268 brouard 7701: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7702: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7703: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7704: 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 \
7705: 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 7706: 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);
7707: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7708: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7709: }
7710: }
1.220 brouard 7711:
1.222 brouard 7712: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7713: 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);
7714: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7715: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7716: }
7717: /* } /\* end i1 *\/ */
1.337 ! brouard 7718: }/* End k1=nres */
1.222 brouard 7719: fprintf(fichtm,"</ul>");
1.126 brouard 7720:
1.222 brouard 7721: fprintf(fichtm,"\
1.126 brouard 7722: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7723: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7724: - 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 7725: But because parameters are usually highly correlated (a higher incidence of disability \
7726: and a higher incidence of recovery can give very close observed transition) it might \
7727: be very useful to look not only at linear confidence intervals estimated from the \
7728: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7729: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7730: covariance matrix of the one-step probabilities. \
7731: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7732:
1.222 brouard 7733: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7734: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7735: fprintf(fichtm,"\
1.126 brouard 7736: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7737: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7738:
1.222 brouard 7739: fprintf(fichtm,"\
1.126 brouard 7740: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7741: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7742: fprintf(fichtm,"\
1.126 brouard 7743: - 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): \
7744: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7745: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7746: fprintf(fichtm,"\
1.126 brouard 7747: - (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): \
7748: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7749: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7750: fprintf(fichtm,"\
1.288 brouard 7751: - 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 7752: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7753: fprintf(fichtm,"\
1.128 brouard 7754: - 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 7755: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7756: fprintf(fichtm,"\
1.288 brouard 7757: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7758: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7759:
7760: /* if(popforecast==1) fprintf(fichtm,"\n */
7761: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7762: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7763: /* <br>",fileres,fileres,fileres,fileres); */
7764: /* else */
7765: /* 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 7766: fflush(fichtm);
1.126 brouard 7767:
1.225 brouard 7768: m=pow(2,cptcoveff);
1.222 brouard 7769: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7770:
1.317 brouard 7771: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7772:
7773: jj1=0;
7774:
7775: fprintf(fichtm," \n<ul>");
1.337 ! brouard 7776: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
! 7777: /* k1=nres; */
! 7778: k1= TKresult[nres];
! 7779: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
! 7780: /* if(m != 1 && TKresult[nres]!= k1) */
! 7781: /* continue; */
1.317 brouard 7782: jj1++;
7783: if (cptcovn > 0) {
7784: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
1.337 ! brouard 7785: for (cpt=1; cpt<=cptcovs;cpt++){
! 7786: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 7787: }
7788: fprintf(fichtm,"\">");
7789:
7790: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7791: fprintf(fichtm,"************ Results for covariates");
1.337 ! brouard 7792: for (cpt=1; cpt<=cptcovs;cpt++){
! 7793: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 7794: }
7795: if(invalidvarcomb[k1]){
7796: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7797: continue;
7798: }
7799: fprintf(fichtm,"</a></li>");
7800: } /* cptcovn >0 */
1.337 ! brouard 7801: } /* End nres */
1.317 brouard 7802: fprintf(fichtm," \n</ul>");
7803:
1.222 brouard 7804: jj1=0;
1.237 brouard 7805:
1.241 brouard 7806: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 7807: /* k1=nres; */
! 7808: k1= TKresult[nres];
! 7809: /* for(k1=1; k1<=m;k1++){ */
! 7810: /* if(m != 1 && TKresult[nres]!= k1) */
! 7811: /* continue; */
1.222 brouard 7812: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7813: jj1++;
1.126 brouard 7814: if (cptcovn > 0) {
1.317 brouard 7815: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
1.337 ! brouard 7816: for (cpt=1; cpt<=cptcovs;cpt++){
! 7817: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 7818: }
7819: fprintf(fichtm,"\"</a>");
7820:
1.126 brouard 7821: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 ! brouard 7822: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcoveff number of variables */
! 7823: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
! 7824: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 7825: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 7826: }
1.237 brouard 7827:
1.321 brouard 7828: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 7829:
1.222 brouard 7830: if(invalidvarcomb[k1]){
7831: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7832: continue;
7833: }
1.337 ! brouard 7834: } /* If cptcovn >0 */
1.126 brouard 7835: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7836: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7837: 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);
7838: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7839: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7840: }
7841: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7842: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7843: true period expectancies (those weighted with period prevalences are also\
7844: drawn in addition to the population based expectancies computed using\
1.314 brouard 7845: 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);
7846: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7847: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7848: /* } /\* end i1 *\/ */
1.241 brouard 7849: }/* End nres */
1.222 brouard 7850: fprintf(fichtm,"</ul>");
7851: fflush(fichtm);
1.126 brouard 7852: }
7853:
7854: /******************* Gnuplot file **************/
1.296 brouard 7855: 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 7856:
7857: char dirfileres[132],optfileres[132];
1.264 brouard 7858: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7859: 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 7860: int lv=0, vlv=0, kl=0;
1.130 brouard 7861: int ng=0;
1.201 brouard 7862: int vpopbased;
1.223 brouard 7863: int ioffset; /* variable offset for columns */
1.270 brouard 7864: int iyearc=1; /* variable column for year of projection */
7865: int iagec=1; /* variable column for age of projection */
1.235 brouard 7866: int nres=0; /* Index of resultline */
1.266 brouard 7867: int istart=1; /* For starting graphs in projections */
1.219 brouard 7868:
1.126 brouard 7869: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7870: /* printf("Problem with file %s",optionfilegnuplot); */
7871: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7872: /* } */
7873:
7874: /*#ifdef windows */
7875: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7876: /*#endif */
1.225 brouard 7877: m=pow(2,cptcoveff);
1.126 brouard 7878:
1.274 brouard 7879: /* diagram of the model */
7880: fprintf(ficgp,"\n#Diagram of the model \n");
7881: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7882: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7883: 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);
7884:
7885: 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);
7886: fprintf(ficgp,"\n#show arrow\nunset label\n");
7887: 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);
7888: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7889: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7890: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7891: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7892:
1.202 brouard 7893: /* Contribution to likelihood */
7894: /* Plot the probability implied in the likelihood */
1.223 brouard 7895: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7896: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7897: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7898: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7899: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7900: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7901: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7902: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7903: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7904: 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));
7905: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7906: 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));
7907: for (i=1; i<= nlstate ; i ++) {
7908: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7909: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7910: 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);
7911: for (j=2; j<= nlstate+ndeath ; j ++) {
7912: 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);
7913: }
7914: fprintf(ficgp,";\nset out; unset ylabel;\n");
7915: }
7916: /* 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 */
7917: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7918: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7919: fprintf(ficgp,"\nset out;unset log\n");
7920: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7921:
1.126 brouard 7922: strcpy(dirfileres,optionfilefiname);
7923: strcpy(optfileres,"vpl");
1.223 brouard 7924: /* 1eme*/
1.238 brouard 7925: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
1.337 ! brouard 7926: /* for (k1=1; k1<= m ; k1 ++){ /\* For each valid combination of covariate *\/ */
1.236 brouard 7927: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 7928: k1=TKresult[nres];
1.238 brouard 7929: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.337 ! brouard 7930: /* if(m != 1 && TKresult[nres]!= k1) */
! 7931: /* continue; */
1.238 brouard 7932: /* We are interested in selected combination by the resultline */
1.246 brouard 7933: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7934: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7935: strcpy(gplotlabel,"(");
1.337 ! brouard 7936: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
! 7937: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 7938: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 7939:
! 7940: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate k get corresponding value lv for combination k1 *\/ */
! 7941: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the value of the covariate corresponding to k1 combination *\\/ *\/ */
! 7942: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
! 7943: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 7944: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 7945: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 7946: /* vlv= nbcode[Tvaraff[k]][lv]; /\* vlv is the value of the covariate lv, 0 or 1 *\/ */
! 7947: /* /\* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv *\/ */
! 7948: /* /\* printf(" V%d=%d ",Tvaraff[k],vlv); *\/ */
! 7949: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 7950: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 7951: /* } */
! 7952: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 7953: /* /\* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); *\/ */
! 7954: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 7955: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.264 brouard 7956: }
7957: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7958: /* printf("\n#\n"); */
1.238 brouard 7959: fprintf(ficgp,"\n#\n");
7960: if(invalidvarcomb[k1]){
1.260 brouard 7961: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7962: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7963: continue;
7964: }
1.235 brouard 7965:
1.241 brouard 7966: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7967: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7968: /* 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 7969: fprintf(ficgp,"set title \"Alive state %d %s model=%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 7970: 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);
7971: /* 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); */
7972: /* k1-1 error should be nres-1*/
1.238 brouard 7973: for (i=1; i<= nlstate ; i ++) {
7974: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7975: else fprintf(ficgp," %%*lf (%%*lf)");
7976: }
1.288 brouard 7977: 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 7978: for (i=1; i<= nlstate ; i ++) {
7979: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7980: else fprintf(ficgp," %%*lf (%%*lf)");
7981: }
1.260 brouard 7982: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7983: for (i=1; i<= nlstate ; i ++) {
7984: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7985: else fprintf(ficgp," %%*lf (%%*lf)");
7986: }
1.265 brouard 7987: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7988:
7989: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7990: if(cptcoveff ==0){
1.271 brouard 7991: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7992: }else{
7993: kl=0;
7994: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 7995: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
7996: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 7997: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7998: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7999: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8000: vlv= nbcode[Tvaraff[k]][lv];
8001: kl++;
8002: /* 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 *\/ */
8003: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8004: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8005: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8006: if(k==cptcoveff){
8007: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
8008: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
8009: }else{
8010: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
8011: kl++;
8012: }
8013: } /* end covariate */
8014: } /* end if no covariate */
8015:
1.296 brouard 8016: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 8017: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 8018: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 8019: if(cptcoveff ==0){
1.245 brouard 8020: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 8021: }else{
8022: kl=0;
8023: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8024: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8025: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 8026: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8027: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8028: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8029: /* vlv= nbcode[Tvaraff[k]][lv]; */
8030: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 8031: kl++;
1.238 brouard 8032: /* 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 *\/ */
8033: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8034: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8035: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8036: if(k==cptcoveff){
1.245 brouard 8037: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 8038: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 8039: }else{
1.332 brouard 8040: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 8041: kl++;
8042: }
8043: } /* end covariate */
8044: } /* end if no covariate */
1.296 brouard 8045: if(prevbcast == 1){
1.268 brouard 8046: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
8047: /* k1-1 error should be nres-1*/
8048: for (i=1; i<= nlstate ; i ++) {
8049: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8050: else fprintf(ficgp," %%*lf (%%*lf)");
8051: }
1.271 brouard 8052: 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 8053: for (i=1; i<= nlstate ; i ++) {
8054: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8055: else fprintf(ficgp," %%*lf (%%*lf)");
8056: }
1.276 brouard 8057: 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 8058: for (i=1; i<= nlstate ; i ++) {
8059: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8060: else fprintf(ficgp," %%*lf (%%*lf)");
8061: }
1.274 brouard 8062: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 8063: } /* end if backprojcast */
1.296 brouard 8064: } /* end if prevbcast */
1.276 brouard 8065: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
8066: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 8067: } /* nres */
1.337 ! brouard 8068: /* } /\* k1 *\/ */
1.201 brouard 8069: } /* cpt */
1.235 brouard 8070:
8071:
1.126 brouard 8072: /*2 eme*/
1.337 ! brouard 8073: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8074: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 8075: k1=TKresult[nres];
! 8076: /* if(m != 1 && TKresult[nres]!= k1) */
! 8077: /* continue; */
1.238 brouard 8078: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 8079: strcpy(gplotlabel,"(");
1.337 ! brouard 8080: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
! 8081: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8082: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8083: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
! 8084: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
! 8085: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
! 8086: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 8087: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 8088: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 8089: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
! 8090: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8091: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 8092: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 8093: /* } */
! 8094: /* /\* for(k=1; k <= ncovds; k++){ *\/ */
! 8095: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 8096: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 8097: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 8098: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 8099: }
1.264 brouard 8100: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8101: fprintf(ficgp,"\n#\n");
1.223 brouard 8102: if(invalidvarcomb[k1]){
8103: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8104: continue;
8105: }
1.219 brouard 8106:
1.241 brouard 8107: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 8108: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 8109: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
8110: if(vpopbased==0){
1.238 brouard 8111: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 8112: }else
1.238 brouard 8113: fprintf(ficgp,"\nreplot ");
8114: for (i=1; i<= nlstate+1 ; i ++) {
8115: k=2*i;
1.261 brouard 8116: 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 8117: for (j=1; j<= nlstate+1 ; j ++) {
8118: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8119: else fprintf(ficgp," %%*lf (%%*lf)");
8120: }
8121: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
8122: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 8123: 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 8124: for (j=1; j<= nlstate+1 ; j ++) {
8125: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8126: else fprintf(ficgp," %%*lf (%%*lf)");
8127: }
8128: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 8129: 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 8130: for (j=1; j<= nlstate+1 ; j ++) {
8131: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8132: else fprintf(ficgp," %%*lf (%%*lf)");
8133: }
8134: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
8135: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
8136: } /* state */
8137: } /* vpopbased */
1.264 brouard 8138: 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 8139: } /* end nres */
1.337 ! brouard 8140: /* } /\* k1 end 2 eme*\/ */
1.238 brouard 8141:
8142:
8143: /*3eme*/
1.337 ! brouard 8144: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8145: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 8146: k1=TKresult[nres];
! 8147: /* if(m != 1 && TKresult[nres]!= k1) */
! 8148: /* continue; */
1.238 brouard 8149:
1.332 brouard 8150: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 8151: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 8152: strcpy(gplotlabel,"(");
1.337 ! brouard 8153: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
! 8154: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8155: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8156: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
! 8157: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
! 8158: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
! 8159: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 8160: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 8161: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 8162: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
! 8163: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8164: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 8165: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 8166: /* } */
! 8167: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 8168: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
! 8169: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
! 8170: }
1.264 brouard 8171: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8172: fprintf(ficgp,"\n#\n");
8173: if(invalidvarcomb[k1]){
8174: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8175: continue;
8176: }
8177:
8178: /* k=2+nlstate*(2*cpt-2); */
8179: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 8180: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 8181: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 8182: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 8183: 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 8184: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8185: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8186: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
8187: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8188: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8189: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 8190:
1.238 brouard 8191: */
8192: for (i=1; i< nlstate ; i ++) {
1.261 brouard 8193: 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 8194: /* 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 8195:
1.238 brouard 8196: }
1.261 brouard 8197: 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 8198: }
1.264 brouard 8199: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 8200: } /* end nres */
1.337 ! brouard 8201: /* } /\* end kl 3eme *\/ */
1.126 brouard 8202:
1.223 brouard 8203: /* 4eme */
1.201 brouard 8204: /* Survival functions (period) from state i in state j by initial state i */
1.337 ! brouard 8205: /* for (k1=1; k1<=m; k1++){ /\* For each covariate and each value *\/ */
1.238 brouard 8206: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 8207: k1=TKresult[nres];
! 8208: /* if(m != 1 && TKresult[nres]!= k1) */
! 8209: /* continue; */
1.238 brouard 8210: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 8211: strcpy(gplotlabel,"(");
1.337 ! brouard 8212: fprintf(ficgp,"\n#\n#\n# Survival functions in state %d : 'LIJ_' files, cov=%d state=%d", cpt, k1, cpt);
! 8213: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
! 8214: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8215: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8216: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
! 8217: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
! 8218: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
! 8219: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 8220: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 8221: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 8222: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
! 8223: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8224: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 8225: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 8226: /* } */
! 8227: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 8228: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
! 8229: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8230: }
1.264 brouard 8231: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8232: fprintf(ficgp,"\n#\n");
8233: if(invalidvarcomb[k1]){
8234: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8235: continue;
1.223 brouard 8236: }
1.238 brouard 8237:
1.241 brouard 8238: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 8239: 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 8240: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8241: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8242: k=3;
8243: for (i=1; i<= nlstate ; i ++){
8244: if(i==1){
8245: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8246: }else{
8247: fprintf(ficgp,", '' ");
8248: }
8249: l=(nlstate+ndeath)*(i-1)+1;
8250: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8251: for (j=2; j<= nlstate+ndeath ; j ++)
8252: fprintf(ficgp,"+$%d",k+l+j-1);
8253: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
8254: } /* nlstate */
1.264 brouard 8255: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8256: } /* end cpt state*/
8257: } /* end nres */
1.337 ! brouard 8258: /* } /\* end covariate k1 *\/ */
1.238 brouard 8259:
1.220 brouard 8260: /* 5eme */
1.201 brouard 8261: /* Survival functions (period) from state i in state j by final state j */
1.337 ! brouard 8262: /* for (k1=1; k1<= m ; k1++){ /\* For each covariate combination if any *\/ */
1.238 brouard 8263: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 8264: k1=TKresult[nres];
! 8265: /* if(m != 1 && TKresult[nres]!= k1) */
! 8266: /* continue; */
1.238 brouard 8267: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 8268: strcpy(gplotlabel,"(");
1.238 brouard 8269: fprintf(ficgp,"\n#\n#\n# Survival functions in state j and all livestates from state i by final state j: 'lij' files, cov=%d state=%d",k1, cpt);
1.337 ! brouard 8270: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
! 8271: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8272: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8273: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
! 8274: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
! 8275: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
! 8276: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 8277: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 8278: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 8279: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
! 8280: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8281: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 8282: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 8283: /* } */
! 8284: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 8285: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
! 8286: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8287: }
1.264 brouard 8288: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8289: fprintf(ficgp,"\n#\n");
8290: if(invalidvarcomb[k1]){
8291: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8292: continue;
8293: }
1.227 brouard 8294:
1.241 brouard 8295: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8296: 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 8297: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8298: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8299: k=3;
8300: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8301: if(j==1)
8302: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8303: else
8304: fprintf(ficgp,", '' ");
8305: l=(nlstate+ndeath)*(cpt-1) +j;
8306: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8307: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8308: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8309: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8310: } /* nlstate */
8311: fprintf(ficgp,", '' ");
8312: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8313: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8314: l=(nlstate+ndeath)*(cpt-1) +j;
8315: if(j < nlstate)
8316: fprintf(ficgp,"$%d +",k+l);
8317: else
8318: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8319: }
1.264 brouard 8320: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8321: } /* end cpt state*/
1.337 ! brouard 8322: /* } /\* end covariate *\/ */
1.238 brouard 8323: } /* end nres */
1.227 brouard 8324:
1.220 brouard 8325: /* 6eme */
1.202 brouard 8326: /* CV preval stable (period) for each covariate */
1.337 ! brouard 8327: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8328: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 8329: k1=TKresult[nres];
! 8330: /* if(m != 1 && TKresult[nres]!= k1) */
! 8331: /* continue; */
1.255 brouard 8332: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8333: strcpy(gplotlabel,"(");
1.288 brouard 8334: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 ! brouard 8335: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
! 8336: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8337: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8338: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
! 8339: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
! 8340: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
! 8341: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 8342: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 8343: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 8344: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
! 8345: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8346: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 8347: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 8348: /* } */
! 8349: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 8350: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
! 8351: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8352: }
1.264 brouard 8353: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8354: fprintf(ficgp,"\n#\n");
1.223 brouard 8355: if(invalidvarcomb[k1]){
1.227 brouard 8356: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8357: continue;
1.223 brouard 8358: }
1.227 brouard 8359:
1.241 brouard 8360: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8361: 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 8362: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8363: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8364: k=3; /* Offset */
1.255 brouard 8365: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8366: if(i==1)
8367: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8368: else
8369: fprintf(ficgp,", '' ");
1.255 brouard 8370: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8371: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8372: for (j=2; j<= nlstate ; j ++)
8373: fprintf(ficgp,"+$%d",k+l+j-1);
8374: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8375: } /* nlstate */
1.264 brouard 8376: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8377: } /* end cpt state*/
8378: } /* end covariate */
1.227 brouard 8379:
8380:
1.220 brouard 8381: /* 7eme */
1.296 brouard 8382: if(prevbcast == 1){
1.288 brouard 8383: /* CV backward prevalence for each covariate */
1.337 ! brouard 8384: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8385: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 8386: k1=TKresult[nres];
! 8387: /* if(m != 1 && TKresult[nres]!= k1) */
! 8388: /* continue; */
1.268 brouard 8389: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8390: strcpy(gplotlabel,"(");
1.288 brouard 8391: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 ! brouard 8392: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
! 8393: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8394: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8395: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
! 8396: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
! 8397: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
! 8398: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 8399: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 8400: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 8401: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
! 8402: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8403: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 8404: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 8405: /* } */
! 8406: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 8407: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
! 8408: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8409: }
1.264 brouard 8410: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8411: fprintf(ficgp,"\n#\n");
8412: if(invalidvarcomb[k1]){
8413: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8414: continue;
8415: }
8416:
1.241 brouard 8417: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8418: 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 8419: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8420: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8421: k=3; /* Offset */
1.268 brouard 8422: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8423: if(i==1)
8424: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8425: else
8426: fprintf(ficgp,", '' ");
8427: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8428: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8429: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8430: /* 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 8431: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8432: /* for (j=2; j<= nlstate ; j ++) */
8433: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8434: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8435: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8436: } /* nlstate */
1.264 brouard 8437: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8438: } /* end cpt state*/
8439: } /* end covariate */
1.296 brouard 8440: } /* End if prevbcast */
1.218 brouard 8441:
1.223 brouard 8442: /* 8eme */
1.218 brouard 8443: if(prevfcast==1){
1.288 brouard 8444: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8445:
1.337 ! brouard 8446: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8447: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 8448: k1=TKresult[nres];
! 8449: /* if(m != 1 && TKresult[nres]!= k1) */
! 8450: /* continue; */
1.211 brouard 8451: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8452: strcpy(gplotlabel,"(");
1.288 brouard 8453: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 ! brouard 8454: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
! 8455: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8456: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8457: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
! 8458: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
! 8459: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
! 8460: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 8461: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 8462: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 8463: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
! 8464: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8465: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 8466: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 8467: /* } */
! 8468: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 8469: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
! 8470: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8471: }
1.264 brouard 8472: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8473: fprintf(ficgp,"\n#\n");
8474: if(invalidvarcomb[k1]){
8475: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8476: continue;
8477: }
8478:
8479: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8480: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8481: 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 8482: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8483: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8484:
8485: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8486: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8487: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8488: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8489: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8490: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8491: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8492: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8493: if(i==istart){
1.227 brouard 8494: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8495: }else{
8496: fprintf(ficgp,",\\\n '' ");
8497: }
8498: if(cptcoveff ==0){ /* No covariate */
8499: ioffset=2; /* Age is in 2 */
8500: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8501: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8502: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8503: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8504: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8505: if(i==nlstate+1){
1.270 brouard 8506: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8507: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8508: fprintf(ficgp,",\\\n '' ");
8509: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8510: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8511: offyear, \
1.268 brouard 8512: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8513: }else
1.227 brouard 8514: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8515: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8516: }else{ /* more than 2 covariates */
1.270 brouard 8517: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8518: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8519: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8520: iyearc=ioffset-1;
8521: iagec=ioffset;
1.227 brouard 8522: fprintf(ficgp," u %d:(",ioffset);
8523: kl=0;
8524: strcpy(gplotcondition,"(");
8525: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8526: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8527: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8528: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8529: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8530: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8531: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8532: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8533: kl++;
8534: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8535: kl++;
8536: if(k <cptcoveff && cptcoveff>1)
8537: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8538: }
8539: strcpy(gplotcondition+strlen(gplotcondition),")");
8540: /* 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 *\/ */
8541: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8542: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8543: /* '' 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*/
8544: if(i==nlstate+1){
1.270 brouard 8545: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8546: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8547: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8548: fprintf(ficgp," u %d:(",iagec);
8549: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8550: iyearc, iagec, offyear, \
8551: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8552: /* '' 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 8553: }else{
8554: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8555: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8556: }
8557: } /* end if covariate */
8558: } /* nlstate */
1.264 brouard 8559: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8560: } /* end cpt state*/
8561: } /* end covariate */
8562: } /* End if prevfcast */
1.227 brouard 8563:
1.296 brouard 8564: if(prevbcast==1){
1.268 brouard 8565: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8566:
1.337 ! brouard 8567: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.268 brouard 8568: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 8569: k1=TKresult[nres];
! 8570: /* if(m != 1 && TKresult[nres]!= k1) */
! 8571: /* continue; */
1.268 brouard 8572: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8573: strcpy(gplotlabel,"(");
8574: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
1.337 ! brouard 8575: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
! 8576: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8577: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8578: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
! 8579: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
! 8580: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
! 8581: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 8582: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 8583: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 8584: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
! 8585: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8586: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 8587: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 8588: /* } */
! 8589: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 8590: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
! 8591: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.268 brouard 8592: }
8593: strcpy(gplotlabel+strlen(gplotlabel),")");
8594: fprintf(ficgp,"\n#\n");
8595: if(invalidvarcomb[k1]){
8596: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8597: continue;
8598: }
8599:
8600: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8601: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8602: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8603: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8604: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8605:
8606: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8607: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8608: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8609: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8610: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8611: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8612: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8613: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8614: if(i==istart){
8615: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8616: }else{
8617: fprintf(ficgp,",\\\n '' ");
8618: }
8619: if(cptcoveff ==0){ /* No covariate */
8620: ioffset=2; /* Age is in 2 */
8621: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8622: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8623: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8624: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8625: fprintf(ficgp," u %d:(", ioffset);
8626: if(i==nlstate+1){
1.270 brouard 8627: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8628: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8629: fprintf(ficgp,",\\\n '' ");
8630: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8631: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8632: offbyear, \
8633: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8634: }else
8635: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8636: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8637: }else{ /* more than 2 covariates */
1.270 brouard 8638: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8639: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8640: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8641: iyearc=ioffset-1;
8642: iagec=ioffset;
1.268 brouard 8643: fprintf(ficgp," u %d:(",ioffset);
8644: kl=0;
8645: strcpy(gplotcondition,"(");
1.337 ! brouard 8646: for (k=1; k<=cptcovs; k++){ /* For each covariate k of the resultline, get corresponding value lv for combination k1 */
! 8647: if(Dummy[Tvresult[nres][k]]==0){ /* To be verified */
! 8648: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
! 8649: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
! 8650: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
! 8651: lv=Tvresult[nres][k];
! 8652: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
! 8653: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
! 8654: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
! 8655: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
! 8656: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
! 8657: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8658: kl++;
! 8659: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
! 8660: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%lg " ,kl,Tvresult[nres][k], kl+1,TinvDoQresult[nres][Tvresult[nres][k]]);
! 8661: kl++;
! 8662: if(k <cptcoveff && cptcoveff>1)
! 8663: sprintf(gplotcondition+strlen(gplotcondition)," && ");
! 8664: }
1.268 brouard 8665: }
8666: strcpy(gplotcondition+strlen(gplotcondition),")");
8667: /* 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 *\/ */
8668: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8669: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8670: /* '' 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*/
8671: if(i==nlstate+1){
1.270 brouard 8672: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8673: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8674: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8675: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8676: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8677: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8678: iyearc,iagec,offbyear, \
8679: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8680: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8681: }else{
8682: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8683: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8684: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8685: }
8686: } /* end if covariate */
8687: } /* nlstate */
8688: fprintf(ficgp,"\nset out; unset label;\n");
8689: } /* end cpt state*/
8690: } /* end covariate */
1.296 brouard 8691: } /* End if prevbcast */
1.268 brouard 8692:
1.227 brouard 8693:
1.238 brouard 8694: /* 9eme writing MLE parameters */
8695: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8696: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8697: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8698: for(k=1; k <=(nlstate+ndeath); k++){
8699: if (k != i) {
1.227 brouard 8700: fprintf(ficgp,"# current state %d\n",k);
8701: for(j=1; j <=ncovmodel; j++){
8702: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8703: jk++;
8704: }
8705: fprintf(ficgp,"\n");
1.126 brouard 8706: }
8707: }
1.223 brouard 8708: }
1.187 brouard 8709: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8710:
1.145 brouard 8711: /*goto avoid;*/
1.238 brouard 8712: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8713: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8714: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8715: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8716: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8717: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8718: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8719: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8720: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8721: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8722: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8723: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8724: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8725: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8726: fprintf(ficgp,"#\n");
1.223 brouard 8727: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8728: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8729: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8730: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8731: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
1.337 ! brouard 8732: /* for(k1=1; k1 <=m; k1++) /\* For each combination of covariate *\/ */
1.237 brouard 8733: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 8734: /* k1=nres; */
! 8735: k1= TKresult[nres];
! 8736: fprintf(ficgp,"\n\n# Resultline k1=%d ",k1);
1.264 brouard 8737: strcpy(gplotlabel,"(");
1.276 brouard 8738: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.337 ! brouard 8739: for (k=1; k<=cptcovs; k++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
! 8740: /* for each resultline nres, and position k, Tvresult[nres][k] gives the name of the variable and
! 8741: TinvDoQresult[nres][Tvresult[nres][k]] gives its value double or integer) */
! 8742: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8743: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
! 8744: }
! 8745: /* if(m != 1 && TKresult[nres]!= k1) */
! 8746: /* continue; */
! 8747: /* fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1); */
! 8748: /* strcpy(gplotlabel,"("); */
! 8749: /* /\*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*\/ */
! 8750: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
! 8751: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
! 8752: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
! 8753: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
! 8754: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
! 8755: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
! 8756: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
! 8757: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
! 8758: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
! 8759: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
! 8760: /* } */
! 8761: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 8762: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
! 8763: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
! 8764: /* } */
1.264 brouard 8765: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8766: fprintf(ficgp,"\n#\n");
1.264 brouard 8767: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8768: fprintf(ficgp,"\nset key outside ");
8769: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8770: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8771: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8772: if (ng==1){
8773: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8774: fprintf(ficgp,"\nunset log y");
8775: }else if (ng==2){
8776: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8777: fprintf(ficgp,"\nset log y");
8778: }else if (ng==3){
8779: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8780: fprintf(ficgp,"\nset log y");
8781: }else
8782: fprintf(ficgp,"\nunset title ");
8783: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8784: i=1;
8785: for(k2=1; k2<=nlstate; k2++) {
8786: k3=i;
8787: for(k=1; k<=(nlstate+ndeath); k++) {
8788: if (k != k2){
8789: switch( ng) {
8790: case 1:
8791: if(nagesqr==0)
8792: fprintf(ficgp," p%d+p%d*x",i,i+1);
8793: else /* nagesqr =1 */
8794: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8795: break;
8796: case 2: /* ng=2 */
8797: if(nagesqr==0)
8798: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8799: else /* nagesqr =1 */
8800: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8801: break;
8802: case 3:
8803: if(nagesqr==0)
8804: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8805: else /* nagesqr =1 */
8806: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8807: break;
8808: }
8809: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8810: ijp=1; /* product no age */
8811: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8812: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8813: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 8814: switch(Typevar[j]){
8815: case 1:
8816: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8817: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
8818: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8819: if(DummyV[j]==0){/* Bug valgrind */
8820: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8821: }else{ /* quantitative */
8822: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8823: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8824: }
8825: ij++;
1.268 brouard 8826: }
1.237 brouard 8827: }
1.329 brouard 8828: }
8829: break;
8830: case 2:
8831: if(cptcovprod >0){
8832: if(j==Tprod[ijp]) { /* */
8833: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8834: if(ijp <=cptcovprod) { /* Product */
8835: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8836: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8837: /* 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)]); */
8838: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8839: }else{ /* Vn is dummy and Vm is quanti */
8840: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8841: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8842: }
8843: }else{ /* Vn*Vm Vn is quanti */
8844: if(DummyV[Tvard[ijp][2]]==0){
8845: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8846: }else{ /* Both quanti */
8847: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8848: }
1.268 brouard 8849: }
1.329 brouard 8850: ijp++;
1.237 brouard 8851: }
1.329 brouard 8852: } /* end Tprod */
8853: }
8854: break;
8855: case 0:
8856: /* simple covariate */
1.264 brouard 8857: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8858: if(Dummy[j]==0){
8859: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8860: }else{ /* quantitative */
8861: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8862: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8863: }
1.329 brouard 8864: /* end simple */
8865: break;
8866: default:
8867: break;
8868: } /* end switch */
1.237 brouard 8869: } /* end j */
1.329 brouard 8870: }else{ /* k=k2 */
8871: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
8872: fprintf(ficgp," (1.");i=i-ncovmodel;
8873: }else
8874: i=i-ncovmodel;
1.223 brouard 8875: }
1.227 brouard 8876:
1.223 brouard 8877: if(ng != 1){
8878: fprintf(ficgp,")/(1");
1.227 brouard 8879:
1.264 brouard 8880: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8881: if(nagesqr==0)
1.264 brouard 8882: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8883: else /* nagesqr =1 */
1.264 brouard 8884: 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 8885:
1.223 brouard 8886: ij=1;
1.329 brouard 8887: ijp=1;
8888: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
8889: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
8890: switch(Typevar[j]){
8891: case 1:
8892: if(cptcovage >0){
8893: if(j==Tage[ij]) { /* Bug valgrind */
8894: if(ij <=cptcovage) { /* Bug valgrind */
8895: if(DummyV[j]==0){/* Bug valgrind */
8896: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
8897: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
8898: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
8899: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
8900: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8901: }else{ /* quantitative */
8902: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8903: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8904: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8905: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8906: }
8907: ij++;
8908: }
8909: }
8910: }
8911: break;
8912: case 2:
8913: if(cptcovprod >0){
8914: if(j==Tprod[ijp]) { /* */
8915: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8916: if(ijp <=cptcovprod) { /* Product */
8917: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8918: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8919: /* 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)]); */
8920: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8921: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
8922: }else{ /* Vn is dummy and Vm is quanti */
8923: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8924: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8925: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8926: }
8927: }else{ /* Vn*Vm Vn is quanti */
8928: if(DummyV[Tvard[ijp][2]]==0){
8929: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8930: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
8931: }else{ /* Both quanti */
8932: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8933: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8934: }
8935: }
8936: ijp++;
8937: }
8938: } /* end Tprod */
8939: } /* end if */
8940: break;
8941: case 0:
8942: /* simple covariate */
8943: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
8944: if(Dummy[j]==0){
8945: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8946: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
8947: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8948: }else{ /* quantitative */
8949: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
8950: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
8951: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8952: }
8953: /* end simple */
8954: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
8955: break;
8956: default:
8957: break;
8958: } /* end switch */
1.223 brouard 8959: }
8960: fprintf(ficgp,")");
8961: }
8962: fprintf(ficgp,")");
8963: if(ng ==2)
1.276 brouard 8964: 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 8965: else /* ng= 3 */
1.276 brouard 8966: 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 8967: }else{ /* end ng <> 1 */
1.223 brouard 8968: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8969: 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 8970: }
8971: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8972: fprintf(ficgp,",");
8973: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8974: fprintf(ficgp,",");
8975: i=i+ncovmodel;
8976: } /* end k */
8977: } /* end k2 */
1.276 brouard 8978: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8979: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.337 ! brouard 8980: } /* end resultline */
1.223 brouard 8981: } /* end ng */
8982: /* avoid: */
8983: fflush(ficgp);
1.126 brouard 8984: } /* end gnuplot */
8985:
8986:
8987: /*************** Moving average **************/
1.219 brouard 8988: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8989: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8990:
1.222 brouard 8991: int i, cpt, cptcod;
8992: int modcovmax =1;
8993: int mobilavrange, mob;
8994: int iage=0;
1.288 brouard 8995: int firstA1=0, firstA2=0;
1.222 brouard 8996:
1.266 brouard 8997: double sum=0., sumr=0.;
1.222 brouard 8998: double age;
1.266 brouard 8999: double *sumnewp, *sumnewm, *sumnewmr;
9000: double *agemingood, *agemaxgood;
9001: double *agemingoodr, *agemaxgoodr;
1.222 brouard 9002:
9003:
1.278 brouard 9004: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
9005: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 9006:
9007: sumnewp = vector(1,ncovcombmax);
9008: sumnewm = vector(1,ncovcombmax);
1.266 brouard 9009: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 9010: agemingood = vector(1,ncovcombmax);
1.266 brouard 9011: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 9012: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 9013: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 9014:
9015: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 9016: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 9017: sumnewp[cptcod]=0.;
1.266 brouard 9018: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
9019: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 9020: }
9021: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
9022:
1.266 brouard 9023: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
9024: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 9025: else mobilavrange=mobilav;
9026: for (age=bage; age<=fage; age++)
9027: for (i=1; i<=nlstate;i++)
9028: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
9029: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9030: /* We keep the original values on the extreme ages bage, fage and for
9031: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
9032: we use a 5 terms etc. until the borders are no more concerned.
9033: */
9034: for (mob=3;mob <=mobilavrange;mob=mob+2){
9035: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 9036: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
9037: sumnewm[cptcod]=0.;
9038: for (i=1; i<=nlstate;i++){
1.222 brouard 9039: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
9040: for (cpt=1;cpt<=(mob-1)/2;cpt++){
9041: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
9042: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
9043: }
9044: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 9045: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9046: } /* end i */
9047: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
9048: } /* end cptcod */
1.222 brouard 9049: }/* end age */
9050: }/* end mob */
1.266 brouard 9051: }else{
9052: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 9053: return -1;
1.266 brouard 9054: }
9055:
9056: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 9057: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
9058: if(invalidvarcomb[cptcod]){
9059: printf("\nCombination (%d) ignored because no cases \n",cptcod);
9060: continue;
9061: }
1.219 brouard 9062:
1.266 brouard 9063: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
9064: sumnewm[cptcod]=0.;
9065: sumnewmr[cptcod]=0.;
9066: for (i=1; i<=nlstate;i++){
9067: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9068: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9069: }
9070: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9071: agemingoodr[cptcod]=age;
9072: }
9073: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9074: agemingood[cptcod]=age;
9075: }
9076: } /* age */
9077: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 9078: sumnewm[cptcod]=0.;
1.266 brouard 9079: sumnewmr[cptcod]=0.;
1.222 brouard 9080: for (i=1; i<=nlstate;i++){
9081: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9082: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9083: }
9084: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9085: agemaxgoodr[cptcod]=age;
1.222 brouard 9086: }
9087: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 9088: agemaxgood[cptcod]=age;
9089: }
9090: } /* age */
9091: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
9092: /* but they will change */
1.288 brouard 9093: firstA1=0;firstA2=0;
1.266 brouard 9094: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
9095: sumnewm[cptcod]=0.;
9096: sumnewmr[cptcod]=0.;
9097: for (i=1; i<=nlstate;i++){
9098: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9099: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9100: }
9101: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9102: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9103: agemaxgoodr[cptcod]=age; /* age min */
9104: for (i=1; i<=nlstate;i++)
9105: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9106: }else{ /* bad we change the value with the values of good ages */
9107: for (i=1; i<=nlstate;i++){
9108: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
9109: } /* i */
9110: } /* end bad */
9111: }else{
9112: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9113: agemaxgood[cptcod]=age;
9114: }else{ /* bad we change the value with the values of good ages */
9115: for (i=1; i<=nlstate;i++){
9116: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
9117: } /* i */
9118: } /* end bad */
9119: }/* end else */
9120: sum=0.;sumr=0.;
9121: for (i=1; i<=nlstate;i++){
9122: sum+=mobaverage[(int)age][i][cptcod];
9123: sumr+=probs[(int)age][i][cptcod];
9124: }
9125: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 9126: if(!firstA1){
9127: firstA1=1;
9128: 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);
9129: }
9130: 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 9131: } /* end bad */
9132: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9133: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 9134: if(!firstA2){
9135: firstA2=1;
9136: 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);
9137: }
9138: 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 9139: } /* end bad */
9140: }/* age */
1.266 brouard 9141:
9142: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 9143: sumnewm[cptcod]=0.;
1.266 brouard 9144: sumnewmr[cptcod]=0.;
1.222 brouard 9145: for (i=1; i<=nlstate;i++){
9146: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9147: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9148: }
9149: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9150: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
9151: agemingoodr[cptcod]=age;
9152: for (i=1; i<=nlstate;i++)
9153: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9154: }else{ /* bad we change the value with the values of good ages */
9155: for (i=1; i<=nlstate;i++){
9156: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
9157: } /* i */
9158: } /* end bad */
9159: }else{
9160: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9161: agemingood[cptcod]=age;
9162: }else{ /* bad */
9163: for (i=1; i<=nlstate;i++){
9164: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
9165: } /* i */
9166: } /* end bad */
9167: }/* end else */
9168: sum=0.;sumr=0.;
9169: for (i=1; i<=nlstate;i++){
9170: sum+=mobaverage[(int)age][i][cptcod];
9171: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 9172: }
1.266 brouard 9173: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 9174: 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 9175: } /* end bad */
9176: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9177: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 9178: 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 9179: } /* end bad */
9180: }/* age */
1.266 brouard 9181:
1.222 brouard 9182:
9183: for (age=bage; age<=fage; age++){
1.235 brouard 9184: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 9185: sumnewp[cptcod]=0.;
9186: sumnewm[cptcod]=0.;
9187: for (i=1; i<=nlstate;i++){
9188: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
9189: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9190: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
9191: }
9192: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
9193: }
9194: /* printf("\n"); */
9195: /* } */
1.266 brouard 9196:
1.222 brouard 9197: /* brutal averaging */
1.266 brouard 9198: /* for (i=1; i<=nlstate;i++){ */
9199: /* for (age=1; age<=bage; age++){ */
9200: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
9201: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9202: /* } */
9203: /* for (age=fage; age<=AGESUP; age++){ */
9204: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
9205: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9206: /* } */
9207: /* } /\* end i status *\/ */
9208: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
9209: /* for (age=1; age<=AGESUP; age++){ */
9210: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
9211: /* mobaverage[(int)age][i][cptcod]=0.; */
9212: /* } */
9213: /* } */
1.222 brouard 9214: }/* end cptcod */
1.266 brouard 9215: free_vector(agemaxgoodr,1, ncovcombmax);
9216: free_vector(agemaxgood,1, ncovcombmax);
9217: free_vector(agemingood,1, ncovcombmax);
9218: free_vector(agemingoodr,1, ncovcombmax);
9219: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 9220: free_vector(sumnewm,1, ncovcombmax);
9221: free_vector(sumnewp,1, ncovcombmax);
9222: return 0;
9223: }/* End movingaverage */
1.218 brouard 9224:
1.126 brouard 9225:
1.296 brouard 9226:
1.126 brouard 9227: /************** Forecasting ******************/
1.296 brouard 9228: /* 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)*/
9229: 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){
9230: /* dateintemean, mean date of interviews
9231: dateprojd, year, month, day of starting projection
9232: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 9233: agemin, agemax range of age
9234: dateprev1 dateprev2 range of dates during which prevalence is computed
9235: */
1.296 brouard 9236: /* double anprojd, mprojd, jprojd; */
9237: /* double anprojf, mprojf, jprojf; */
1.267 brouard 9238: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 9239: double agec; /* generic age */
1.296 brouard 9240: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 9241: double *popeffectif,*popcount;
9242: double ***p3mat;
1.218 brouard 9243: /* double ***mobaverage; */
1.126 brouard 9244: char fileresf[FILENAMELENGTH];
9245:
9246: agelim=AGESUP;
1.211 brouard 9247: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9248: in each health status at the date of interview (if between dateprev1 and dateprev2).
9249: We still use firstpass and lastpass as another selection.
9250: */
1.214 brouard 9251: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9252: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 9253:
1.201 brouard 9254: strcpy(fileresf,"F_");
9255: strcat(fileresf,fileresu);
1.126 brouard 9256: if((ficresf=fopen(fileresf,"w"))==NULL) {
9257: printf("Problem with forecast resultfile: %s\n", fileresf);
9258: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
9259: }
1.235 brouard 9260: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
9261: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 9262:
1.225 brouard 9263: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 9264:
9265:
9266: stepsize=(int) (stepm+YEARM-1)/YEARM;
9267: if (stepm<=12) stepsize=1;
9268: if(estepm < stepm){
9269: printf ("Problem %d lower than %d\n",estepm, stepm);
9270: }
1.270 brouard 9271: else{
9272: hstepm=estepm;
9273: }
9274: if(estepm > stepm){ /* Yes every two year */
9275: stepsize=2;
9276: }
1.296 brouard 9277: hstepm=hstepm/stepm;
1.126 brouard 9278:
1.296 brouard 9279:
9280: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9281: /* fractional in yp1 *\/ */
9282: /* aintmean=yp; */
9283: /* yp2=modf((yp1*12),&yp); */
9284: /* mintmean=yp; */
9285: /* yp1=modf((yp2*30.5),&yp); */
9286: /* jintmean=yp; */
9287: /* if(jintmean==0) jintmean=1; */
9288: /* if(mintmean==0) mintmean=1; */
1.126 brouard 9289:
1.296 brouard 9290:
9291: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
9292: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
9293: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 9294: i1=pow(2,cptcoveff);
1.126 brouard 9295: if (cptcovn < 1){i1=1;}
9296:
1.296 brouard 9297: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 9298:
9299: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 9300:
1.126 brouard 9301: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 9302: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9303: 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 9304: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9305: continue;
1.227 brouard 9306: if(invalidvarcomb[k]){
9307: printf("\nCombination (%d) projection ignored because no cases \n",k);
9308: continue;
9309: }
9310: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
9311: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9312: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); */
9313: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 9314: }
1.235 brouard 9315: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 9316: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 9317: }
1.227 brouard 9318: fprintf(ficresf," yearproj age");
9319: for(j=1; j<=nlstate+ndeath;j++){
9320: for(i=1; i<=nlstate;i++)
9321: fprintf(ficresf," p%d%d",i,j);
9322: fprintf(ficresf," wp.%d",j);
9323: }
1.296 brouard 9324: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 9325: fprintf(ficresf,"\n");
1.296 brouard 9326: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 9327: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
9328: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 9329: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
9330: nhstepm = nhstepm/hstepm;
9331: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9332: oldm=oldms;savm=savms;
1.268 brouard 9333: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 9334: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 9335: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 9336: for (h=0; h<=nhstepm; h++){
9337: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 9338: break;
9339: }
9340: }
9341: fprintf(ficresf,"\n");
9342: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9343: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
9344: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* TnsdVar[Tvaraff] correct */
1.296 brouard 9345: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9346:
9347: for(j=1; j<=nlstate+ndeath;j++) {
9348: ppij=0.;
9349: for(i=1; i<=nlstate;i++) {
1.278 brouard 9350: if (mobilav>=1)
9351: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9352: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9353: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9354: }
1.268 brouard 9355: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9356: } /* end i */
9357: fprintf(ficresf," %.3f", ppij);
9358: }/* end j */
1.227 brouard 9359: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9360: } /* end agec */
1.266 brouard 9361: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9362: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9363: } /* end yearp */
9364: } /* end k */
1.219 brouard 9365:
1.126 brouard 9366: fclose(ficresf);
1.215 brouard 9367: printf("End of Computing forecasting \n");
9368: fprintf(ficlog,"End of Computing forecasting\n");
9369:
1.126 brouard 9370: }
9371:
1.269 brouard 9372: /************** Back Forecasting ******************/
1.296 brouard 9373: /* 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){ */
9374: 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){
9375: /* back1, year, month, day of starting backprojection
1.267 brouard 9376: agemin, agemax range of age
9377: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9378: anback2 year of end of backprojection (same day and month as back1).
9379: prevacurrent and prev are prevalences.
1.267 brouard 9380: */
9381: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9382: double agec; /* generic age */
1.302 brouard 9383: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9384: double *popeffectif,*popcount;
9385: double ***p3mat;
9386: /* double ***mobaverage; */
9387: char fileresfb[FILENAMELENGTH];
9388:
1.268 brouard 9389: agelim=AGEINF;
1.267 brouard 9390: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9391: in each health status at the date of interview (if between dateprev1 and dateprev2).
9392: We still use firstpass and lastpass as another selection.
9393: */
9394: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9395: /* firstpass, lastpass, stepm, weightopt, model); */
9396:
9397: /*Do we need to compute prevalence again?*/
9398:
9399: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9400:
9401: strcpy(fileresfb,"FB_");
9402: strcat(fileresfb,fileresu);
9403: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9404: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9405: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9406: }
9407: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9408: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9409:
9410: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9411:
9412:
9413: stepsize=(int) (stepm+YEARM-1)/YEARM;
9414: if (stepm<=12) stepsize=1;
9415: if(estepm < stepm){
9416: printf ("Problem %d lower than %d\n",estepm, stepm);
9417: }
1.270 brouard 9418: else{
9419: hstepm=estepm;
9420: }
9421: if(estepm >= stepm){ /* Yes every two year */
9422: stepsize=2;
9423: }
1.267 brouard 9424:
9425: hstepm=hstepm/stepm;
1.296 brouard 9426: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9427: /* fractional in yp1 *\/ */
9428: /* aintmean=yp; */
9429: /* yp2=modf((yp1*12),&yp); */
9430: /* mintmean=yp; */
9431: /* yp1=modf((yp2*30.5),&yp); */
9432: /* jintmean=yp; */
9433: /* if(jintmean==0) jintmean=1; */
9434: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9435:
9436: i1=pow(2,cptcoveff);
9437: if (cptcovn < 1){i1=1;}
9438:
1.296 brouard 9439: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9440: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9441:
9442: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9443:
9444: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9445: for(k=1; k<=i1;k++){
9446: if(i1 != 1 && TKresult[nres]!= k)
9447: continue;
9448: if(invalidvarcomb[k]){
9449: printf("\nCombination (%d) projection ignored because no cases \n",k);
9450: continue;
9451: }
1.268 brouard 9452: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9453: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9454: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.267 brouard 9455: }
9456: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9457: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9458: }
9459: fprintf(ficresfb," yearbproj age");
9460: for(j=1; j<=nlstate+ndeath;j++){
9461: for(i=1; i<=nlstate;i++)
1.268 brouard 9462: fprintf(ficresfb," b%d%d",i,j);
9463: fprintf(ficresfb," b.%d",j);
1.267 brouard 9464: }
1.296 brouard 9465: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9466: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9467: fprintf(ficresfb,"\n");
1.296 brouard 9468: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9469: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9470: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9471: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9472: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9473: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9474: nhstepm = nhstepm/hstepm;
9475: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9476: oldm=oldms;savm=savms;
1.268 brouard 9477: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9478: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9479: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9480: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9481: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9482: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9483: for (h=0; h<=nhstepm; h++){
1.268 brouard 9484: if (h*hstepm/YEARM*stepm ==-yearp) {
9485: break;
9486: }
9487: }
9488: fprintf(ficresfb,"\n");
9489: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9490: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.296 brouard 9491: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9492: for(i=1; i<=nlstate+ndeath;i++) {
9493: ppij=0.;ppi=0.;
9494: for(j=1; j<=nlstate;j++) {
9495: /* if (mobilav==1) */
1.269 brouard 9496: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9497: ppi=ppi+prevacurrent[(int)agec][j][k];
9498: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9499: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9500: /* else { */
9501: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9502: /* } */
1.268 brouard 9503: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9504: } /* end j */
9505: if(ppi <0.99){
9506: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9507: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9508: }
9509: fprintf(ficresfb," %.3f", ppij);
9510: }/* end j */
1.267 brouard 9511: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9512: } /* end agec */
9513: } /* end yearp */
9514: } /* end k */
1.217 brouard 9515:
1.267 brouard 9516: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9517:
1.267 brouard 9518: fclose(ficresfb);
9519: printf("End of Computing Back forecasting \n");
9520: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9521:
1.267 brouard 9522: }
1.217 brouard 9523:
1.269 brouard 9524: /* Variance of prevalence limit: varprlim */
9525: 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 9526: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9527:
9528: char fileresvpl[FILENAMELENGTH];
9529: FILE *ficresvpl;
9530: double **oldm, **savm;
9531: double **varpl; /* Variances of prevalence limits by age */
9532: int i1, k, nres, j ;
9533:
9534: strcpy(fileresvpl,"VPL_");
9535: strcat(fileresvpl,fileresu);
9536: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9537: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9538: exit(0);
9539: }
1.288 brouard 9540: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9541: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9542:
9543: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9544: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9545:
9546: i1=pow(2,cptcoveff);
9547: if (cptcovn < 1){i1=1;}
9548:
1.337 ! brouard 9549: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
! 9550: k=TKresult[nres];
! 9551: /* for(k=1; k<=i1;k++){ /\* We find the combination equivalent to result line values of dummies *\/ */
1.269 brouard 9552: if(i1 != 1 && TKresult[nres]!= k)
9553: continue;
9554: fprintf(ficresvpl,"\n#****** ");
9555: printf("\n#****** ");
9556: fprintf(ficlog,"\n#****** ");
1.337 ! brouard 9557: for(j=1;j<=cptcovs;j++) {
! 9558: fprintf(ficresvpl,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
! 9559: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
! 9560: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
! 9561: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 9562: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.269 brouard 9563: }
1.337 ! brouard 9564: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
! 9565: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
! 9566: /* fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
! 9567: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
! 9568: /* } */
1.269 brouard 9569: fprintf(ficresvpl,"******\n");
9570: printf("******\n");
9571: fprintf(ficlog,"******\n");
9572:
9573: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9574: oldm=oldms;savm=savms;
9575: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9576: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9577: /*}*/
9578: }
9579:
9580: fclose(ficresvpl);
1.288 brouard 9581: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9582: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9583:
9584: }
9585: /* Variance of back prevalence: varbprlim */
9586: 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){
9587: /*------- Variance of back (stable) prevalence------*/
9588:
9589: char fileresvbl[FILENAMELENGTH];
9590: FILE *ficresvbl;
9591:
9592: double **oldm, **savm;
9593: double **varbpl; /* Variances of back prevalence limits by age */
9594: int i1, k, nres, j ;
9595:
9596: strcpy(fileresvbl,"VBL_");
9597: strcat(fileresvbl,fileresu);
9598: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9599: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9600: exit(0);
9601: }
9602: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9603: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9604:
9605:
9606: i1=pow(2,cptcoveff);
9607: if (cptcovn < 1){i1=1;}
9608:
1.337 ! brouard 9609: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
! 9610: k=TKresult[nres];
! 9611: /* for(k=1; k<=i1;k++){ */
! 9612: /* if(i1 != 1 && TKresult[nres]!= k) */
! 9613: /* continue; */
1.269 brouard 9614: fprintf(ficresvbl,"\n#****** ");
9615: printf("\n#****** ");
9616: fprintf(ficlog,"\n#****** ");
1.337 ! brouard 9617: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */
! 9618: printf(" V%d=%lg ",Tvqresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
! 9619: fprintf(ficresvbl," V%d=%lg ",Tvqresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
! 9620: fprintf(ficlog," V%d=%lg ",Tvqresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
! 9621: /* for(j=1;j<=cptcoveff;j++) { */
! 9622: /* fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 9623: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 9624: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 9625: /* } */
! 9626: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
! 9627: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
! 9628: /* fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
! 9629: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
1.269 brouard 9630: }
9631: fprintf(ficresvbl,"******\n");
9632: printf("******\n");
9633: fprintf(ficlog,"******\n");
9634:
9635: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9636: oldm=oldms;savm=savms;
9637:
9638: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9639: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9640: /*}*/
9641: }
9642:
9643: fclose(ficresvbl);
9644: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9645: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9646:
9647: } /* End of varbprlim */
9648:
1.126 brouard 9649: /************** Forecasting *****not tested NB*************/
1.227 brouard 9650: /* 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 9651:
1.227 brouard 9652: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9653: /* int *popage; */
9654: /* double calagedatem, agelim, kk1, kk2; */
9655: /* double *popeffectif,*popcount; */
9656: /* double ***p3mat,***tabpop,***tabpopprev; */
9657: /* /\* double ***mobaverage; *\/ */
9658: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9659:
1.227 brouard 9660: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9661: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9662: /* agelim=AGESUP; */
9663: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9664:
1.227 brouard 9665: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9666:
9667:
1.227 brouard 9668: /* strcpy(filerespop,"POP_"); */
9669: /* strcat(filerespop,fileresu); */
9670: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9671: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9672: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9673: /* } */
9674: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9675: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9676:
1.227 brouard 9677: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9678:
1.227 brouard 9679: /* /\* if (mobilav!=0) { *\/ */
9680: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9681: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9682: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9683: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9684: /* /\* } *\/ */
9685: /* /\* } *\/ */
1.126 brouard 9686:
1.227 brouard 9687: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9688: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9689:
1.227 brouard 9690: /* agelim=AGESUP; */
1.126 brouard 9691:
1.227 brouard 9692: /* hstepm=1; */
9693: /* hstepm=hstepm/stepm; */
1.218 brouard 9694:
1.227 brouard 9695: /* if (popforecast==1) { */
9696: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9697: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9698: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9699: /* } */
9700: /* popage=ivector(0,AGESUP); */
9701: /* popeffectif=vector(0,AGESUP); */
9702: /* popcount=vector(0,AGESUP); */
1.126 brouard 9703:
1.227 brouard 9704: /* i=1; */
9705: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9706:
1.227 brouard 9707: /* imx=i; */
9708: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9709: /* } */
1.218 brouard 9710:
1.227 brouard 9711: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9712: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9713: /* k=k+1; */
9714: /* fprintf(ficrespop,"\n#******"); */
9715: /* for(j=1;j<=cptcoveff;j++) { */
9716: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9717: /* } */
9718: /* fprintf(ficrespop,"******\n"); */
9719: /* fprintf(ficrespop,"# Age"); */
9720: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9721: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9722:
1.227 brouard 9723: /* for (cpt=0; cpt<=0;cpt++) { */
9724: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9725:
1.227 brouard 9726: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9727: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9728: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9729:
1.227 brouard 9730: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9731: /* oldm=oldms;savm=savms; */
9732: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9733:
1.227 brouard 9734: /* for (h=0; h<=nhstepm; h++){ */
9735: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9736: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9737: /* } */
9738: /* for(j=1; j<=nlstate+ndeath;j++) { */
9739: /* kk1=0.;kk2=0; */
9740: /* for(i=1; i<=nlstate;i++) { */
9741: /* if (mobilav==1) */
9742: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9743: /* else { */
9744: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9745: /* } */
9746: /* } */
9747: /* if (h==(int)(calagedatem+12*cpt)){ */
9748: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9749: /* /\*fprintf(ficrespop," %.3f", kk1); */
9750: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9751: /* } */
9752: /* } */
9753: /* for(i=1; i<=nlstate;i++){ */
9754: /* kk1=0.; */
9755: /* for(j=1; j<=nlstate;j++){ */
9756: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9757: /* } */
9758: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9759: /* } */
1.218 brouard 9760:
1.227 brouard 9761: /* if (h==(int)(calagedatem+12*cpt)) */
9762: /* for(j=1; j<=nlstate;j++) */
9763: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9764: /* } */
9765: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9766: /* } */
9767: /* } */
1.218 brouard 9768:
1.227 brouard 9769: /* /\******\/ */
1.218 brouard 9770:
1.227 brouard 9771: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9772: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9773: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9774: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9775: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9776:
1.227 brouard 9777: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9778: /* oldm=oldms;savm=savms; */
9779: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9780: /* for (h=0; h<=nhstepm; h++){ */
9781: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9782: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9783: /* } */
9784: /* for(j=1; j<=nlstate+ndeath;j++) { */
9785: /* kk1=0.;kk2=0; */
9786: /* for(i=1; i<=nlstate;i++) { */
9787: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9788: /* } */
9789: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9790: /* } */
9791: /* } */
9792: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9793: /* } */
9794: /* } */
9795: /* } */
9796: /* } */
1.218 brouard 9797:
1.227 brouard 9798: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9799:
1.227 brouard 9800: /* if (popforecast==1) { */
9801: /* free_ivector(popage,0,AGESUP); */
9802: /* free_vector(popeffectif,0,AGESUP); */
9803: /* free_vector(popcount,0,AGESUP); */
9804: /* } */
9805: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9806: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9807: /* fclose(ficrespop); */
9808: /* } /\* End of popforecast *\/ */
1.218 brouard 9809:
1.126 brouard 9810: int fileappend(FILE *fichier, char *optionfich)
9811: {
9812: if((fichier=fopen(optionfich,"a"))==NULL) {
9813: printf("Problem with file: %s\n", optionfich);
9814: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9815: return (0);
9816: }
9817: fflush(fichier);
9818: return (1);
9819: }
9820:
9821:
9822: /**************** function prwizard **********************/
9823: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9824: {
9825:
9826: /* Wizard to print covariance matrix template */
9827:
1.164 brouard 9828: char ca[32], cb[32];
9829: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9830: int numlinepar;
9831:
9832: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9833: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9834: for(i=1; i <=nlstate; i++){
9835: jj=0;
9836: for(j=1; j <=nlstate+ndeath; j++){
9837: if(j==i) continue;
9838: jj++;
9839: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9840: printf("%1d%1d",i,j);
9841: fprintf(ficparo,"%1d%1d",i,j);
9842: for(k=1; k<=ncovmodel;k++){
9843: /* printf(" %lf",param[i][j][k]); */
9844: /* fprintf(ficparo," %lf",param[i][j][k]); */
9845: printf(" 0.");
9846: fprintf(ficparo," 0.");
9847: }
9848: printf("\n");
9849: fprintf(ficparo,"\n");
9850: }
9851: }
9852: printf("# Scales (for hessian or gradient estimation)\n");
9853: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9854: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9855: for(i=1; i <=nlstate; i++){
9856: jj=0;
9857: for(j=1; j <=nlstate+ndeath; j++){
9858: if(j==i) continue;
9859: jj++;
9860: fprintf(ficparo,"%1d%1d",i,j);
9861: printf("%1d%1d",i,j);
9862: fflush(stdout);
9863: for(k=1; k<=ncovmodel;k++){
9864: /* printf(" %le",delti3[i][j][k]); */
9865: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9866: printf(" 0.");
9867: fprintf(ficparo," 0.");
9868: }
9869: numlinepar++;
9870: printf("\n");
9871: fprintf(ficparo,"\n");
9872: }
9873: }
9874: printf("# Covariance matrix\n");
9875: /* # 121 Var(a12)\n\ */
9876: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9877: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9878: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9879: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9880: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9881: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9882: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9883: fflush(stdout);
9884: fprintf(ficparo,"# Covariance matrix\n");
9885: /* # 121 Var(a12)\n\ */
9886: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9887: /* # ...\n\ */
9888: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9889:
9890: for(itimes=1;itimes<=2;itimes++){
9891: jj=0;
9892: for(i=1; i <=nlstate; i++){
9893: for(j=1; j <=nlstate+ndeath; j++){
9894: if(j==i) continue;
9895: for(k=1; k<=ncovmodel;k++){
9896: jj++;
9897: ca[0]= k+'a'-1;ca[1]='\0';
9898: if(itimes==1){
9899: printf("#%1d%1d%d",i,j,k);
9900: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9901: }else{
9902: printf("%1d%1d%d",i,j,k);
9903: fprintf(ficparo,"%1d%1d%d",i,j,k);
9904: /* printf(" %.5le",matcov[i][j]); */
9905: }
9906: ll=0;
9907: for(li=1;li <=nlstate; li++){
9908: for(lj=1;lj <=nlstate+ndeath; lj++){
9909: if(lj==li) continue;
9910: for(lk=1;lk<=ncovmodel;lk++){
9911: ll++;
9912: if(ll<=jj){
9913: cb[0]= lk +'a'-1;cb[1]='\0';
9914: if(ll<jj){
9915: if(itimes==1){
9916: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9917: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9918: }else{
9919: printf(" 0.");
9920: fprintf(ficparo," 0.");
9921: }
9922: }else{
9923: if(itimes==1){
9924: printf(" Var(%s%1d%1d)",ca,i,j);
9925: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9926: }else{
9927: printf(" 0.");
9928: fprintf(ficparo," 0.");
9929: }
9930: }
9931: }
9932: } /* end lk */
9933: } /* end lj */
9934: } /* end li */
9935: printf("\n");
9936: fprintf(ficparo,"\n");
9937: numlinepar++;
9938: } /* end k*/
9939: } /*end j */
9940: } /* end i */
9941: } /* end itimes */
9942:
9943: } /* end of prwizard */
9944: /******************* Gompertz Likelihood ******************************/
9945: double gompertz(double x[])
9946: {
1.302 brouard 9947: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9948: int i,n=0; /* n is the size of the sample */
9949:
1.220 brouard 9950: for (i=1;i<=imx ; i++) {
1.126 brouard 9951: sump=sump+weight[i];
9952: /* sump=sump+1;*/
9953: num=num+1;
9954: }
1.302 brouard 9955: L=0.0;
9956: /* agegomp=AGEGOMP; */
1.126 brouard 9957: /* for (i=0; i<=imx; i++)
9958: 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]);*/
9959:
1.302 brouard 9960: for (i=1;i<=imx ; i++) {
9961: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9962: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9963: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9964: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9965: * +
9966: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9967: */
9968: if (wav[i] > 1 || agedc[i] < AGESUP) {
9969: if (cens[i] == 1){
9970: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9971: } else if (cens[i] == 0){
1.126 brouard 9972: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9973: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9974: } else
9975: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9976: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9977: L=L+A*weight[i];
1.126 brouard 9978: /* 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 9979: }
9980: }
1.126 brouard 9981:
1.302 brouard 9982: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9983:
9984: return -2*L*num/sump;
9985: }
9986:
1.136 brouard 9987: #ifdef GSL
9988: /******************* Gompertz_f Likelihood ******************************/
9989: double gompertz_f(const gsl_vector *v, void *params)
9990: {
1.302 brouard 9991: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9992: double *x= (double *) v->data;
9993: int i,n=0; /* n is the size of the sample */
9994:
9995: for (i=0;i<=imx-1 ; i++) {
9996: sump=sump+weight[i];
9997: /* sump=sump+1;*/
9998: num=num+1;
9999: }
10000:
10001:
10002: /* for (i=0; i<=imx; i++)
10003: 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]);*/
10004: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
10005: for (i=1;i<=imx ; i++)
10006: {
10007: if (cens[i] == 1 && wav[i]>1)
10008: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
10009:
10010: if (cens[i] == 0 && wav[i]>1)
10011: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
10012: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
10013:
10014: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
10015: if (wav[i] > 1 ) { /* ??? */
10016: LL=LL+A*weight[i];
10017: /* 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]);*/
10018: }
10019: }
10020:
10021: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
10022: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
10023:
10024: return -2*LL*num/sump;
10025: }
10026: #endif
10027:
1.126 brouard 10028: /******************* Printing html file ***********/
1.201 brouard 10029: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 10030: int lastpass, int stepm, int weightopt, char model[],\
10031: int imx, double p[],double **matcov,double agemortsup){
10032: int i,k;
10033:
10034: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
10035: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
10036: for (i=1;i<=2;i++)
10037: 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 10038: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 10039: fprintf(fichtm,"</ul>");
10040:
10041: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
10042:
10043: 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>");
10044:
10045: for (k=agegomp;k<(agemortsup-2);k++)
10046: 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]);
10047:
10048:
10049: fflush(fichtm);
10050: }
10051:
10052: /******************* Gnuplot file **************/
1.201 brouard 10053: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 10054:
10055: char dirfileres[132],optfileres[132];
1.164 brouard 10056:
1.126 brouard 10057: int ng;
10058:
10059:
10060: /*#ifdef windows */
10061: fprintf(ficgp,"cd \"%s\" \n",pathc);
10062: /*#endif */
10063:
10064:
10065: strcpy(dirfileres,optionfilefiname);
10066: strcpy(optfileres,"vpl");
1.199 brouard 10067: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 10068: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 10069: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 10070: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 10071: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
10072:
10073: }
10074:
1.136 brouard 10075: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
10076: {
1.126 brouard 10077:
1.136 brouard 10078: /*-------- data file ----------*/
10079: FILE *fic;
10080: char dummy[]=" ";
1.240 brouard 10081: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 10082: int lstra;
1.136 brouard 10083: int linei, month, year,iout;
1.302 brouard 10084: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 10085: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 10086: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 10087: char *stratrunc;
1.223 brouard 10088:
1.240 brouard 10089: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
10090: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 10091: for(v=1;v<NCOVMAX;v++){
10092: DummyV[v]=0;
10093: FixedV[v]=0;
10094: }
1.126 brouard 10095:
1.240 brouard 10096: for(v=1; v <=ncovcol;v++){
10097: DummyV[v]=0;
10098: FixedV[v]=0;
10099: }
10100: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
10101: DummyV[v]=1;
10102: FixedV[v]=0;
10103: }
10104: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
10105: DummyV[v]=0;
10106: FixedV[v]=1;
10107: }
10108: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
10109: DummyV[v]=1;
10110: FixedV[v]=1;
10111: }
10112: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
10113: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10114: 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]);
10115: }
1.126 brouard 10116:
1.136 brouard 10117: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 10118: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10119: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 10120: }
1.126 brouard 10121:
1.302 brouard 10122: /* Is it a BOM UTF-8 Windows file? */
10123: /* First data line */
10124: linei=0;
10125: while(fgets(line, MAXLINE, fic)) {
10126: noffset=0;
10127: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10128: {
10129: noffset=noffset+3;
10130: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
10131: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
10132: fflush(ficlog); return 1;
10133: }
10134: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
10135: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
10136: {
10137: noffset=noffset+2;
1.304 brouard 10138: 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);
10139: 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 10140: fflush(ficlog); return 1;
10141: }
10142: else if( line[0] == 0 && line[1] == 0)
10143: {
10144: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10145: noffset=noffset+4;
1.304 brouard 10146: 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);
10147: 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 10148: fflush(ficlog); return 1;
10149: }
10150: } else{
10151: ;/*printf(" Not a BOM file\n");*/
10152: }
10153: /* If line starts with a # it is a comment */
10154: if (line[noffset] == '#') {
10155: linei=linei+1;
10156: break;
10157: }else{
10158: break;
10159: }
10160: }
10161: fclose(fic);
10162: if((fic=fopen(datafile,"r"))==NULL) {
10163: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10164: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
10165: }
10166: /* Not a Bom file */
10167:
1.136 brouard 10168: i=1;
10169: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
10170: linei=linei+1;
10171: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
10172: if(line[j] == '\t')
10173: line[j] = ' ';
10174: }
10175: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
10176: ;
10177: };
10178: line[j+1]=0; /* Trims blanks at end of line */
10179: if(line[0]=='#'){
10180: fprintf(ficlog,"Comment line\n%s\n",line);
10181: printf("Comment line\n%s\n",line);
10182: continue;
10183: }
10184: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 10185: strcpy(line, linetmp);
1.223 brouard 10186:
10187: /* Loops on waves */
10188: for (j=maxwav;j>=1;j--){
10189: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 10190: cutv(stra, strb, line, ' ');
10191: if(strb[0]=='.') { /* Missing value */
10192: lval=-1;
10193: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
10194: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
10195: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
10196: 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);
10197: 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);
10198: return 1;
10199: }
10200: }else{
10201: errno=0;
10202: /* what_kind_of_number(strb); */
10203: dval=strtod(strb,&endptr);
10204: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
10205: /* if(strb != endptr && *endptr == '\0') */
10206: /* dval=dlval; */
10207: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10208: if( strb[0]=='\0' || (*endptr != '\0')){
10209: 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);
10210: 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);
10211: return 1;
10212: }
10213: cotqvar[j][iv][i]=dval;
10214: cotvar[j][ntv+iv][i]=dval;
10215: }
10216: strcpy(line,stra);
1.223 brouard 10217: }/* end loop ntqv */
1.225 brouard 10218:
1.223 brouard 10219: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 10220: cutv(stra, strb, line, ' ');
10221: if(strb[0]=='.') { /* Missing value */
10222: lval=-1;
10223: }else{
10224: errno=0;
10225: lval=strtol(strb,&endptr,10);
10226: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10227: if( strb[0]=='\0' || (*endptr != '\0')){
10228: 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);
10229: 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);
10230: return 1;
10231: }
10232: }
10233: if(lval <-1 || lval >1){
10234: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10235: 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 10236: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10237: For example, for multinomial values like 1, 2 and 3,\n \
10238: build V1=0 V2=0 for the reference value (1),\n \
10239: V1=1 V2=0 for (2) \n \
1.223 brouard 10240: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10241: output of IMaCh is often meaningless.\n \
1.319 brouard 10242: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 10243: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10244: 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 10245: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10246: For example, for multinomial values like 1, 2 and 3,\n \
10247: build V1=0 V2=0 for the reference value (1),\n \
10248: V1=1 V2=0 for (2) \n \
1.223 brouard 10249: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10250: output of IMaCh is often meaningless.\n \
1.319 brouard 10251: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 10252: return 1;
10253: }
10254: cotvar[j][iv][i]=(double)(lval);
10255: strcpy(line,stra);
1.223 brouard 10256: }/* end loop ntv */
1.225 brouard 10257:
1.223 brouard 10258: /* Statuses at wave */
1.137 brouard 10259: cutv(stra, strb, line, ' ');
1.223 brouard 10260: if(strb[0]=='.') { /* Missing value */
1.238 brouard 10261: lval=-1;
1.136 brouard 10262: }else{
1.238 brouard 10263: errno=0;
10264: lval=strtol(strb,&endptr,10);
10265: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10266: if( strb[0]=='\0' || (*endptr != '\0')){
10267: 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);
10268: 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);
10269: return 1;
10270: }
1.136 brouard 10271: }
1.225 brouard 10272:
1.136 brouard 10273: s[j][i]=lval;
1.225 brouard 10274:
1.223 brouard 10275: /* Date of Interview */
1.136 brouard 10276: strcpy(line,stra);
10277: cutv(stra, strb,line,' ');
1.169 brouard 10278: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10279: }
1.169 brouard 10280: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 10281: month=99;
10282: year=9999;
1.136 brouard 10283: }else{
1.225 brouard 10284: 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);
10285: 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);
10286: return 1;
1.136 brouard 10287: }
10288: anint[j][i]= (double) year;
1.302 brouard 10289: mint[j][i]= (double)month;
10290: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
10291: /* 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]); */
10292: /* 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]); */
10293: /* } */
1.136 brouard 10294: strcpy(line,stra);
1.223 brouard 10295: } /* End loop on waves */
1.225 brouard 10296:
1.223 brouard 10297: /* Date of death */
1.136 brouard 10298: cutv(stra, strb,line,' ');
1.169 brouard 10299: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10300: }
1.169 brouard 10301: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 10302: month=99;
10303: year=9999;
10304: }else{
1.141 brouard 10305: 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 10306: 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);
10307: return 1;
1.136 brouard 10308: }
10309: andc[i]=(double) year;
10310: moisdc[i]=(double) month;
10311: strcpy(line,stra);
10312:
1.223 brouard 10313: /* Date of birth */
1.136 brouard 10314: cutv(stra, strb,line,' ');
1.169 brouard 10315: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10316: }
1.169 brouard 10317: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 10318: month=99;
10319: year=9999;
10320: }else{
1.141 brouard 10321: 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);
10322: 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 10323: return 1;
1.136 brouard 10324: }
10325: if (year==9999) {
1.141 brouard 10326: 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);
10327: 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 10328: return 1;
10329:
1.136 brouard 10330: }
10331: annais[i]=(double)(year);
1.302 brouard 10332: moisnais[i]=(double)(month);
10333: for (j=1;j<=maxwav;j++){
10334: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
10335: 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]);
10336: 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]);
10337: }
10338: }
10339:
1.136 brouard 10340: strcpy(line,stra);
1.225 brouard 10341:
1.223 brouard 10342: /* Sample weight */
1.136 brouard 10343: cutv(stra, strb,line,' ');
10344: errno=0;
10345: dval=strtod(strb,&endptr);
10346: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 10347: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
10348: 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 10349: fflush(ficlog);
10350: return 1;
10351: }
10352: weight[i]=dval;
10353: strcpy(line,stra);
1.225 brouard 10354:
1.223 brouard 10355: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10356: cutv(stra, strb, line, ' ');
10357: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10358: lval=-1;
1.311 brouard 10359: coqvar[iv][i]=NAN;
10360: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10361: }else{
1.225 brouard 10362: errno=0;
10363: /* what_kind_of_number(strb); */
10364: dval=strtod(strb,&endptr);
10365: /* if(strb != endptr && *endptr == '\0') */
10366: /* dval=dlval; */
10367: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10368: if( strb[0]=='\0' || (*endptr != '\0')){
10369: 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);
10370: 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);
10371: return 1;
10372: }
10373: coqvar[iv][i]=dval;
1.226 brouard 10374: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10375: }
10376: strcpy(line,stra);
10377: }/* end loop nqv */
1.136 brouard 10378:
1.223 brouard 10379: /* Covariate values */
1.136 brouard 10380: for (j=ncovcol;j>=1;j--){
10381: cutv(stra, strb,line,' ');
1.223 brouard 10382: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10383: lval=-1;
1.136 brouard 10384: }else{
1.225 brouard 10385: errno=0;
10386: lval=strtol(strb,&endptr,10);
10387: if( strb[0]=='\0' || (*endptr != '\0')){
10388: 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);
10389: 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);
10390: return 1;
10391: }
1.136 brouard 10392: }
10393: if(lval <-1 || lval >1){
1.225 brouard 10394: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10395: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10396: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10397: For example, for multinomial values like 1, 2 and 3,\n \
10398: build V1=0 V2=0 for the reference value (1),\n \
10399: V1=1 V2=0 for (2) \n \
1.136 brouard 10400: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10401: output of IMaCh is often meaningless.\n \
1.136 brouard 10402: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10403: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10404: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10405: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10406: For example, for multinomial values like 1, 2 and 3,\n \
10407: build V1=0 V2=0 for the reference value (1),\n \
10408: V1=1 V2=0 for (2) \n \
1.136 brouard 10409: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10410: output of IMaCh is often meaningless.\n \
1.136 brouard 10411: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10412: return 1;
1.136 brouard 10413: }
10414: covar[j][i]=(double)(lval);
10415: strcpy(line,stra);
10416: }
10417: lstra=strlen(stra);
1.225 brouard 10418:
1.136 brouard 10419: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10420: stratrunc = &(stra[lstra-9]);
10421: num[i]=atol(stratrunc);
10422: }
10423: else
10424: num[i]=atol(stra);
10425: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10426: 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;}*/
10427:
10428: i=i+1;
10429: } /* End loop reading data */
1.225 brouard 10430:
1.136 brouard 10431: *imax=i-1; /* Number of individuals */
10432: fclose(fic);
1.225 brouard 10433:
1.136 brouard 10434: return (0);
1.164 brouard 10435: /* endread: */
1.225 brouard 10436: printf("Exiting readdata: ");
10437: fclose(fic);
10438: return (1);
1.223 brouard 10439: }
1.126 brouard 10440:
1.234 brouard 10441: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10442: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10443: while (*p2 == ' ')
1.234 brouard 10444: p2++;
10445: /* while ((*p1++ = *p2++) !=0) */
10446: /* ; */
10447: /* do */
10448: /* while (*p2 == ' ') */
10449: /* p2++; */
10450: /* while (*p1++ == *p2++); */
10451: *stri=p2;
1.145 brouard 10452: }
10453:
1.330 brouard 10454: int decoderesult( char resultline[], int nres)
1.230 brouard 10455: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10456: {
1.235 brouard 10457: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10458: char resultsav[MAXLINE];
1.330 brouard 10459: /* int resultmodel[MAXLINE]; */
1.334 brouard 10460: /* int modelresult[MAXLINE]; */
1.230 brouard 10461: char stra[80], strb[80], strc[80], strd[80],stre[80];
10462:
1.234 brouard 10463: removefirstspace(&resultline);
1.332 brouard 10464: printf("decoderesult:%s\n",resultline);
1.230 brouard 10465:
1.332 brouard 10466: strcpy(resultsav,resultline);
10467: printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline);
1.230 brouard 10468: if (strlen(resultsav) >1){
1.334 brouard 10469: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 10470: }
1.253 brouard 10471: if(j == 0){ /* Resultline but no = */
10472: TKresult[nres]=0; /* Combination for the nresult and the model */
10473: return (0);
10474: }
1.234 brouard 10475: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.334 brouard 10476: 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);
10477: 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 10478: /* return 1;*/
1.234 brouard 10479: }
1.334 brouard 10480: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 10481: if(nbocc(resultsav,'=') >1){
1.318 brouard 10482: 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 10483: /* 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 10484: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 10485: /* If a blank, then strc="V4=" and strd='\0' */
10486: if(strc[0]=='\0'){
10487: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
10488: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
10489: return 1;
10490: }
1.234 brouard 10491: }else
10492: cutl(strc,strd,resultsav,'=');
1.318 brouard 10493: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10494:
1.230 brouard 10495: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10496: 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 10497: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10498: /* cptcovsel++; */
10499: if (nbocc(stra,'=') >0)
10500: strcpy(resultsav,stra); /* and analyzes it */
10501: }
1.235 brouard 10502: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10503: /* 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 10504: 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 10505: if(Typevar[k1]==0){ /* Single covariate in model */
10506: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 10507: match=0;
1.318 brouard 10508: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10509: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10510: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10511: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10512: break;
10513: }
10514: }
10515: if(match == 0){
1.332 brouard 10516: 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]);
10517: 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 10518: return 1;
1.234 brouard 10519: }
1.332 brouard 10520: }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*/
10521: /* We feed resultmodel[k1]=k2; */
10522: match=0;
10523: 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 */
10524: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10525: 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 10526: resultmodel[nres][k1]=k2; /* Added here */
10527: printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]);
10528: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10529: break;
10530: }
10531: }
10532: if(match == 0){
10533: 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 10534: 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 10535: return 1;
10536: }
10537: }else if(Typevar[k1]==2){ /* Product No age We want to get the position in the resultline of the product in the model line*/
10538: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
10539: match=0;
10540: 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]);
10541: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10542: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10543: /* modelresult[k2]=k1; */
10544: printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]);
10545: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10546: }
10547: }
10548: if(match == 0){
10549: 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 10550: 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 10551: return 1;
10552: }
10553: match=0;
10554: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10555: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10556: /* modelresult[k2]=k1;*/
10557: printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]);
10558: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10559: break;
10560: }
10561: }
10562: if(match == 0){
10563: 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 10564: 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 10565: return 1;
10566: }
10567: }/* End of testing */
1.333 brouard 10568: }/* End loop cptcovt */
1.235 brouard 10569: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10570: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 10571: 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)
10572: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 10573: match=0;
1.318 brouard 10574: 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 10575: if(Typevar[k1]==0){ /* Single only */
1.237 brouard 10576: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 brouard 10577: 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 10578: 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 10579: ++match;
10580: }
10581: }
10582: }
10583: if(match == 0){
1.332 brouard 10584: printf("Error in result line: variable V%d is missing in model; result: %s, model=%s\n",Tvarsel[k2], resultline, model);
10585: 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 10586: return 1;
1.234 brouard 10587: }else if(match > 1){
10588: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 10589: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
10590: return 1;
1.234 brouard 10591: }
10592: }
1.334 brouard 10593: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 10594: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10595: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 10596: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10597: /* 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*/
10598: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 10599: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10600: /* 1 0 0 0 */
10601: /* 2 1 0 0 */
10602: /* 3 0 1 0 */
1.330 brouard 10603: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 10604: /* 5 0 0 1 */
1.330 brouard 10605: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 10606: /* 7 0 1 1 */
10607: /* 8 1 1 1 */
1.237 brouard 10608: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10609: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10610: /* V5*age V5 known which value for nres? */
10611: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 10612: 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.
10613: * loop on position k1 in the MODEL LINE */
1.331 brouard 10614: /* k counting number of combination of single dummies in the equation model */
10615: /* k4 counting single dummies in the equation model */
10616: /* k4q counting single quantitatives in the equation model */
1.334 brouard 10617: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Dummy and Single, k1 is sorting according to MODEL, but k3 to resultline */
10618: /* 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 10619: /* 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 10620: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 10621: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
10622: /* k3 is the position in the nres result line of the k1th variable of the model equation */
10623: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
10624: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
10625: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 10626: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 10627: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 10628: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 10629: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
10630: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
10631: 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 10632: 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 10633: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 10634: /* Tinvresult[nres][4]=1 */
1.334 brouard 10635: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
10636: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
10637: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10638: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 10639: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 10640: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.332 brouard 10641: 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 10642: k4++;;
1.331 brouard 10643: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 10644: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 10645: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 10646: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 10647: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
10648: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
10649: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 10650: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
10651: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10652: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
10653: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10654: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10655: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 10656: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 10657: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 10658: precov[nres][k1]=Tvalsel[k3q];
10659: 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 10660: k4q++;;
1.331 brouard 10661: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
10662: /* Tvar[k1]; */ /* Age variable */
1.332 brouard 10663: /* Wrong we want the value of variable name Tvar[k1] */
10664:
10665: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
1.331 brouard 10666: 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 10667: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
1.332 brouard 10668: precov[nres][k1]=Tvalsel[k3];
10669: 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 10670: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.332 brouard 10671: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.331 brouard 10672: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 10673: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
1.332 brouard 10674: precov[nres][k1]=Tvalsel[k3q];
1.334 brouard 10675: 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 10676: }else if(Typevar[k1]==2 ){ /* For product quant or dummy (not with age) */
1.332 brouard 10677: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
10678: 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 10679: }else{
1.332 brouard 10680: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
10681: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 10682: }
10683: }
1.234 brouard 10684:
1.334 brouard 10685: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 10686: return (0);
10687: }
1.235 brouard 10688:
1.230 brouard 10689: int decodemodel( char model[], int lastobs)
10690: /**< This routine decodes the model and returns:
1.224 brouard 10691: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10692: * - nagesqr = 1 if age*age in the model, otherwise 0.
10693: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10694: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10695: * - cptcovage number of covariates with age*products =2
10696: * - cptcovs number of simple covariates
10697: * - 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
10698: * which is a new column after the 9 (ncovcol) variables.
1.319 brouard 10699: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10700: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10701: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10702: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10703: */
1.319 brouard 10704: /* 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 10705: {
1.238 brouard 10706: int i, j, k, ks, v;
1.227 brouard 10707: int j1, k1, k2, k3, k4;
1.136 brouard 10708: char modelsav[80];
1.145 brouard 10709: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10710: char *strpt;
1.136 brouard 10711:
1.145 brouard 10712: /*removespace(model);*/
1.136 brouard 10713: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10714: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10715: if (strstr(model,"AGE") !=0){
1.192 brouard 10716: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10717: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10718: return 1;
10719: }
1.141 brouard 10720: if (strstr(model,"v") !=0){
10721: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
10722: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
10723: return 1;
10724: }
1.187 brouard 10725: strcpy(modelsav,model);
10726: if ((strpt=strstr(model,"age*age")) !=0){
10727: printf(" strpt=%s, model=%s\n",strpt, model);
10728: if(strpt != model){
1.234 brouard 10729: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10730: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10731: corresponding column of parameters.\n",model);
1.234 brouard 10732: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10733: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10734: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10735: return 1;
1.225 brouard 10736: }
1.187 brouard 10737: nagesqr=1;
10738: if (strstr(model,"+age*age") !=0)
1.234 brouard 10739: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10740: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10741: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10742: else
1.234 brouard 10743: substrchaine(modelsav, model, "age*age");
1.187 brouard 10744: }else
10745: nagesqr=0;
10746: if (strlen(modelsav) >1){
10747: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10748: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10749: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10750: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10751: * cst, age and age*age
10752: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10753: /* including age products which are counted in cptcovage.
10754: * but the covariates which are products must be treated
10755: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10756: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10757: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10758:
10759:
1.187 brouard 10760: /* Design
10761: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10762: * < ncovcol=8 >
10763: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10764: * k= 1 2 3 4 5 6 7 8
10765: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10766: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10767: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10768: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10769: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10770: * Tage[++cptcovage]=k
10771: * if products, new covar are created after ncovcol with k1
10772: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10773: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10774: * 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
10775: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10776: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10777: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10778: * < ncovcol=8 >
10779: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10780: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10781: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10782: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10783: * p Tprod[1]@2={ 6, 5}
10784: *p Tvard[1][1]@4= {7, 8, 5, 6}
10785: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10786: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10787: *How to reorganize? Tvars(orted)
1.187 brouard 10788: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10789: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
10790: * {2, 1, 4, 8, 5, 6, 3, 7}
10791: * Struct []
10792: */
1.225 brouard 10793:
1.187 brouard 10794: /* This loop fills the array Tvar from the string 'model'.*/
10795: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
10796: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
10797: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
10798: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
10799: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
10800: /* k=1 Tvar[1]=2 (from V2) */
10801: /* k=5 Tvar[5] */
10802: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 10803: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 10804: /* } */
1.198 brouard 10805: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 10806: /*
10807: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 10808: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
10809: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
10810: }
1.187 brouard 10811: cptcovage=0;
1.319 brouard 10812: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
10813: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
10814: 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" */
10815: if (nbocc(modelsav,'+')==0)
10816: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 10817: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
10818: /*scanf("%d",i);*/
1.319 brouard 10819: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
10820: 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 10821: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
10822: /* covar is not filled and then is empty */
10823: cptcovprod--;
10824: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 10825: 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 10826: Typevar[k]=1; /* 1 for age product */
1.319 brouard 10827: cptcovage++; /* Counts the number of covariates which include age as a product */
10828: 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 10829: /*printf("stre=%s ", stre);*/
10830: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
10831: cptcovprod--;
10832: cutl(stre,strb,strc,'V');
10833: Tvar[k]=atoi(stre);
10834: Typevar[k]=1; /* 1 for age product */
10835: cptcovage++;
10836: Tage[cptcovage]=k;
10837: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
10838: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
10839: cptcovn++;
10840: cptcovprodnoage++;k1++;
10841: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10842: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10843: because this model-covariate is a construction we invent a new column
10844: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.335 brouard 10845: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
1.319 brouard 10846: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
10847: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=4 etc */
1.335 brouard 10848: /* Please remark that the new variables are model dependent */
10849: /* If we have 4 variable but the model uses only 3, like in
10850: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
10851: * k= 1 2 3 4 5 6 7 8
10852: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
10853: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
10854: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
10855: */
1.234 brouard 10856: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10857: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10858: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 10859: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 10860: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 brouard 10861: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 10862: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 brouard 10863: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 10864: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10865: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10866: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10867: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10868: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10869: for (i=1; i<=lastobs;i++){
10870: /* Computes the new covariate which is a product of
10871: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10872: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10873: }
10874: } /* End age is not in the model */
10875: } /* End if model includes a product */
1.319 brouard 10876: else { /* not a product */
1.234 brouard 10877: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10878: /* scanf("%d",i);*/
10879: cutl(strd,strc,strb,'V');
10880: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10881: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10882: Tvar[k]=atoi(strd);
10883: Typevar[k]=0; /* 0 for simple covariates */
10884: }
10885: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10886: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10887: scanf("%d",i);*/
1.187 brouard 10888: } /* end of loop + on total covariates */
10889: } /* end if strlen(modelsave == 0) age*age might exist */
10890: } /* end if strlen(model == 0) */
1.136 brouard 10891:
10892: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10893: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10894:
1.136 brouard 10895: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10896: printf("cptcovprod=%d ", cptcovprod);
10897: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10898: scanf("%d ",i);*/
10899:
10900:
1.230 brouard 10901: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10902: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10903: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10904: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10905: k = 1 2 3 4 5 6 7 8 9
10906: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 10907: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 10908: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10909: Dummy[k] 1 0 0 0 3 1 1 2 3
10910: Tmodelind[combination of covar]=k;
1.225 brouard 10911: */
10912: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10913: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10914: /* 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 10915: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 10916: printf("Model=1+age+%s\n\
1.227 brouard 10917: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10918: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10919: 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 10920: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 10921: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10922: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10923: 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 10924: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10925: 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 */
10926: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10927: Fixed[k]= 0;
10928: Dummy[k]= 0;
1.225 brouard 10929: ncoveff++;
1.232 brouard 10930: ncovf++;
1.234 brouard 10931: nsd++;
10932: modell[k].maintype= FTYPE;
10933: TvarsD[nsd]=Tvar[k];
10934: TvarsDind[nsd]=k;
1.330 brouard 10935: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 10936: TvarF[ncovf]=Tvar[k];
10937: TvarFind[ncovf]=k;
10938: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10939: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10940: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10941: Fixed[k]= 0;
10942: Dummy[k]= 0;
10943: ncoveff++;
10944: ncovf++;
10945: modell[k].maintype= FTYPE;
10946: TvarF[ncovf]=Tvar[k];
1.330 brouard 10947: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 10948: TvarFind[ncovf]=k;
1.230 brouard 10949: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10950: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10951: }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 10952: Fixed[k]= 0;
10953: Dummy[k]= 1;
1.230 brouard 10954: nqfveff++;
1.234 brouard 10955: modell[k].maintype= FTYPE;
10956: modell[k].subtype= FQ;
10957: nsq++;
1.334 brouard 10958: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
10959: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 10960: ncovf++;
1.234 brouard 10961: TvarF[ncovf]=Tvar[k];
10962: TvarFind[ncovf]=k;
1.231 brouard 10963: 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 10964: 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 10965: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10966: Fixed[k]= 1;
10967: Dummy[k]= 0;
1.225 brouard 10968: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10969: modell[k].maintype= VTYPE;
10970: modell[k].subtype= VD;
10971: nsd++;
10972: TvarsD[nsd]=Tvar[k];
10973: TvarsDind[nsd]=k;
1.330 brouard 10974: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 10975: ncovv++; /* Only simple time varying variables */
10976: TvarV[ncovv]=Tvar[k];
1.242 brouard 10977: 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 10978: 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 */
10979: 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 10980: 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);
10981: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10982: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10983: Fixed[k]= 1;
10984: Dummy[k]= 1;
10985: nqtveff++;
10986: modell[k].maintype= VTYPE;
10987: modell[k].subtype= VQ;
10988: ncovv++; /* Only simple time varying variables */
10989: nsq++;
1.334 brouard 10990: 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) */
10991: 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 10992: TvarV[ncovv]=Tvar[k];
1.242 brouard 10993: 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 10994: 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 */
10995: 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 10996: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10997: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10998: 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 10999: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 11000: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 11001: ncova++;
11002: TvarA[ncova]=Tvar[k];
11003: TvarAind[ncova]=k;
1.231 brouard 11004: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 11005: Fixed[k]= 2;
11006: Dummy[k]= 2;
11007: modell[k].maintype= ATYPE;
11008: modell[k].subtype= APFD;
11009: /* ncoveff++; */
1.227 brouard 11010: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 11011: Fixed[k]= 2;
11012: Dummy[k]= 3;
11013: modell[k].maintype= ATYPE;
11014: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
11015: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 11016: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 11017: Fixed[k]= 3;
11018: Dummy[k]= 2;
11019: modell[k].maintype= ATYPE;
11020: modell[k].subtype= APVD; /* Product age * varying dummy */
11021: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 11022: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 11023: Fixed[k]= 3;
11024: Dummy[k]= 3;
11025: modell[k].maintype= ATYPE;
11026: modell[k].subtype= APVQ; /* Product age * varying quantitative */
11027: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 11028: }
11029: }else if (Typevar[k] == 2) { /* product without age */
11030: k1=Tposprod[k];
11031: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 11032: if(Tvard[k1][2] <=ncovcol){
11033: Fixed[k]= 1;
11034: Dummy[k]= 0;
11035: modell[k].maintype= FTYPE;
11036: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
11037: ncovf++; /* Fixed variables without age */
11038: TvarF[ncovf]=Tvar[k];
11039: TvarFind[ncovf]=k;
11040: }else if(Tvard[k1][2] <=ncovcol+nqv){
11041: Fixed[k]= 0; /* or 2 ?*/
11042: Dummy[k]= 1;
11043: modell[k].maintype= FTYPE;
11044: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
11045: ncovf++; /* Varying variables without age */
11046: TvarF[ncovf]=Tvar[k];
11047: TvarFind[ncovf]=k;
11048: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11049: Fixed[k]= 1;
11050: Dummy[k]= 0;
11051: modell[k].maintype= VTYPE;
11052: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
11053: ncovv++; /* Varying variables without age */
11054: TvarV[ncovv]=Tvar[k];
11055: TvarVind[ncovv]=k;
11056: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11057: Fixed[k]= 1;
11058: Dummy[k]= 1;
11059: modell[k].maintype= VTYPE;
11060: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
11061: ncovv++; /* Varying variables without age */
11062: TvarV[ncovv]=Tvar[k];
11063: TvarVind[ncovv]=k;
11064: }
1.227 brouard 11065: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 11066: if(Tvard[k1][2] <=ncovcol){
11067: Fixed[k]= 0; /* or 2 ?*/
11068: Dummy[k]= 1;
11069: modell[k].maintype= FTYPE;
11070: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
11071: ncovf++; /* Fixed variables without age */
11072: TvarF[ncovf]=Tvar[k];
11073: TvarFind[ncovf]=k;
11074: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11075: Fixed[k]= 1;
11076: Dummy[k]= 1;
11077: modell[k].maintype= VTYPE;
11078: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
11079: ncovv++; /* Varying variables without age */
11080: TvarV[ncovv]=Tvar[k];
11081: TvarVind[ncovv]=k;
11082: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11083: Fixed[k]= 1;
11084: Dummy[k]= 1;
11085: modell[k].maintype= VTYPE;
11086: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
11087: ncovv++; /* Varying variables without age */
11088: TvarV[ncovv]=Tvar[k];
11089: TvarVind[ncovv]=k;
11090: ncovv++; /* Varying variables without age */
11091: TvarV[ncovv]=Tvar[k];
11092: TvarVind[ncovv]=k;
11093: }
1.227 brouard 11094: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 11095: if(Tvard[k1][2] <=ncovcol){
11096: Fixed[k]= 1;
11097: Dummy[k]= 1;
11098: modell[k].maintype= VTYPE;
11099: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
11100: ncovv++; /* Varying variables without age */
11101: TvarV[ncovv]=Tvar[k];
11102: TvarVind[ncovv]=k;
11103: }else if(Tvard[k1][2] <=ncovcol+nqv){
11104: Fixed[k]= 1;
11105: Dummy[k]= 1;
11106: modell[k].maintype= VTYPE;
11107: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
11108: ncovv++; /* Varying variables without age */
11109: TvarV[ncovv]=Tvar[k];
11110: TvarVind[ncovv]=k;
11111: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11112: Fixed[k]= 1;
11113: Dummy[k]= 0;
11114: modell[k].maintype= VTYPE;
11115: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
11116: ncovv++; /* Varying variables without age */
11117: TvarV[ncovv]=Tvar[k];
11118: TvarVind[ncovv]=k;
11119: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11120: Fixed[k]= 1;
11121: Dummy[k]= 1;
11122: modell[k].maintype= VTYPE;
11123: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
11124: ncovv++; /* Varying variables without age */
11125: TvarV[ncovv]=Tvar[k];
11126: TvarVind[ncovv]=k;
11127: }
1.227 brouard 11128: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 11129: if(Tvard[k1][2] <=ncovcol){
11130: Fixed[k]= 1;
11131: Dummy[k]= 1;
11132: modell[k].maintype= VTYPE;
11133: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
11134: ncovv++; /* Varying variables without age */
11135: TvarV[ncovv]=Tvar[k];
11136: TvarVind[ncovv]=k;
11137: }else if(Tvard[k1][2] <=ncovcol+nqv){
11138: Fixed[k]= 1;
11139: Dummy[k]= 1;
11140: modell[k].maintype= VTYPE;
11141: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
11142: ncovv++; /* Varying variables without age */
11143: TvarV[ncovv]=Tvar[k];
11144: TvarVind[ncovv]=k;
11145: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11146: Fixed[k]= 1;
11147: Dummy[k]= 1;
11148: modell[k].maintype= VTYPE;
11149: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
11150: ncovv++; /* Varying variables without age */
11151: TvarV[ncovv]=Tvar[k];
11152: TvarVind[ncovv]=k;
11153: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11154: Fixed[k]= 1;
11155: Dummy[k]= 1;
11156: modell[k].maintype= VTYPE;
11157: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
11158: ncovv++; /* Varying variables without age */
11159: TvarV[ncovv]=Tvar[k];
11160: TvarVind[ncovv]=k;
11161: }
1.227 brouard 11162: }else{
1.240 brouard 11163: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11164: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11165: } /*end k1*/
1.225 brouard 11166: }else{
1.226 brouard 11167: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
11168: 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 11169: }
1.227 brouard 11170: 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 11171: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 11172: 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]);
11173: }
11174: /* Searching for doublons in the model */
11175: for(k1=1; k1<= cptcovt;k1++){
11176: for(k2=1; k2 <k1;k2++){
1.285 brouard 11177: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
11178: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 11179: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
11180: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 11181: 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]);
11182: 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 11183: return(1);
11184: }
11185: }else if (Typevar[k1] ==2){
11186: k3=Tposprod[k1];
11187: k4=Tposprod[k2];
11188: 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])) ){
11189: 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]]);
11190: 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);
11191: return(1);
11192: }
11193: }
1.227 brouard 11194: }
11195: }
1.225 brouard 11196: }
11197: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
11198: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 11199: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
11200: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 11201: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 11202: /*endread:*/
1.225 brouard 11203: printf("Exiting decodemodel: ");
11204: return (1);
1.136 brouard 11205: }
11206:
1.169 brouard 11207: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 11208: {/* Check ages at death */
1.136 brouard 11209: int i, m;
1.218 brouard 11210: int firstone=0;
11211:
1.136 brouard 11212: for (i=1; i<=imx; i++) {
11213: for(m=2; (m<= maxwav); m++) {
11214: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
11215: anint[m][i]=9999;
1.216 brouard 11216: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
11217: s[m][i]=-1;
1.136 brouard 11218: }
11219: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 11220: *nberr = *nberr + 1;
1.218 brouard 11221: if(firstone == 0){
11222: firstone=1;
1.260 brouard 11223: 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 11224: }
1.262 brouard 11225: 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 11226: s[m][i]=-1; /* Droping the death status */
1.136 brouard 11227: }
11228: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 11229: (*nberr)++;
1.259 brouard 11230: 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 11231: 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 11232: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 11233: }
11234: }
11235: }
11236:
11237: for (i=1; i<=imx; i++) {
11238: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
11239: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 11240: 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 11241: if (s[m][i] >= nlstate+1) {
1.169 brouard 11242: if(agedc[i]>0){
11243: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 11244: agev[m][i]=agedc[i];
1.214 brouard 11245: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 11246: }else {
1.136 brouard 11247: if ((int)andc[i]!=9999){
11248: nbwarn++;
11249: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
11250: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
11251: agev[m][i]=-1;
11252: }
11253: }
1.169 brouard 11254: } /* agedc > 0 */
1.214 brouard 11255: } /* end if */
1.136 brouard 11256: else if(s[m][i] !=9){ /* Standard case, age in fractional
11257: years but with the precision of a month */
11258: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
11259: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
11260: agev[m][i]=1;
11261: else if(agev[m][i] < *agemin){
11262: *agemin=agev[m][i];
11263: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
11264: }
11265: else if(agev[m][i] >*agemax){
11266: *agemax=agev[m][i];
1.156 brouard 11267: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 11268: }
11269: /*agev[m][i]=anint[m][i]-annais[i];*/
11270: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 11271: } /* en if 9*/
1.136 brouard 11272: else { /* =9 */
1.214 brouard 11273: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 11274: agev[m][i]=1;
11275: s[m][i]=-1;
11276: }
11277: }
1.214 brouard 11278: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 11279: agev[m][i]=1;
1.214 brouard 11280: else{
11281: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11282: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11283: agev[m][i]=0;
11284: }
11285: } /* End for lastpass */
11286: }
1.136 brouard 11287:
11288: for (i=1; i<=imx; i++) {
11289: for(m=firstpass; (m<=lastpass); m++){
11290: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 11291: (*nberr)++;
1.136 brouard 11292: 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);
11293: 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);
11294: return 1;
11295: }
11296: }
11297: }
11298:
11299: /*for (i=1; i<=imx; i++){
11300: for (m=firstpass; (m<lastpass); m++){
11301: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
11302: }
11303:
11304: }*/
11305:
11306:
1.139 brouard 11307: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
11308: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 11309:
11310: return (0);
1.164 brouard 11311: /* endread:*/
1.136 brouard 11312: printf("Exiting calandcheckages: ");
11313: return (1);
11314: }
11315:
1.172 brouard 11316: #if defined(_MSC_VER)
11317: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11318: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11319: //#include "stdafx.h"
11320: //#include <stdio.h>
11321: //#include <tchar.h>
11322: //#include <windows.h>
11323: //#include <iostream>
11324: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
11325:
11326: LPFN_ISWOW64PROCESS fnIsWow64Process;
11327:
11328: BOOL IsWow64()
11329: {
11330: BOOL bIsWow64 = FALSE;
11331:
11332: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
11333: // (HANDLE, PBOOL);
11334:
11335: //LPFN_ISWOW64PROCESS fnIsWow64Process;
11336:
11337: HMODULE module = GetModuleHandle(_T("kernel32"));
11338: const char funcName[] = "IsWow64Process";
11339: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
11340: GetProcAddress(module, funcName);
11341:
11342: if (NULL != fnIsWow64Process)
11343: {
11344: if (!fnIsWow64Process(GetCurrentProcess(),
11345: &bIsWow64))
11346: //throw std::exception("Unknown error");
11347: printf("Unknown error\n");
11348: }
11349: return bIsWow64 != FALSE;
11350: }
11351: #endif
1.177 brouard 11352:
1.191 brouard 11353: void syscompilerinfo(int logged)
1.292 brouard 11354: {
11355: #include <stdint.h>
11356:
11357: /* #include "syscompilerinfo.h"*/
1.185 brouard 11358: /* command line Intel compiler 32bit windows, XP compatible:*/
11359: /* /GS /W3 /Gy
11360: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
11361: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
11362: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 11363: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
11364: */
11365: /* 64 bits */
1.185 brouard 11366: /*
11367: /GS /W3 /Gy
11368: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
11369: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
11370: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
11371: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
11372: /* Optimization are useless and O3 is slower than O2 */
11373: /*
11374: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
11375: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
11376: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
11377: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
11378: */
1.186 brouard 11379: /* Link is */ /* /OUT:"visual studio
1.185 brouard 11380: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
11381: /PDB:"visual studio
11382: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
11383: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
11384: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
11385: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
11386: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
11387: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
11388: uiAccess='false'"
11389: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
11390: /NOLOGO /TLBID:1
11391: */
1.292 brouard 11392:
11393:
1.177 brouard 11394: #if defined __INTEL_COMPILER
1.178 brouard 11395: #if defined(__GNUC__)
11396: struct utsname sysInfo; /* For Intel on Linux and OS/X */
11397: #endif
1.177 brouard 11398: #elif defined(__GNUC__)
1.179 brouard 11399: #ifndef __APPLE__
1.174 brouard 11400: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 11401: #endif
1.177 brouard 11402: struct utsname sysInfo;
1.178 brouard 11403: int cross = CROSS;
11404: if (cross){
11405: printf("Cross-");
1.191 brouard 11406: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 11407: }
1.174 brouard 11408: #endif
11409:
1.191 brouard 11410: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 11411: #if defined(__clang__)
1.191 brouard 11412: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 11413: #endif
11414: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 11415: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 11416: #endif
11417: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 11418: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 11419: #endif
11420: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 11421: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 11422: #endif
11423: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 11424: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 11425: #endif
11426: #if defined(_MSC_VER)
1.191 brouard 11427: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 11428: #endif
11429: #if defined(__PGI)
1.191 brouard 11430: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 11431: #endif
11432: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 11433: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 11434: #endif
1.191 brouard 11435: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 11436:
1.167 brouard 11437: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
11438: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
11439: // Windows (x64 and x86)
1.191 brouard 11440: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 11441: #elif __unix__ // all unices, not all compilers
11442: // Unix
1.191 brouard 11443: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 11444: #elif __linux__
11445: // linux
1.191 brouard 11446: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 11447: #elif __APPLE__
1.174 brouard 11448: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 11449: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11450: #endif
11451:
11452: /* __MINGW32__ */
11453: /* __CYGWIN__ */
11454: /* __MINGW64__ */
11455: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11456: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11457: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11458: /* _WIN64 // Defined for applications for Win64. */
11459: /* _M_X64 // Defined for compilations that target x64 processors. */
11460: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11461:
1.167 brouard 11462: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11463: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11464: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11465: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11466: #else
1.191 brouard 11467: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11468: #endif
11469:
1.169 brouard 11470: #if defined(__GNUC__)
11471: # if defined(__GNUC_PATCHLEVEL__)
11472: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11473: + __GNUC_MINOR__ * 100 \
11474: + __GNUC_PATCHLEVEL__)
11475: # else
11476: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11477: + __GNUC_MINOR__ * 100)
11478: # endif
1.174 brouard 11479: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11480: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11481:
11482: if (uname(&sysInfo) != -1) {
11483: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11484: 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 11485: }
11486: else
11487: perror("uname() error");
1.179 brouard 11488: //#ifndef __INTEL_COMPILER
11489: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11490: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11491: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11492: #endif
1.169 brouard 11493: #endif
1.172 brouard 11494:
1.286 brouard 11495: // void main ()
1.172 brouard 11496: // {
1.169 brouard 11497: #if defined(_MSC_VER)
1.174 brouard 11498: if (IsWow64()){
1.191 brouard 11499: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11500: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11501: }
11502: else{
1.191 brouard 11503: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11504: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11505: }
1.172 brouard 11506: // printf("\nPress Enter to continue...");
11507: // getchar();
11508: // }
11509:
1.169 brouard 11510: #endif
11511:
1.167 brouard 11512:
1.219 brouard 11513: }
1.136 brouard 11514:
1.219 brouard 11515: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11516: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 11517: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 11518: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11519: /* double ftolpl = 1.e-10; */
1.180 brouard 11520: double age, agebase, agelim;
1.203 brouard 11521: double tot;
1.180 brouard 11522:
1.202 brouard 11523: strcpy(filerespl,"PL_");
11524: strcat(filerespl,fileresu);
11525: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11526: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11527: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11528: }
1.288 brouard 11529: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11530: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11531: pstamp(ficrespl);
1.288 brouard 11532: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11533: fprintf(ficrespl,"#Age ");
11534: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11535: fprintf(ficrespl,"\n");
1.180 brouard 11536:
1.219 brouard 11537: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11538:
1.219 brouard 11539: agebase=ageminpar;
11540: agelim=agemaxpar;
1.180 brouard 11541:
1.227 brouard 11542: /* i1=pow(2,ncoveff); */
1.234 brouard 11543: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11544: if (cptcovn < 1){i1=1;}
1.180 brouard 11545:
1.337 ! brouard 11546: /* for(k=1; k<=i1;k++){ /\* For each combination k of dummy covariates in the model *\/ */
1.238 brouard 11547: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 11548: k=TKresult[nres];
! 11549: /* if(i1 != 1 && TKresult[nres]!= k) /\* We found the combination k corresponding to the resultline value of dummies *\/ */
! 11550: /* continue; */
1.235 brouard 11551:
1.238 brouard 11552: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11553: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11554: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11555: /* k=k+1; */
11556: /* to clean */
1.332 brouard 11557: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11558: fprintf(ficrespl,"#******");
11559: printf("#******");
11560: fprintf(ficlog,"#******");
1.337 ! brouard 11561: for(j=1;j<=cptcovs ;j++) {/**< cptcovs number of SIMPLE covariates in the model or resultline V2+V1 =2 (dummy or quantit or time varying) */
1.332 brouard 11562: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
1.337 ! brouard 11563: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 11564: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 11565: fprintf(ficrespl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
! 11566: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
! 11567: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
! 11568: }
! 11569: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 11570: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 11571: /* fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 11572: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 11573: /* } */
1.238 brouard 11574: fprintf(ficrespl,"******\n");
11575: printf("******\n");
11576: fprintf(ficlog,"******\n");
11577: if(invalidvarcomb[k]){
11578: printf("\nCombination (%d) ignored because no case \n",k);
11579: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
11580: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
11581: continue;
11582: }
1.219 brouard 11583:
1.238 brouard 11584: fprintf(ficrespl,"#Age ");
1.337 ! brouard 11585: /* for(j=1;j<=cptcoveff;j++) { */
! 11586: /* fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 11587: /* } */
! 11588: for(j=1;j<=cptcovs;j++) { /* New the quanti variable is added */
! 11589: fprintf(ficrespl,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11590: }
11591: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
11592: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 11593:
1.238 brouard 11594: for (age=agebase; age<=agelim; age++){
11595: /* for (age=agebase; age<=agebase; age++){ */
1.337 ! brouard 11596: /**< Computes the prevalence limit in each live state at age x and for covariate combination (k and) nres */
! 11597: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres); /* Nicely done */
1.238 brouard 11598: fprintf(ficrespl,"%.0f ",age );
1.337 ! brouard 11599: /* for(j=1;j<=cptcoveff;j++) */
! 11600: /* fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 11601: for(j=1;j<=cptcovs;j++)
! 11602: fprintf(ficrespl,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11603: tot=0.;
11604: for(i=1; i<=nlstate;i++){
11605: tot += prlim[i][i];
11606: fprintf(ficrespl," %.5f", prlim[i][i]);
11607: }
11608: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
11609: } /* Age */
11610: /* was end of cptcod */
1.337 ! brouard 11611: } /* nres */
! 11612: /* } /\* for each combination *\/ */
1.219 brouard 11613: return 0;
1.180 brouard 11614: }
11615:
1.218 brouard 11616: 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 11617: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 11618:
11619: /* Computes the back prevalence limit for any combination of covariate values
11620: * at any age between ageminpar and agemaxpar
11621: */
1.235 brouard 11622: int i, j, k, i1, nres=0 ;
1.217 brouard 11623: /* double ftolpl = 1.e-10; */
11624: double age, agebase, agelim;
11625: double tot;
1.218 brouard 11626: /* double ***mobaverage; */
11627: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 11628:
11629: strcpy(fileresplb,"PLB_");
11630: strcat(fileresplb,fileresu);
11631: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 11632: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
11633: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 11634: }
1.288 brouard 11635: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
11636: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 11637: pstamp(ficresplb);
1.288 brouard 11638: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 11639: fprintf(ficresplb,"#Age ");
11640: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
11641: fprintf(ficresplb,"\n");
11642:
1.218 brouard 11643:
11644: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
11645:
11646: agebase=ageminpar;
11647: agelim=agemaxpar;
11648:
11649:
1.227 brouard 11650: i1=pow(2,cptcoveff);
1.218 brouard 11651: if (cptcovn < 1){i1=1;}
1.227 brouard 11652:
1.238 brouard 11653: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11654: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11655: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11656: continue;
1.332 brouard 11657: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11658: fprintf(ficresplb,"#******");
11659: printf("#******");
11660: fprintf(ficlog,"#******");
11661: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
1.332 brouard 11662: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11663: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11664: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11665: }
11666: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 11667: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
11668: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
11669: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 11670: }
11671: fprintf(ficresplb,"******\n");
11672: printf("******\n");
11673: fprintf(ficlog,"******\n");
11674: if(invalidvarcomb[k]){
11675: printf("\nCombination (%d) ignored because no cases \n",k);
11676: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
11677: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
11678: continue;
11679: }
1.218 brouard 11680:
1.238 brouard 11681: fprintf(ficresplb,"#Age ");
11682: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 11683: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11684: }
11685: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
11686: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 11687:
11688:
1.238 brouard 11689: for (age=agebase; age<=agelim; age++){
11690: /* for (age=agebase; age<=agebase; age++){ */
11691: if(mobilavproj > 0){
11692: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11693: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11694: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11695: }else if (mobilavproj == 0){
11696: 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);
11697: 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);
11698: exit(1);
11699: }else{
11700: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11701: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11702: /* printf("TOTOT\n"); */
11703: /* exit(1); */
1.238 brouard 11704: }
11705: fprintf(ficresplb,"%.0f ",age );
11706: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11707: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11708: tot=0.;
11709: for(i=1; i<=nlstate;i++){
11710: tot += bprlim[i][i];
11711: fprintf(ficresplb," %.5f", bprlim[i][i]);
11712: }
11713: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11714: } /* Age */
11715: /* was end of cptcod */
1.255 brouard 11716: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 11717: } /* end of any combination */
11718: } /* end of nres */
1.218 brouard 11719: /* hBijx(p, bage, fage); */
11720: /* fclose(ficrespijb); */
11721:
11722: return 0;
1.217 brouard 11723: }
1.218 brouard 11724:
1.180 brouard 11725: int hPijx(double *p, int bage, int fage){
11726: /*------------- h Pij x at various ages ------------*/
1.336 brouard 11727: /* to be optimized with precov */
1.180 brouard 11728: int stepsize;
11729: int agelim;
11730: int hstepm;
11731: int nhstepm;
1.235 brouard 11732: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11733:
11734: double agedeb;
11735: double ***p3mat;
11736:
1.337 ! brouard 11737: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
! 11738: if((ficrespij=fopen(filerespij,"w"))==NULL) {
! 11739: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
! 11740: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
! 11741: }
! 11742: printf("Computing pij: result on file '%s' \n", filerespij);
! 11743: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
! 11744:
! 11745: stepsize=(int) (stepm+YEARM-1)/YEARM;
! 11746: /*if (stepm<=24) stepsize=2;*/
! 11747:
! 11748: agelim=AGESUP;
! 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(ficrespij);
! 11754: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
! 11755: i1= pow(2,cptcoveff);
! 11756: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
! 11757: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
! 11758: /* k=k+1; */
! 11759: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
! 11760: k=TKresult[nres];
! 11761: /* for(k=1; k<=i1;k++){ */
! 11762: /* if(i1 != 1 && TKresult[nres]!= k) */
! 11763: /* continue; */
! 11764: fprintf(ficrespij,"\n#****** ");
! 11765: for(j=1;j<=cptcovs;j++){
! 11766: fprintf(ficrespij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
! 11767: /* fprintf(ficrespij,"@wV%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 11768: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
! 11769: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 11770: /* fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
! 11771: }
! 11772: fprintf(ficrespij,"******\n");
! 11773:
! 11774: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
! 11775: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
! 11776: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
! 11777:
! 11778: /* nhstepm=nhstepm*YEARM; aff par mois*/
! 11779:
! 11780: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
! 11781: oldm=oldms;savm=savms;
! 11782: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
! 11783: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
! 11784: for(i=1; i<=nlstate;i++)
! 11785: for(j=1; j<=nlstate+ndeath;j++)
! 11786: fprintf(ficrespij," %1d-%1d",i,j);
! 11787: fprintf(ficrespij,"\n");
! 11788: for (h=0; h<=nhstepm; h++){
! 11789: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
! 11790: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.183 brouard 11791: for(i=1; i<=nlstate;i++)
11792: for(j=1; j<=nlstate+ndeath;j++)
1.337 ! brouard 11793: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.183 brouard 11794: fprintf(ficrespij,"\n");
11795: }
1.337 ! brouard 11796: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
! 11797: fprintf(ficrespij,"\n");
1.180 brouard 11798: }
1.337 ! brouard 11799: }
! 11800: /*}*/
! 11801: return 0;
1.180 brouard 11802: }
1.218 brouard 11803:
11804: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 11805: /*------------- h Bij x at various ages ------------*/
1.336 brouard 11806: /* To be optimized with precov */
1.217 brouard 11807: int stepsize;
1.218 brouard 11808: /* int agelim; */
11809: int ageminl;
1.217 brouard 11810: int hstepm;
11811: int nhstepm;
1.238 brouard 11812: int h, i, i1, j, k, nres;
1.218 brouard 11813:
1.217 brouard 11814: double agedeb;
11815: double ***p3mat;
1.218 brouard 11816:
11817: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
11818: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
11819: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11820: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11821: }
11822: printf("Computing pij back: result on file '%s' \n", filerespijb);
11823: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
11824:
11825: stepsize=(int) (stepm+YEARM-1)/YEARM;
11826: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 11827:
1.218 brouard 11828: /* agelim=AGESUP; */
1.289 brouard 11829: ageminl=AGEINF; /* was 30 */
1.218 brouard 11830: hstepm=stepsize*YEARM; /* Every year of age */
11831: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11832:
11833: /* hstepm=1; aff par mois*/
11834: pstamp(ficrespijb);
1.255 brouard 11835: 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 11836: i1= pow(2,cptcoveff);
1.218 brouard 11837: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11838: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11839: /* k=k+1; */
1.238 brouard 11840: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 ! brouard 11841: k=TKresult[nres];
! 11842: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
! 11843: /* if(i1 != 1 && TKresult[nres]!= k) */
! 11844: /* continue; */
! 11845: fprintf(ficrespijb,"\n#****** ");
! 11846: for(j=1;j<=cptcovs;j++){
! 11847: fprintf(ficrespij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
! 11848: /* for(j=1;j<=cptcoveff;j++) */
! 11849: /* fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
! 11850: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
! 11851: /* fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
! 11852: }
! 11853: fprintf(ficrespijb,"******\n");
! 11854: if(invalidvarcomb[k]){ /* Is it necessary here? */
! 11855: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
! 11856: continue;
! 11857: }
! 11858:
! 11859: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
! 11860: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
! 11861: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
! 11862: 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 */
! 11863: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
! 11864:
! 11865: /* nhstepm=nhstepm*YEARM; aff par mois*/
! 11866:
! 11867: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
! 11868: /* and memory limitations if stepm is small */
! 11869:
! 11870: /* oldm=oldms;savm=savms; */
! 11871: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
! 11872: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
! 11873: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
! 11874: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
! 11875: for(i=1; i<=nlstate;i++)
! 11876: for(j=1; j<=nlstate+ndeath;j++)
! 11877: fprintf(ficrespijb," %1d-%1d",i,j);
! 11878: fprintf(ficrespijb,"\n");
! 11879: for (h=0; h<=nhstepm; h++){
! 11880: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
! 11881: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
! 11882: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
1.217 brouard 11883: for(i=1; i<=nlstate;i++)
11884: for(j=1; j<=nlstate+ndeath;j++)
1.337 ! brouard 11885: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.217 brouard 11886: fprintf(ficrespijb,"\n");
1.337 ! brouard 11887: }
! 11888: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
! 11889: fprintf(ficrespijb,"\n");
! 11890: } /* end age deb */
! 11891: /* } /\* end combination *\/ */
1.238 brouard 11892: } /* end nres */
1.218 brouard 11893: return 0;
11894: } /* hBijx */
1.217 brouard 11895:
1.180 brouard 11896:
1.136 brouard 11897: /***********************************************/
11898: /**************** Main Program *****************/
11899: /***********************************************/
11900:
11901: int main(int argc, char *argv[])
11902: {
11903: #ifdef GSL
11904: const gsl_multimin_fminimizer_type *T;
11905: size_t iteri = 0, it;
11906: int rval = GSL_CONTINUE;
11907: int status = GSL_SUCCESS;
11908: double ssval;
11909: #endif
11910: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11911: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11912: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11913: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11914: int jj, ll, li, lj, lk;
1.136 brouard 11915: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11916: int num_filled;
1.136 brouard 11917: int itimes;
11918: int NDIM=2;
11919: int vpopbased=0;
1.235 brouard 11920: int nres=0;
1.258 brouard 11921: int endishere=0;
1.277 brouard 11922: int noffset=0;
1.274 brouard 11923: int ncurrv=0; /* Temporary variable */
11924:
1.164 brouard 11925: char ca[32], cb[32];
1.136 brouard 11926: /* FILE *fichtm; *//* Html File */
11927: /* FILE *ficgp;*/ /*Gnuplot File */
11928: struct stat info;
1.191 brouard 11929: double agedeb=0.;
1.194 brouard 11930:
11931: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11932: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11933:
1.165 brouard 11934: double fret;
1.191 brouard 11935: double dum=0.; /* Dummy variable */
1.136 brouard 11936: double ***p3mat;
1.218 brouard 11937: /* double ***mobaverage; */
1.319 brouard 11938: double wald;
1.164 brouard 11939:
11940: char line[MAXLINE];
1.197 brouard 11941: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11942:
1.234 brouard 11943: char modeltemp[MAXLINE];
1.332 brouard 11944: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 11945:
1.136 brouard 11946: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11947: char *tok, *val; /* pathtot */
1.334 brouard 11948: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.195 brouard 11949: int c, h , cpt, c2;
1.191 brouard 11950: int jl=0;
11951: int i1, j1, jk, stepsize=0;
1.194 brouard 11952: int count=0;
11953:
1.164 brouard 11954: int *tab;
1.136 brouard 11955: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11956: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11957: /* double anprojf, mprojf, jprojf; */
11958: /* double jintmean,mintmean,aintmean; */
11959: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11960: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11961: double yrfproj= 10.0; /* Number of years of forward projections */
11962: double yrbproj= 10.0; /* Number of years of backward projections */
11963: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11964: int mobilav=0,popforecast=0;
1.191 brouard 11965: int hstepm=0, nhstepm=0;
1.136 brouard 11966: int agemortsup;
11967: float sumlpop=0.;
11968: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11969: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11970:
1.191 brouard 11971: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11972: double ftolpl=FTOL;
11973: double **prlim;
1.217 brouard 11974: double **bprlim;
1.317 brouard 11975: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
11976: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11977: double ***paramstart; /* Matrix of starting parameter values */
11978: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11979: double **matcov; /* Matrix of covariance */
1.203 brouard 11980: double **hess; /* Hessian matrix */
1.136 brouard 11981: double ***delti3; /* Scale */
11982: double *delti; /* Scale */
11983: double ***eij, ***vareij;
11984: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11985:
1.136 brouard 11986: double *epj, vepp;
1.164 brouard 11987:
1.273 brouard 11988: double dateprev1, dateprev2;
1.296 brouard 11989: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11990: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11991:
1.217 brouard 11992:
1.136 brouard 11993: double **ximort;
1.145 brouard 11994: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11995: int *dcwave;
11996:
1.164 brouard 11997: char z[1]="c";
1.136 brouard 11998:
11999: /*char *strt;*/
12000: char strtend[80];
1.126 brouard 12001:
1.164 brouard 12002:
1.126 brouard 12003: /* setlocale (LC_ALL, ""); */
12004: /* bindtextdomain (PACKAGE, LOCALEDIR); */
12005: /* textdomain (PACKAGE); */
12006: /* setlocale (LC_CTYPE, ""); */
12007: /* setlocale (LC_MESSAGES, ""); */
12008:
12009: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 12010: rstart_time = time(NULL);
12011: /* (void) gettimeofday(&start_time,&tzp);*/
12012: start_time = *localtime(&rstart_time);
1.126 brouard 12013: curr_time=start_time;
1.157 brouard 12014: /*tml = *localtime(&start_time.tm_sec);*/
12015: /* strcpy(strstart,asctime(&tml)); */
12016: strcpy(strstart,asctime(&start_time));
1.126 brouard 12017:
12018: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 12019: /* tp.tm_sec = tp.tm_sec +86400; */
12020: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 12021: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
12022: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
12023: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 12024: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 12025: /* strt=asctime(&tmg); */
12026: /* printf("Time(after) =%s",strstart); */
12027: /* (void) time (&time_value);
12028: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
12029: * tm = *localtime(&time_value);
12030: * strstart=asctime(&tm);
12031: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
12032: */
12033:
12034: nberr=0; /* Number of errors and warnings */
12035: nbwarn=0;
1.184 brouard 12036: #ifdef WIN32
12037: _getcwd(pathcd, size);
12038: #else
1.126 brouard 12039: getcwd(pathcd, size);
1.184 brouard 12040: #endif
1.191 brouard 12041: syscompilerinfo(0);
1.196 brouard 12042: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 12043: if(argc <=1){
12044: printf("\nEnter the parameter file name: ");
1.205 brouard 12045: if(!fgets(pathr,FILENAMELENGTH,stdin)){
12046: printf("ERROR Empty parameter file name\n");
12047: goto end;
12048: }
1.126 brouard 12049: i=strlen(pathr);
12050: if(pathr[i-1]=='\n')
12051: pathr[i-1]='\0';
1.156 brouard 12052: i=strlen(pathr);
1.205 brouard 12053: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 12054: pathr[i-1]='\0';
1.205 brouard 12055: }
12056: i=strlen(pathr);
12057: if( i==0 ){
12058: printf("ERROR Empty parameter file name\n");
12059: goto end;
12060: }
12061: for (tok = pathr; tok != NULL; ){
1.126 brouard 12062: printf("Pathr |%s|\n",pathr);
12063: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
12064: printf("val= |%s| pathr=%s\n",val,pathr);
12065: strcpy (pathtot, val);
12066: if(pathr[0] == '\0') break; /* Dirty */
12067: }
12068: }
1.281 brouard 12069: else if (argc<=2){
12070: strcpy(pathtot,argv[1]);
12071: }
1.126 brouard 12072: else{
12073: strcpy(pathtot,argv[1]);
1.281 brouard 12074: strcpy(z,argv[2]);
12075: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 12076: }
12077: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
12078: /*cygwin_split_path(pathtot,path,optionfile);
12079: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
12080: /* cutv(path,optionfile,pathtot,'\\');*/
12081:
12082: /* Split argv[0], imach program to get pathimach */
12083: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
12084: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12085: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12086: /* strcpy(pathimach,argv[0]); */
12087: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
12088: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
12089: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 12090: #ifdef WIN32
12091: _chdir(path); /* Can be a relative path */
12092: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
12093: #else
1.126 brouard 12094: chdir(path); /* Can be a relative path */
1.184 brouard 12095: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
12096: #endif
12097: printf("Current directory %s!\n",pathcd);
1.126 brouard 12098: strcpy(command,"mkdir ");
12099: strcat(command,optionfilefiname);
12100: if((outcmd=system(command)) != 0){
1.169 brouard 12101: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 12102: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
12103: /* fclose(ficlog); */
12104: /* exit(1); */
12105: }
12106: /* if((imk=mkdir(optionfilefiname))<0){ */
12107: /* perror("mkdir"); */
12108: /* } */
12109:
12110: /*-------- arguments in the command line --------*/
12111:
1.186 brouard 12112: /* Main Log file */
1.126 brouard 12113: strcat(filelog, optionfilefiname);
12114: strcat(filelog,".log"); /* */
12115: if((ficlog=fopen(filelog,"w"))==NULL) {
12116: printf("Problem with logfile %s\n",filelog);
12117: goto end;
12118: }
12119: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 12120: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 12121: fprintf(ficlog,"\nEnter the parameter file name: \n");
12122: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
12123: path=%s \n\
12124: optionfile=%s\n\
12125: optionfilext=%s\n\
1.156 brouard 12126: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 12127:
1.197 brouard 12128: syscompilerinfo(1);
1.167 brouard 12129:
1.126 brouard 12130: printf("Local time (at start):%s",strstart);
12131: fprintf(ficlog,"Local time (at start): %s",strstart);
12132: fflush(ficlog);
12133: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 12134: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 12135:
12136: /* */
12137: strcpy(fileres,"r");
12138: strcat(fileres, optionfilefiname);
1.201 brouard 12139: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 12140: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 12141: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 12142:
1.186 brouard 12143: /* Main ---------arguments file --------*/
1.126 brouard 12144:
12145: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 12146: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
12147: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 12148: fflush(ficlog);
1.149 brouard 12149: /* goto end; */
12150: exit(70);
1.126 brouard 12151: }
12152:
12153: strcpy(filereso,"o");
1.201 brouard 12154: strcat(filereso,fileresu);
1.126 brouard 12155: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
12156: printf("Problem with Output resultfile: %s\n", filereso);
12157: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
12158: fflush(ficlog);
12159: goto end;
12160: }
1.278 brouard 12161: /*-------- Rewriting parameter file ----------*/
12162: strcpy(rfileres,"r"); /* "Rparameterfile */
12163: strcat(rfileres,optionfilefiname); /* Parameter file first name */
12164: strcat(rfileres,"."); /* */
12165: strcat(rfileres,optionfilext); /* Other files have txt extension */
12166: if((ficres =fopen(rfileres,"w"))==NULL) {
12167: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
12168: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
12169: fflush(ficlog);
12170: goto end;
12171: }
12172: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 12173:
1.278 brouard 12174:
1.126 brouard 12175: /* Reads comments: lines beginning with '#' */
12176: numlinepar=0;
1.277 brouard 12177: /* Is it a BOM UTF-8 Windows file? */
12178: /* First parameter line */
1.197 brouard 12179: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 12180: noffset=0;
12181: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12182: {
12183: noffset=noffset+3;
12184: printf("# File is an UTF8 Bom.\n"); // 0xBF
12185: }
1.302 brouard 12186: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12187: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 12188: {
12189: noffset=noffset+2;
12190: printf("# File is an UTF16BE BOM file\n");
12191: }
12192: else if( line[0] == 0 && line[1] == 0)
12193: {
12194: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12195: noffset=noffset+4;
12196: printf("# File is an UTF16BE BOM file\n");
12197: }
12198: } else{
12199: ;/*printf(" Not a BOM file\n");*/
12200: }
12201:
1.197 brouard 12202: /* If line starts with a # it is a comment */
1.277 brouard 12203: if (line[noffset] == '#') {
1.197 brouard 12204: numlinepar++;
12205: fputs(line,stdout);
12206: fputs(line,ficparo);
1.278 brouard 12207: fputs(line,ficres);
1.197 brouard 12208: fputs(line,ficlog);
12209: continue;
12210: }else
12211: break;
12212: }
12213: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
12214: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
12215: if (num_filled != 5) {
12216: printf("Should be 5 parameters\n");
1.283 brouard 12217: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 12218: }
1.126 brouard 12219: numlinepar++;
1.197 brouard 12220: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 12221: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12222: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12223: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 12224: }
12225: /* Second parameter line */
12226: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 12227: /* while(fscanf(ficpar,"%[^\n]", line)) { */
12228: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 12229: if (line[0] == '#') {
12230: numlinepar++;
1.283 brouard 12231: printf("%s",line);
12232: fprintf(ficres,"%s",line);
12233: fprintf(ficparo,"%s",line);
12234: fprintf(ficlog,"%s",line);
1.197 brouard 12235: continue;
12236: }else
12237: break;
12238: }
1.223 brouard 12239: 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", \
12240: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
12241: if (num_filled != 11) {
12242: 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 12243: printf("but line=%s\n",line);
1.283 brouard 12244: 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");
12245: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 12246: }
1.286 brouard 12247: if( lastpass > maxwav){
12248: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12249: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12250: fflush(ficlog);
12251: goto end;
12252: }
12253: 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 12254: 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 12255: 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 12256: 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 12257: }
1.203 brouard 12258: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 12259: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 12260: /* Third parameter line */
12261: while(fgets(line, MAXLINE, ficpar)) {
12262: /* If line starts with a # it is a comment */
12263: if (line[0] == '#') {
12264: numlinepar++;
1.283 brouard 12265: printf("%s",line);
12266: fprintf(ficres,"%s",line);
12267: fprintf(ficparo,"%s",line);
12268: fprintf(ficlog,"%s",line);
1.197 brouard 12269: continue;
12270: }else
12271: break;
12272: }
1.201 brouard 12273: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 12274: if (num_filled != 1){
1.302 brouard 12275: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
12276: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 12277: model[0]='\0';
12278: goto end;
12279: }
12280: else{
12281: if (model[0]=='+'){
12282: for(i=1; i<=strlen(model);i++)
12283: modeltemp[i-1]=model[i];
1.201 brouard 12284: strcpy(model,modeltemp);
1.197 brouard 12285: }
12286: }
1.199 brouard 12287: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 12288: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 12289: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
12290: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
12291: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 12292: }
12293: /* 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); */
12294: /* numlinepar=numlinepar+3; /\* In general *\/ */
12295: /* 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 12296: /* 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); */
12297: /* 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 12298: fflush(ficlog);
1.190 brouard 12299: /* if(model[0]=='#'|| model[0]== '\0'){ */
12300: if(model[0]=='#'){
1.279 brouard 12301: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
12302: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
12303: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 12304: if(mle != -1){
1.279 brouard 12305: 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 12306: exit(1);
12307: }
12308: }
1.126 brouard 12309: while((c=getc(ficpar))=='#' && c!= EOF){
12310: ungetc(c,ficpar);
12311: fgets(line, MAXLINE, ficpar);
12312: numlinepar++;
1.195 brouard 12313: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
12314: z[0]=line[1];
12315: }
12316: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 12317: fputs(line, stdout);
12318: //puts(line);
1.126 brouard 12319: fputs(line,ficparo);
12320: fputs(line,ficlog);
12321: }
12322: ungetc(c,ficpar);
12323:
12324:
1.290 brouard 12325: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
12326: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
12327: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
12328: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 12329: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
12330: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
12331: v1+v2*age+v2*v3 makes cptcovn = 3
12332: */
12333: if (strlen(model)>1)
1.187 brouard 12334: 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 12335: else
1.187 brouard 12336: ncovmodel=2; /* Constant and age */
1.133 brouard 12337: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
12338: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 12339: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
12340: 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);
12341: 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);
12342: fflush(stdout);
12343: fclose (ficlog);
12344: goto end;
12345: }
1.126 brouard 12346: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12347: delti=delti3[1][1];
12348: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
12349: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 12350: /* We could also provide initial parameters values giving by simple logistic regression
12351: * only one way, that is without matrix product. We will have nlstate maximizations */
12352: /* for(i=1;i<nlstate;i++){ */
12353: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12354: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12355: /* } */
1.126 brouard 12356: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 12357: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
12358: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12359: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12360: fclose (ficparo);
12361: fclose (ficlog);
12362: goto end;
12363: exit(0);
1.220 brouard 12364: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 12365: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 12366: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
12367: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12368: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12369: matcov=matrix(1,npar,1,npar);
1.203 brouard 12370: hess=matrix(1,npar,1,npar);
1.220 brouard 12371: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 12372: /* Read guessed parameters */
1.126 brouard 12373: /* Reads comments: lines beginning with '#' */
12374: while((c=getc(ficpar))=='#' && c!= EOF){
12375: ungetc(c,ficpar);
12376: fgets(line, MAXLINE, ficpar);
12377: numlinepar++;
1.141 brouard 12378: fputs(line,stdout);
1.126 brouard 12379: fputs(line,ficparo);
12380: fputs(line,ficlog);
12381: }
12382: ungetc(c,ficpar);
12383:
12384: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 12385: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 12386: for(i=1; i <=nlstate; i++){
1.234 brouard 12387: j=0;
1.126 brouard 12388: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 12389: if(jj==i) continue;
12390: j++;
1.292 brouard 12391: while((c=getc(ficpar))=='#' && c!= EOF){
12392: ungetc(c,ficpar);
12393: fgets(line, MAXLINE, ficpar);
12394: numlinepar++;
12395: fputs(line,stdout);
12396: fputs(line,ficparo);
12397: fputs(line,ficlog);
12398: }
12399: ungetc(c,ficpar);
1.234 brouard 12400: fscanf(ficpar,"%1d%1d",&i1,&j1);
12401: if ((i1 != i) || (j1 != jj)){
12402: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 12403: It might be a problem of design; if ncovcol and the model are correct\n \
12404: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 12405: exit(1);
12406: }
12407: fprintf(ficparo,"%1d%1d",i1,j1);
12408: if(mle==1)
12409: printf("%1d%1d",i,jj);
12410: fprintf(ficlog,"%1d%1d",i,jj);
12411: for(k=1; k<=ncovmodel;k++){
12412: fscanf(ficpar," %lf",¶m[i][j][k]);
12413: if(mle==1){
12414: printf(" %lf",param[i][j][k]);
12415: fprintf(ficlog," %lf",param[i][j][k]);
12416: }
12417: else
12418: fprintf(ficlog," %lf",param[i][j][k]);
12419: fprintf(ficparo," %lf",param[i][j][k]);
12420: }
12421: fscanf(ficpar,"\n");
12422: numlinepar++;
12423: if(mle==1)
12424: printf("\n");
12425: fprintf(ficlog,"\n");
12426: fprintf(ficparo,"\n");
1.126 brouard 12427: }
12428: }
12429: fflush(ficlog);
1.234 brouard 12430:
1.251 brouard 12431: /* Reads parameters values */
1.126 brouard 12432: p=param[1][1];
1.251 brouard 12433: pstart=paramstart[1][1];
1.126 brouard 12434:
12435: /* Reads comments: lines beginning with '#' */
12436: while((c=getc(ficpar))=='#' && c!= EOF){
12437: ungetc(c,ficpar);
12438: fgets(line, MAXLINE, ficpar);
12439: numlinepar++;
1.141 brouard 12440: fputs(line,stdout);
1.126 brouard 12441: fputs(line,ficparo);
12442: fputs(line,ficlog);
12443: }
12444: ungetc(c,ficpar);
12445:
12446: for(i=1; i <=nlstate; i++){
12447: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 12448: fscanf(ficpar,"%1d%1d",&i1,&j1);
12449: if ( (i1-i) * (j1-j) != 0){
12450: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
12451: exit(1);
12452: }
12453: printf("%1d%1d",i,j);
12454: fprintf(ficparo,"%1d%1d",i1,j1);
12455: fprintf(ficlog,"%1d%1d",i1,j1);
12456: for(k=1; k<=ncovmodel;k++){
12457: fscanf(ficpar,"%le",&delti3[i][j][k]);
12458: printf(" %le",delti3[i][j][k]);
12459: fprintf(ficparo," %le",delti3[i][j][k]);
12460: fprintf(ficlog," %le",delti3[i][j][k]);
12461: }
12462: fscanf(ficpar,"\n");
12463: numlinepar++;
12464: printf("\n");
12465: fprintf(ficparo,"\n");
12466: fprintf(ficlog,"\n");
1.126 brouard 12467: }
12468: }
12469: fflush(ficlog);
1.234 brouard 12470:
1.145 brouard 12471: /* Reads covariance matrix */
1.126 brouard 12472: delti=delti3[1][1];
1.220 brouard 12473:
12474:
1.126 brouard 12475: /* 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 12476:
1.126 brouard 12477: /* Reads comments: lines beginning with '#' */
12478: while((c=getc(ficpar))=='#' && c!= EOF){
12479: ungetc(c,ficpar);
12480: fgets(line, MAXLINE, ficpar);
12481: numlinepar++;
1.141 brouard 12482: fputs(line,stdout);
1.126 brouard 12483: fputs(line,ficparo);
12484: fputs(line,ficlog);
12485: }
12486: ungetc(c,ficpar);
1.220 brouard 12487:
1.126 brouard 12488: matcov=matrix(1,npar,1,npar);
1.203 brouard 12489: hess=matrix(1,npar,1,npar);
1.131 brouard 12490: for(i=1; i <=npar; i++)
12491: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12492:
1.194 brouard 12493: /* Scans npar lines */
1.126 brouard 12494: for(i=1; i <=npar; i++){
1.226 brouard 12495: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12496: if(count != 3){
1.226 brouard 12497: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12498: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12499: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12500: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12501: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12502: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12503: exit(1);
1.220 brouard 12504: }else{
1.226 brouard 12505: if(mle==1)
12506: printf("%1d%1d%d",i1,j1,jk);
12507: }
12508: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12509: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12510: for(j=1; j <=i; j++){
1.226 brouard 12511: fscanf(ficpar," %le",&matcov[i][j]);
12512: if(mle==1){
12513: printf(" %.5le",matcov[i][j]);
12514: }
12515: fprintf(ficlog," %.5le",matcov[i][j]);
12516: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12517: }
12518: fscanf(ficpar,"\n");
12519: numlinepar++;
12520: if(mle==1)
1.220 brouard 12521: printf("\n");
1.126 brouard 12522: fprintf(ficlog,"\n");
12523: fprintf(ficparo,"\n");
12524: }
1.194 brouard 12525: /* End of read covariance matrix npar lines */
1.126 brouard 12526: for(i=1; i <=npar; i++)
12527: for(j=i+1;j<=npar;j++)
1.226 brouard 12528: matcov[i][j]=matcov[j][i];
1.126 brouard 12529:
12530: if(mle==1)
12531: printf("\n");
12532: fprintf(ficlog,"\n");
12533:
12534: fflush(ficlog);
12535:
12536: } /* End of mle != -3 */
1.218 brouard 12537:
1.186 brouard 12538: /* Main data
12539: */
1.290 brouard 12540: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12541: /* num=lvector(1,n); */
12542: /* moisnais=vector(1,n); */
12543: /* annais=vector(1,n); */
12544: /* moisdc=vector(1,n); */
12545: /* andc=vector(1,n); */
12546: /* weight=vector(1,n); */
12547: /* agedc=vector(1,n); */
12548: /* cod=ivector(1,n); */
12549: /* for(i=1;i<=n;i++){ */
12550: num=lvector(firstobs,lastobs);
12551: moisnais=vector(firstobs,lastobs);
12552: annais=vector(firstobs,lastobs);
12553: moisdc=vector(firstobs,lastobs);
12554: andc=vector(firstobs,lastobs);
12555: weight=vector(firstobs,lastobs);
12556: agedc=vector(firstobs,lastobs);
12557: cod=ivector(firstobs,lastobs);
12558: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 12559: num[i]=0;
12560: moisnais[i]=0;
12561: annais[i]=0;
12562: moisdc[i]=0;
12563: andc[i]=0;
12564: agedc[i]=0;
12565: cod[i]=0;
12566: weight[i]=1.0; /* Equal weights, 1 by default */
12567: }
1.290 brouard 12568: mint=matrix(1,maxwav,firstobs,lastobs);
12569: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 12570: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.336 brouard 12571: /* printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel)); */
1.126 brouard 12572: tab=ivector(1,NCOVMAX);
1.144 brouard 12573: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 12574: 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 12575:
1.136 brouard 12576: /* Reads data from file datafile */
12577: if (readdata(datafile, firstobs, lastobs, &imx)==1)
12578: goto end;
12579:
12580: /* Calculation of the number of parameters from char model */
1.234 brouard 12581: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 12582: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
12583: k=3 V4 Tvar[k=3]= 4 (from V4)
12584: k=2 V1 Tvar[k=2]= 1 (from V1)
12585: k=1 Tvar[1]=2 (from V2)
1.234 brouard 12586: */
12587:
12588: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
12589: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 12590: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 brouard 12591: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 12592: TvarsD=ivector(1,NCOVMAX); /* */
12593: TvarsQind=ivector(1,NCOVMAX); /* */
12594: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 12595: TvarF=ivector(1,NCOVMAX); /* */
12596: TvarFind=ivector(1,NCOVMAX); /* */
12597: TvarV=ivector(1,NCOVMAX); /* */
12598: TvarVind=ivector(1,NCOVMAX); /* */
12599: TvarA=ivector(1,NCOVMAX); /* */
12600: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 12601: TvarFD=ivector(1,NCOVMAX); /* */
12602: TvarFDind=ivector(1,NCOVMAX); /* */
12603: TvarFQ=ivector(1,NCOVMAX); /* */
12604: TvarFQind=ivector(1,NCOVMAX); /* */
12605: TvarVD=ivector(1,NCOVMAX); /* */
12606: TvarVDind=ivector(1,NCOVMAX); /* */
12607: TvarVQ=ivector(1,NCOVMAX); /* */
12608: TvarVQind=ivector(1,NCOVMAX); /* */
12609:
1.230 brouard 12610: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 12611: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 12612: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
12613: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
12614: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 12615: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
12616: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
12617: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
12618: */
12619: /* For model-covariate k tells which data-covariate to use but
12620: because this model-covariate is a construction we invent a new column
12621: ncovcol + k1
12622: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
12623: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 12624: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
12625: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 12626: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
12627: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 12628: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 12629: */
1.145 brouard 12630: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
12631: 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 12632: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
12633: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 brouard 12634: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 12635: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 12636: 4 covariates (3 plus signs)
12637: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 12638: */
12639: for(i=1;i<NCOVMAX;i++)
12640: Tage[i]=0;
1.230 brouard 12641: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 12642: * individual dummy, fixed or varying:
12643: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
12644: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 12645: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
12646: * V1 df, V2 qf, V3 & V4 dv, V5 qv
12647: * Tmodelind[1]@9={9,0,3,2,}*/
12648: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
12649: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 12650: * individual quantitative, fixed or varying:
12651: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
12652: * 3, 1, 0, 0, 0, 0, 0, 0},
12653: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 12654: /* Main decodemodel */
12655:
1.187 brouard 12656:
1.223 brouard 12657: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 12658: goto end;
12659:
1.137 brouard 12660: if((double)(lastobs-imx)/(double)imx > 1.10){
12661: nbwarn++;
12662: 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);
12663: 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);
12664: }
1.136 brouard 12665: /* if(mle==1){*/
1.137 brouard 12666: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
12667: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 12668: }
12669:
12670: /*-calculation of age at interview from date of interview and age at death -*/
12671: agev=matrix(1,maxwav,1,imx);
12672:
12673: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
12674: goto end;
12675:
1.126 brouard 12676:
1.136 brouard 12677: agegomp=(int)agemin;
1.290 brouard 12678: free_vector(moisnais,firstobs,lastobs);
12679: free_vector(annais,firstobs,lastobs);
1.126 brouard 12680: /* free_matrix(mint,1,maxwav,1,n);
12681: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 12682: /* free_vector(moisdc,1,n); */
12683: /* free_vector(andc,1,n); */
1.145 brouard 12684: /* */
12685:
1.126 brouard 12686: wav=ivector(1,imx);
1.214 brouard 12687: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
12688: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
12689: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
12690: 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.*/
12691: bh=imatrix(1,lastpass-firstpass+2,1,imx);
12692: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 12693:
12694: /* Concatenates waves */
1.214 brouard 12695: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
12696: Death is a valid wave (if date is known).
12697: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
12698: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
12699: and mw[mi+1][i]. dh depends on stepm.
12700: */
12701:
1.126 brouard 12702: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 12703: /* Concatenates waves */
1.145 brouard 12704:
1.290 brouard 12705: free_vector(moisdc,firstobs,lastobs);
12706: free_vector(andc,firstobs,lastobs);
1.215 brouard 12707:
1.126 brouard 12708: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12709: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12710: ncodemax[1]=1;
1.145 brouard 12711: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12712: cptcoveff=0;
1.220 brouard 12713: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 brouard 12714: 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 12715: }
12716:
12717: ncovcombmax=pow(2,cptcoveff);
12718: invalidvarcomb=ivector(1, ncovcombmax);
12719: for(i=1;i<ncovcombmax;i++)
12720: invalidvarcomb[i]=0;
12721:
1.211 brouard 12722: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12723: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12724: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12725:
1.200 brouard 12726: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12727: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12728: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12729: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12730: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12731: * (currently 0 or 1) in the data.
12732: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12733: * corresponding modality (h,j).
12734: */
12735:
1.145 brouard 12736: h=0;
12737: /*if (cptcovn > 0) */
1.126 brouard 12738: m=pow(2,cptcoveff);
12739:
1.144 brouard 12740: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12741: * For k=4 covariates, h goes from 1 to m=2**k
12742: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12743: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 12744: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
12745: *______________________________ *______________________
12746: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
12747: * 2 2 1 1 1 * 1 0 0 0 1
12748: * 3 i=2 1 2 1 1 * 2 0 0 1 0
12749: * 4 2 2 1 1 * 3 0 0 1 1
12750: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
12751: * 6 2 1 2 1 * 5 0 1 0 1
12752: * 7 i=4 1 2 2 1 * 6 0 1 1 0
12753: * 8 2 2 2 1 * 7 0 1 1 1
12754: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
12755: * 10 2 1 1 2 * 9 1 0 0 1
12756: * 11 i=6 1 2 1 2 * 10 1 0 1 0
12757: * 12 2 2 1 2 * 11 1 0 1 1
12758: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
12759: * 14 2 1 2 2 * 13 1 1 0 1
12760: * 15 i=8 1 2 2 2 * 14 1 1 1 0
12761: * 16 2 2 2 2 * 15 1 1 1 1
12762: */
1.212 brouard 12763: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 12764: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
12765: * and the value of each covariate?
12766: * V1=1, V2=1, V3=2, V4=1 ?
12767: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
12768: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
12769: * In order to get the real value in the data, we use nbcode
12770: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
12771: * We are keeping this crazy system in order to be able (in the future?)
12772: * to have more than 2 values (0 or 1) for a covariate.
12773: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
12774: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
12775: * bbbbbbbb
12776: * 76543210
12777: * h-1 00000101 (6-1=5)
1.219 brouard 12778: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 12779: * &
12780: * 1 00000001 (1)
1.219 brouard 12781: * 00000000 = 1 & ((h-1) >> (k-1))
12782: * +1= 00000001 =1
1.211 brouard 12783: *
12784: * h=14, k=3 => h'=h-1=13, k'=k-1=2
12785: * h' 1101 =2^3+2^2+0x2^1+2^0
12786: * >>k' 11
12787: * & 00000001
12788: * = 00000001
12789: * +1 = 00000010=2 = codtabm(14,3)
12790: * Reverse h=6 and m=16?
12791: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
12792: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
12793: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
12794: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
12795: * V3=decodtabm(14,3,2**4)=2
12796: * h'=13 1101 =2^3+2^2+0x2^1+2^0
12797: *(h-1) >> (j-1) 0011 =13 >> 2
12798: * &1 000000001
12799: * = 000000001
12800: * +1= 000000010 =2
12801: * 2211
12802: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
12803: * V3=2
1.220 brouard 12804: * codtabm and decodtabm are identical
1.211 brouard 12805: */
12806:
1.145 brouard 12807:
12808: free_ivector(Ndum,-1,NCOVMAX);
12809:
12810:
1.126 brouard 12811:
1.186 brouard 12812: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 12813: strcpy(optionfilegnuplot,optionfilefiname);
12814: if(mle==-3)
1.201 brouard 12815: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 12816: strcat(optionfilegnuplot,".gp");
12817:
12818: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
12819: printf("Problem with file %s",optionfilegnuplot);
12820: }
12821: else{
1.204 brouard 12822: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 12823: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 12824: //fprintf(ficgp,"set missing 'NaNq'\n");
12825: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 12826: }
12827: /* fclose(ficgp);*/
1.186 brouard 12828:
12829:
12830: /* Initialisation of --------- index.htm --------*/
1.126 brouard 12831:
12832: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
12833: if(mle==-3)
1.201 brouard 12834: strcat(optionfilehtm,"-MORT_");
1.126 brouard 12835: strcat(optionfilehtm,".htm");
12836: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 12837: printf("Problem with %s \n",optionfilehtm);
12838: exit(0);
1.126 brouard 12839: }
12840:
12841: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
12842: strcat(optionfilehtmcov,"-cov.htm");
12843: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
12844: printf("Problem with %s \n",optionfilehtmcov), exit(0);
12845: }
12846: else{
12847: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
12848: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12849: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 12850: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
12851: }
12852:
1.335 brouard 12853: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
12854: <title>IMaCh %s</title></head>\n\
12855: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
12856: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
12857: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
12858: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
12859: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
12860:
12861: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12862: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 12863: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.337 ! brouard 12864: This file: <a href=\"%s\">%s</a></br>Title=%s <br>Datafile=<a href=\"%s\">%s</a> Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 12865: \n\
12866: <hr size=\"2\" color=\"#EC5E5E\">\
12867: <ul><li><h4>Parameter files</h4>\n\
12868: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
12869: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
12870: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
12871: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
12872: - Date and time at start: %s</ul>\n",\
1.335 brouard 12873: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 12874: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
12875: fileres,fileres,\
12876: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
12877: fflush(fichtm);
12878:
12879: strcpy(pathr,path);
12880: strcat(pathr,optionfilefiname);
1.184 brouard 12881: #ifdef WIN32
12882: _chdir(optionfilefiname); /* Move to directory named optionfile */
12883: #else
1.126 brouard 12884: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12885: #endif
12886:
1.126 brouard 12887:
1.220 brouard 12888: /* Calculates basic frequencies. Computes observed prevalence at single age
12889: and for any valid combination of covariates
1.126 brouard 12890: and prints on file fileres'p'. */
1.251 brouard 12891: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12892: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12893:
12894: fprintf(fichtm,"\n");
1.286 brouard 12895: 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 12896: ftol, stepm);
12897: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12898: ncurrv=1;
12899: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12900: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12901: ncurrv=i;
12902: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12903: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12904: ncurrv=i;
12905: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12906: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12907: ncurrv=i;
12908: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12909: 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", \
12910: nlstate, ndeath, maxwav, mle, weightopt);
12911:
12912: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12913: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12914:
12915:
1.317 brouard 12916: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12917: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12918: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12919: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12920: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12921: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12922: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12923: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12924: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12925:
1.126 brouard 12926: /* For Powell, parameters are in a vector p[] starting at p[1]
12927: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12928: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12929:
12930: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12931: /* For mortality only */
1.126 brouard 12932: if (mle==-3){
1.136 brouard 12933: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12934: for(i=1;i<=NDIM;i++)
12935: for(j=1;j<=NDIM;j++)
12936: ximort[i][j]=0.;
1.186 brouard 12937: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12938: cens=ivector(firstobs,lastobs);
12939: ageexmed=vector(firstobs,lastobs);
12940: agecens=vector(firstobs,lastobs);
12941: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12942:
1.126 brouard 12943: for (i=1; i<=imx; i++){
12944: dcwave[i]=-1;
12945: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12946: if (s[m][i]>nlstate) {
12947: dcwave[i]=m;
12948: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12949: break;
12950: }
1.126 brouard 12951: }
1.226 brouard 12952:
1.126 brouard 12953: for (i=1; i<=imx; i++) {
12954: if (wav[i]>0){
1.226 brouard 12955: ageexmed[i]=agev[mw[1][i]][i];
12956: j=wav[i];
12957: agecens[i]=1.;
12958:
12959: if (ageexmed[i]> 1 && wav[i] > 0){
12960: agecens[i]=agev[mw[j][i]][i];
12961: cens[i]= 1;
12962: }else if (ageexmed[i]< 1)
12963: cens[i]= -1;
12964: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12965: cens[i]=0 ;
1.126 brouard 12966: }
12967: else cens[i]=-1;
12968: }
12969:
12970: for (i=1;i<=NDIM;i++) {
12971: for (j=1;j<=NDIM;j++)
1.226 brouard 12972: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12973: }
12974:
1.302 brouard 12975: p[1]=0.0268; p[NDIM]=0.083;
12976: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12977:
12978:
1.136 brouard 12979: #ifdef GSL
12980: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12981: #else
1.126 brouard 12982: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12983: #endif
1.201 brouard 12984: strcpy(filerespow,"POW-MORT_");
12985: strcat(filerespow,fileresu);
1.126 brouard 12986: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12987: printf("Problem with resultfile: %s\n", filerespow);
12988: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12989: }
1.136 brouard 12990: #ifdef GSL
12991: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12992: #else
1.126 brouard 12993: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12994: #endif
1.126 brouard 12995: /* for (i=1;i<=nlstate;i++)
12996: for(j=1;j<=nlstate+ndeath;j++)
12997: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12998: */
12999: fprintf(ficrespow,"\n");
1.136 brouard 13000: #ifdef GSL
13001: /* gsl starts here */
13002: T = gsl_multimin_fminimizer_nmsimplex;
13003: gsl_multimin_fminimizer *sfm = NULL;
13004: gsl_vector *ss, *x;
13005: gsl_multimin_function minex_func;
13006:
13007: /* Initial vertex size vector */
13008: ss = gsl_vector_alloc (NDIM);
13009:
13010: if (ss == NULL){
13011: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
13012: }
13013: /* Set all step sizes to 1 */
13014: gsl_vector_set_all (ss, 0.001);
13015:
13016: /* Starting point */
1.126 brouard 13017:
1.136 brouard 13018: x = gsl_vector_alloc (NDIM);
13019:
13020: if (x == NULL){
13021: gsl_vector_free(ss);
13022: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
13023: }
13024:
13025: /* Initialize method and iterate */
13026: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 13027: /* gsl_vector_set(x, 0, 0.0268); */
13028: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 13029: gsl_vector_set(x, 0, p[1]);
13030: gsl_vector_set(x, 1, p[2]);
13031:
13032: minex_func.f = &gompertz_f;
13033: minex_func.n = NDIM;
13034: minex_func.params = (void *)&p; /* ??? */
13035:
13036: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
13037: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
13038:
13039: printf("Iterations beginning .....\n\n");
13040: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
13041:
13042: iteri=0;
13043: while (rval == GSL_CONTINUE){
13044: iteri++;
13045: status = gsl_multimin_fminimizer_iterate(sfm);
13046:
13047: if (status) printf("error: %s\n", gsl_strerror (status));
13048: fflush(0);
13049:
13050: if (status)
13051: break;
13052:
13053: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
13054: ssval = gsl_multimin_fminimizer_size (sfm);
13055:
13056: if (rval == GSL_SUCCESS)
13057: printf ("converged to a local maximum at\n");
13058:
13059: printf("%5d ", iteri);
13060: for (it = 0; it < NDIM; it++){
13061: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
13062: }
13063: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
13064: }
13065:
13066: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
13067:
13068: gsl_vector_free(x); /* initial values */
13069: gsl_vector_free(ss); /* inital step size */
13070: for (it=0; it<NDIM; it++){
13071: p[it+1]=gsl_vector_get(sfm->x,it);
13072: fprintf(ficrespow," %.12lf", p[it]);
13073: }
13074: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
13075: #endif
13076: #ifdef POWELL
13077: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
13078: #endif
1.126 brouard 13079: fclose(ficrespow);
13080:
1.203 brouard 13081: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 13082:
13083: for(i=1; i <=NDIM; i++)
13084: for(j=i+1;j<=NDIM;j++)
1.220 brouard 13085: matcov[i][j]=matcov[j][i];
1.126 brouard 13086:
13087: printf("\nCovariance matrix\n ");
1.203 brouard 13088: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 13089: for(i=1; i <=NDIM; i++) {
13090: for(j=1;j<=NDIM;j++){
1.220 brouard 13091: printf("%f ",matcov[i][j]);
13092: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 13093: }
1.203 brouard 13094: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 13095: }
13096:
13097: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 13098: for (i=1;i<=NDIM;i++) {
1.126 brouard 13099: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 13100: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
13101: }
1.302 brouard 13102: lsurv=vector(agegomp,AGESUP);
13103: lpop=vector(agegomp,AGESUP);
13104: tpop=vector(agegomp,AGESUP);
1.126 brouard 13105: lsurv[agegomp]=100000;
13106:
13107: for (k=agegomp;k<=AGESUP;k++) {
13108: agemortsup=k;
13109: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
13110: }
13111:
13112: for (k=agegomp;k<agemortsup;k++)
13113: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
13114:
13115: for (k=agegomp;k<agemortsup;k++){
13116: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
13117: sumlpop=sumlpop+lpop[k];
13118: }
13119:
13120: tpop[agegomp]=sumlpop;
13121: for (k=agegomp;k<(agemortsup-3);k++){
13122: /* tpop[k+1]=2;*/
13123: tpop[k+1]=tpop[k]-lpop[k];
13124: }
13125:
13126:
13127: printf("\nAge lx qx dx Lx Tx e(x)\n");
13128: for (k=agegomp;k<(agemortsup-2);k++)
13129: 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]);
13130:
13131:
13132: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 13133: ageminpar=50;
13134: agemaxpar=100;
1.194 brouard 13135: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
13136: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13137: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13138: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
13139: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13140: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13141: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13142: }else{
13143: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
13144: 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 13145: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 13146: }
1.201 brouard 13147: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 13148: stepm, weightopt,\
13149: model,imx,p,matcov,agemortsup);
13150:
1.302 brouard 13151: free_vector(lsurv,agegomp,AGESUP);
13152: free_vector(lpop,agegomp,AGESUP);
13153: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 13154: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 13155: free_ivector(dcwave,firstobs,lastobs);
13156: free_vector(agecens,firstobs,lastobs);
13157: free_vector(ageexmed,firstobs,lastobs);
13158: free_ivector(cens,firstobs,lastobs);
1.220 brouard 13159: #ifdef GSL
1.136 brouard 13160: #endif
1.186 brouard 13161: } /* Endof if mle==-3 mortality only */
1.205 brouard 13162: /* Standard */
13163: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
13164: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13165: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 13166: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 13167: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
13168: for (k=1; k<=npar;k++)
13169: printf(" %d %8.5f",k,p[k]);
13170: printf("\n");
1.205 brouard 13171: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
13172: /* mlikeli uses func not funcone */
1.247 brouard 13173: /* for(i=1;i<nlstate;i++){ */
13174: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
13175: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
13176: /* } */
1.205 brouard 13177: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
13178: }
13179: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
13180: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13181: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
13182: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13183: }
13184: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 13185: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13186: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 brouard 13187: /* exit(0); */
1.126 brouard 13188: for (k=1; k<=npar;k++)
13189: printf(" %d %8.5f",k,p[k]);
13190: printf("\n");
13191:
13192: /*--------- results files --------------*/
1.283 brouard 13193: /* 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 13194:
13195:
13196: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13197: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 13198: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13199:
13200: printf("#model= 1 + age ");
13201: fprintf(ficres,"#model= 1 + age ");
13202: fprintf(ficlog,"#model= 1 + age ");
13203: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
13204: </ul>", model);
13205:
13206: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
13207: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
13208: if(nagesqr==1){
13209: printf(" + age*age ");
13210: fprintf(ficres," + age*age ");
13211: fprintf(ficlog," + age*age ");
13212: fprintf(fichtm, "<th>+ age*age</th>");
13213: }
13214: for(j=1;j <=ncovmodel-2;j++){
13215: if(Typevar[j]==0) {
13216: printf(" + V%d ",Tvar[j]);
13217: fprintf(ficres," + V%d ",Tvar[j]);
13218: fprintf(ficlog," + V%d ",Tvar[j]);
13219: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13220: }else if(Typevar[j]==1) {
13221: printf(" + V%d*age ",Tvar[j]);
13222: fprintf(ficres," + V%d*age ",Tvar[j]);
13223: fprintf(ficlog," + V%d*age ",Tvar[j]);
13224: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13225: }else if(Typevar[j]==2) {
13226: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13227: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13228: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13229: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13230: }
13231: }
13232: printf("\n");
13233: fprintf(ficres,"\n");
13234: fprintf(ficlog,"\n");
13235: fprintf(fichtm, "</tr>");
13236: fprintf(fichtm, "\n");
13237:
13238:
1.126 brouard 13239: for(i=1,jk=1; i <=nlstate; i++){
13240: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 13241: if (k != i) {
1.319 brouard 13242: fprintf(fichtm, "<tr>");
1.225 brouard 13243: printf("%d%d ",i,k);
13244: fprintf(ficlog,"%d%d ",i,k);
13245: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 13246: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13247: for(j=1; j <=ncovmodel; j++){
13248: printf("%12.7f ",p[jk]);
13249: fprintf(ficlog,"%12.7f ",p[jk]);
13250: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 13251: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 13252: jk++;
13253: }
13254: printf("\n");
13255: fprintf(ficlog,"\n");
13256: fprintf(ficres,"\n");
1.319 brouard 13257: fprintf(fichtm, "</tr>\n");
1.225 brouard 13258: }
1.126 brouard 13259: }
13260: }
1.319 brouard 13261: /* fprintf(fichtm,"</tr>\n"); */
13262: fprintf(fichtm,"</table>\n");
13263: fprintf(fichtm, "\n");
13264:
1.203 brouard 13265: if(mle != 0){
13266: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 13267: ftolhess=ftol; /* Usually correct */
1.203 brouard 13268: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
13269: 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");
13270: 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 13271: 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 13272: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
13273: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
13274: if(nagesqr==1){
13275: printf(" + age*age ");
13276: fprintf(ficres," + age*age ");
13277: fprintf(ficlog," + age*age ");
13278: fprintf(fichtm, "<th>+ age*age</th>");
13279: }
13280: for(j=1;j <=ncovmodel-2;j++){
13281: if(Typevar[j]==0) {
13282: printf(" + V%d ",Tvar[j]);
13283: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13284: }else if(Typevar[j]==1) {
13285: printf(" + V%d*age ",Tvar[j]);
13286: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13287: }else if(Typevar[j]==2) {
13288: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13289: }
13290: }
13291: fprintf(fichtm, "</tr>\n");
13292:
1.203 brouard 13293: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 13294: for(k=1; k <=(nlstate+ndeath); k++){
13295: if (k != i) {
1.319 brouard 13296: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 13297: printf("%d%d ",i,k);
13298: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 13299: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13300: for(j=1; j <=ncovmodel; j++){
1.319 brouard 13301: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 13302: 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]));
13303: 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 13304: if(fabs(wald) > 1.96){
1.321 brouard 13305: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 13306: }else{
13307: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
13308: }
1.324 brouard 13309: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 13310: 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 13311: jk++;
13312: }
13313: printf("\n");
13314: fprintf(ficlog,"\n");
1.319 brouard 13315: fprintf(fichtm, "</tr>\n");
1.225 brouard 13316: }
13317: }
1.193 brouard 13318: }
1.203 brouard 13319: } /* end of hesscov and Wald tests */
1.319 brouard 13320: fprintf(fichtm,"</table>\n");
1.225 brouard 13321:
1.203 brouard 13322: /* */
1.126 brouard 13323: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
13324: printf("# Scales (for hessian or gradient estimation)\n");
13325: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
13326: for(i=1,jk=1; i <=nlstate; i++){
13327: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 13328: if (j!=i) {
13329: fprintf(ficres,"%1d%1d",i,j);
13330: printf("%1d%1d",i,j);
13331: fprintf(ficlog,"%1d%1d",i,j);
13332: for(k=1; k<=ncovmodel;k++){
13333: printf(" %.5e",delti[jk]);
13334: fprintf(ficlog," %.5e",delti[jk]);
13335: fprintf(ficres," %.5e",delti[jk]);
13336: jk++;
13337: }
13338: printf("\n");
13339: fprintf(ficlog,"\n");
13340: fprintf(ficres,"\n");
13341: }
1.126 brouard 13342: }
13343: }
13344:
13345: 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 13346: if(mle >= 1) /* To big for the screen */
1.126 brouard 13347: 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");
13348: 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");
13349: /* # 121 Var(a12)\n\ */
13350: /* # 122 Cov(b12,a12) Var(b12)\n\ */
13351: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
13352: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
13353: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
13354: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
13355: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
13356: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
13357:
13358:
13359: /* Just to have a covariance matrix which will be more understandable
13360: even is we still don't want to manage dictionary of variables
13361: */
13362: for(itimes=1;itimes<=2;itimes++){
13363: jj=0;
13364: for(i=1; i <=nlstate; i++){
1.225 brouard 13365: for(j=1; j <=nlstate+ndeath; j++){
13366: if(j==i) continue;
13367: for(k=1; k<=ncovmodel;k++){
13368: jj++;
13369: ca[0]= k+'a'-1;ca[1]='\0';
13370: if(itimes==1){
13371: if(mle>=1)
13372: printf("#%1d%1d%d",i,j,k);
13373: fprintf(ficlog,"#%1d%1d%d",i,j,k);
13374: fprintf(ficres,"#%1d%1d%d",i,j,k);
13375: }else{
13376: if(mle>=1)
13377: printf("%1d%1d%d",i,j,k);
13378: fprintf(ficlog,"%1d%1d%d",i,j,k);
13379: fprintf(ficres,"%1d%1d%d",i,j,k);
13380: }
13381: ll=0;
13382: for(li=1;li <=nlstate; li++){
13383: for(lj=1;lj <=nlstate+ndeath; lj++){
13384: if(lj==li) continue;
13385: for(lk=1;lk<=ncovmodel;lk++){
13386: ll++;
13387: if(ll<=jj){
13388: cb[0]= lk +'a'-1;cb[1]='\0';
13389: if(ll<jj){
13390: if(itimes==1){
13391: if(mle>=1)
13392: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13393: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13394: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13395: }else{
13396: if(mle>=1)
13397: printf(" %.5e",matcov[jj][ll]);
13398: fprintf(ficlog," %.5e",matcov[jj][ll]);
13399: fprintf(ficres," %.5e",matcov[jj][ll]);
13400: }
13401: }else{
13402: if(itimes==1){
13403: if(mle>=1)
13404: printf(" Var(%s%1d%1d)",ca,i,j);
13405: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
13406: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
13407: }else{
13408: if(mle>=1)
13409: printf(" %.7e",matcov[jj][ll]);
13410: fprintf(ficlog," %.7e",matcov[jj][ll]);
13411: fprintf(ficres," %.7e",matcov[jj][ll]);
13412: }
13413: }
13414: }
13415: } /* end lk */
13416: } /* end lj */
13417: } /* end li */
13418: if(mle>=1)
13419: printf("\n");
13420: fprintf(ficlog,"\n");
13421: fprintf(ficres,"\n");
13422: numlinepar++;
13423: } /* end k*/
13424: } /*end j */
1.126 brouard 13425: } /* end i */
13426: } /* end itimes */
13427:
13428: fflush(ficlog);
13429: fflush(ficres);
1.225 brouard 13430: while(fgets(line, MAXLINE, ficpar)) {
13431: /* If line starts with a # it is a comment */
13432: if (line[0] == '#') {
13433: numlinepar++;
13434: fputs(line,stdout);
13435: fputs(line,ficparo);
13436: fputs(line,ficlog);
1.299 brouard 13437: fputs(line,ficres);
1.225 brouard 13438: continue;
13439: }else
13440: break;
13441: }
13442:
1.209 brouard 13443: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
13444: /* ungetc(c,ficpar); */
13445: /* fgets(line, MAXLINE, ficpar); */
13446: /* fputs(line,stdout); */
13447: /* fputs(line,ficparo); */
13448: /* } */
13449: /* ungetc(c,ficpar); */
1.126 brouard 13450:
13451: estepm=0;
1.209 brouard 13452: 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 13453:
13454: if (num_filled != 6) {
13455: 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);
13456: 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);
13457: goto end;
13458: }
13459: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
13460: }
13461: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
13462: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
13463:
1.209 brouard 13464: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 13465: if (estepm==0 || estepm < stepm) estepm=stepm;
13466: if (fage <= 2) {
13467: bage = ageminpar;
13468: fage = agemaxpar;
13469: }
13470:
13471: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 13472: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
13473: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 13474:
1.186 brouard 13475: /* Other stuffs, more or less useful */
1.254 brouard 13476: while(fgets(line, MAXLINE, ficpar)) {
13477: /* If line starts with a # it is a comment */
13478: if (line[0] == '#') {
13479: numlinepar++;
13480: fputs(line,stdout);
13481: fputs(line,ficparo);
13482: fputs(line,ficlog);
1.299 brouard 13483: fputs(line,ficres);
1.254 brouard 13484: continue;
13485: }else
13486: break;
13487: }
13488:
13489: 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){
13490:
13491: if (num_filled != 7) {
13492: 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);
13493: 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);
13494: goto end;
13495: }
13496: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13497: 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);
13498: 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);
13499: 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 13500: }
1.254 brouard 13501:
13502: while(fgets(line, MAXLINE, ficpar)) {
13503: /* If line starts with a # it is a comment */
13504: if (line[0] == '#') {
13505: numlinepar++;
13506: fputs(line,stdout);
13507: fputs(line,ficparo);
13508: fputs(line,ficlog);
1.299 brouard 13509: fputs(line,ficres);
1.254 brouard 13510: continue;
13511: }else
13512: break;
1.126 brouard 13513: }
13514:
13515:
13516: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13517: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13518:
1.254 brouard 13519: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13520: if (num_filled != 1) {
13521: 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);
13522: 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);
13523: goto end;
13524: }
13525: printf("pop_based=%d\n",popbased);
13526: fprintf(ficlog,"pop_based=%d\n",popbased);
13527: fprintf(ficparo,"pop_based=%d\n",popbased);
13528: fprintf(ficres,"pop_based=%d\n",popbased);
13529: }
13530:
1.258 brouard 13531: /* Results */
1.332 brouard 13532: /* Value of covariate in each resultine will be compututed (if product) and sorted according to model rank */
13533: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
13534: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 13535: endishere=0;
1.258 brouard 13536: nresult=0;
1.308 brouard 13537: parameterline=0;
1.258 brouard 13538: do{
13539: if(!fgets(line, MAXLINE, ficpar)){
13540: endishere=1;
1.308 brouard 13541: parameterline=15;
1.258 brouard 13542: }else if (line[0] == '#') {
13543: /* If line starts with a # it is a comment */
1.254 brouard 13544: numlinepar++;
13545: fputs(line,stdout);
13546: fputs(line,ficparo);
13547: fputs(line,ficlog);
1.299 brouard 13548: fputs(line,ficres);
1.254 brouard 13549: continue;
1.258 brouard 13550: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13551: parameterline=11;
1.296 brouard 13552: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 13553: parameterline=12;
1.307 brouard 13554: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 13555: parameterline=13;
1.307 brouard 13556: }
1.258 brouard 13557: else{
13558: parameterline=14;
1.254 brouard 13559: }
1.308 brouard 13560: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 13561: case 11:
1.296 brouard 13562: 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)){
13563: 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 13564: 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);
13565: 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);
13566: 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);
13567: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 13568: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
13569: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 13570: prvforecast = 1;
13571: }
13572: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 13573: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13574: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13575: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 13576: prvforecast = 2;
13577: }
13578: else {
13579: 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);
13580: 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);
13581: goto end;
1.258 brouard 13582: }
1.254 brouard 13583: break;
1.258 brouard 13584: case 12:
1.296 brouard 13585: 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)){
13586: 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);
13587: 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);
13588: 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);
13589: 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);
13590: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 13591: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
13592: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 13593: prvbackcast = 1;
13594: }
13595: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 13596: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13597: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13598: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 13599: prvbackcast = 2;
13600: }
13601: else {
13602: 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);
13603: 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);
13604: goto end;
1.258 brouard 13605: }
1.230 brouard 13606: break;
1.258 brouard 13607: case 13:
1.332 brouard 13608: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 13609: nresult++; /* Sum of resultlines */
1.332 brouard 13610: printf("Result %d: result:%s\n",nresult, resultlineori);
13611: /* removefirstspace(&resultlineori); */
13612:
13613: if(strstr(resultlineori,"v") !=0){
13614: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
13615: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
13616: return 1;
13617: }
13618: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
13619: printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori);
1.318 brouard 13620: if(nresult > MAXRESULTLINESPONE-1){
13621: 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);
13622: 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 13623: goto end;
13624: }
1.332 brouard 13625:
1.310 brouard 13626: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 13627: fprintf(ficparo,"result: %s\n",resultline);
13628: fprintf(ficres,"result: %s\n",resultline);
13629: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 13630: } else
13631: goto end;
1.307 brouard 13632: break;
13633: case 14:
13634: printf("Error: Unknown command '%s'\n",line);
13635: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 13636: if(line[0] == ' ' || line[0] == '\n'){
13637: printf("It should not be an empty line '%s'\n",line);
13638: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
13639: }
1.307 brouard 13640: if(ncovmodel >=2 && nresult==0 ){
13641: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
13642: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 13643: }
1.307 brouard 13644: /* goto end; */
13645: break;
1.308 brouard 13646: case 15:
13647: printf("End of resultlines.\n");
13648: fprintf(ficlog,"End of resultlines.\n");
13649: break;
13650: default: /* parameterline =0 */
1.307 brouard 13651: nresult=1;
13652: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 13653: } /* End switch parameterline */
13654: }while(endishere==0); /* End do */
1.126 brouard 13655:
1.230 brouard 13656: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 13657: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 13658:
13659: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 13660: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 13661: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13662: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13663: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 13664: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13665: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13666: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13667: }else{
1.270 brouard 13668: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 13669: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
13670: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
13671: if(prvforecast==1){
13672: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
13673: jprojd=jproj1;
13674: mprojd=mproj1;
13675: anprojd=anproj1;
13676: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
13677: jprojf=jproj2;
13678: mprojf=mproj2;
13679: anprojf=anproj2;
13680: } else if(prvforecast == 2){
13681: dateprojd=dateintmean;
13682: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
13683: dateprojf=dateintmean+yrfproj;
13684: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
13685: }
13686: if(prvbackcast==1){
13687: datebackd=(jback1+12*mback1+365*anback1)/365;
13688: jbackd=jback1;
13689: mbackd=mback1;
13690: anbackd=anback1;
13691: datebackf=(jback2+12*mback2+365*anback2)/365;
13692: jbackf=jback2;
13693: mbackf=mback2;
13694: anbackf=anback2;
13695: } else if(prvbackcast == 2){
13696: datebackd=dateintmean;
13697: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
13698: datebackf=dateintmean-yrbproj;
13699: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
13700: }
13701:
13702: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 13703: }
13704: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 13705: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
13706: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 13707:
1.225 brouard 13708: /*------------ free_vector -------------*/
13709: /* chdir(path); */
1.220 brouard 13710:
1.215 brouard 13711: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
13712: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
13713: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
13714: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 13715: free_lvector(num,firstobs,lastobs);
13716: free_vector(agedc,firstobs,lastobs);
1.126 brouard 13717: /*free_matrix(covar,0,NCOVMAX,1,n);*/
13718: /*free_matrix(covar,1,NCOVMAX,1,n);*/
13719: fclose(ficparo);
13720: fclose(ficres);
1.220 brouard 13721:
13722:
1.186 brouard 13723: /* Other results (useful)*/
1.220 brouard 13724:
13725:
1.126 brouard 13726: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13727: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13728: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 13729: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 13730: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13731: fclose(ficrespl);
13732:
13733: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13734: /*#include "hpijx.h"*/
1.332 brouard 13735: /** 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?*/
13736: /* calls hpxij with combination k */
1.180 brouard 13737: hPijx(p, bage, fage);
1.145 brouard 13738: fclose(ficrespij);
1.227 brouard 13739:
1.220 brouard 13740: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 13741: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 13742: k=1;
1.126 brouard 13743: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13744:
1.269 brouard 13745: /* Prevalence for each covariate combination in probs[age][status][cov] */
13746: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13747: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 13748: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 13749: for(k=1;k<=ncovcombmax;k++)
13750: probs[i][j][k]=0.;
1.269 brouard 13751: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
13752: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 13753: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 13754: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13755: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 13756: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 13757: for(k=1;k<=ncovcombmax;k++)
13758: mobaverages[i][j][k]=0.;
1.219 brouard 13759: mobaverage=mobaverages;
13760: if (mobilav!=0) {
1.235 brouard 13761: printf("Movingaveraging observed prevalence\n");
1.258 brouard 13762: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 13763: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
13764: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
13765: printf(" Error in movingaverage mobilav=%d\n",mobilav);
13766: }
1.269 brouard 13767: } else if (mobilavproj !=0) {
1.235 brouard 13768: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 13769: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 13770: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
13771: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
13772: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
13773: }
1.269 brouard 13774: }else{
13775: printf("Internal error moving average\n");
13776: fflush(stdout);
13777: exit(1);
1.219 brouard 13778: }
13779: }/* end if moving average */
1.227 brouard 13780:
1.126 brouard 13781: /*---------- Forecasting ------------------*/
1.296 brouard 13782: if(prevfcast==1){
13783: /* /\* if(stepm ==1){*\/ */
13784: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13785: /*This done previously after freqsummary.*/
13786: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
13787: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
13788:
13789: /* } else if (prvforecast==2){ */
13790: /* /\* if(stepm ==1){*\/ */
13791: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13792: /* } */
13793: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
13794: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 13795: }
1.269 brouard 13796:
1.296 brouard 13797: /* Prevbcasting */
13798: if(prevbcast==1){
1.219 brouard 13799: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13800: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13801: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13802:
13803: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
13804:
13805: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 13806:
1.219 brouard 13807: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
13808: fclose(ficresplb);
13809:
1.222 brouard 13810: hBijx(p, bage, fage, mobaverage);
13811: fclose(ficrespijb);
1.219 brouard 13812:
1.296 brouard 13813: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
13814: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
13815: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
13816: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
13817: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
13818: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
13819:
13820:
1.269 brouard 13821: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13822:
13823:
1.269 brouard 13824: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 13825: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13826: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13827: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 13828: } /* end Prevbcasting */
1.268 brouard 13829:
1.186 brouard 13830:
13831: /* ------ Other prevalence ratios------------ */
1.126 brouard 13832:
1.215 brouard 13833: free_ivector(wav,1,imx);
13834: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
13835: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
13836: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 13837:
13838:
1.127 brouard 13839: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 13840:
1.201 brouard 13841: strcpy(filerese,"E_");
13842: strcat(filerese,fileresu);
1.126 brouard 13843: if((ficreseij=fopen(filerese,"w"))==NULL) {
13844: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13845: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13846: }
1.208 brouard 13847: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
13848: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 13849:
13850: pstamp(ficreseij);
1.219 brouard 13851:
1.235 brouard 13852: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13853: if (cptcovn < 1){i1=1;}
13854:
13855: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13856: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13857: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13858: continue;
1.219 brouard 13859: fprintf(ficreseij,"\n#****** ");
1.235 brouard 13860: printf("\n#****** ");
1.225 brouard 13861: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 13862: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
13863: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 13864: }
13865: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.337 ! brouard 13866: printf(" V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
! 13867: fprintf(ficreseij,"V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 13868: }
13869: fprintf(ficreseij,"******\n");
1.235 brouard 13870: printf("******\n");
1.219 brouard 13871:
13872: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13873: oldm=oldms;savm=savms;
1.330 brouard 13874: /* 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 13875: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 13876:
1.219 brouard 13877: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 13878: }
13879: fclose(ficreseij);
1.208 brouard 13880: printf("done evsij\n");fflush(stdout);
13881: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 13882:
1.218 brouard 13883:
1.227 brouard 13884: /*---------- State-specific expectancies and variances ------------*/
1.336 brouard 13885: /* Should be moved in a function */
1.201 brouard 13886: strcpy(filerest,"T_");
13887: strcat(filerest,fileresu);
1.127 brouard 13888: if((ficrest=fopen(filerest,"w"))==NULL) {
13889: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
13890: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
13891: }
1.208 brouard 13892: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
13893: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 13894: strcpy(fileresstde,"STDE_");
13895: strcat(fileresstde,fileresu);
1.126 brouard 13896: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 13897: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
13898: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 13899: }
1.227 brouard 13900: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
13901: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 13902:
1.201 brouard 13903: strcpy(filerescve,"CVE_");
13904: strcat(filerescve,fileresu);
1.126 brouard 13905: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 13906: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
13907: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 13908: }
1.227 brouard 13909: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
13910: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 13911:
1.201 brouard 13912: strcpy(fileresv,"V_");
13913: strcat(fileresv,fileresu);
1.126 brouard 13914: if((ficresvij=fopen(fileresv,"w"))==NULL) {
13915: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
13916: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
13917: }
1.227 brouard 13918: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
13919: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 13920:
1.235 brouard 13921: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13922: if (cptcovn < 1){i1=1;}
13923:
1.334 brouard 13924: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
13925: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
13926: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
13927: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
13928: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
13929: /* */
13930: 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 13931: continue;
1.321 brouard 13932: printf("\n# model %s \n#****** Result for:", model);
13933: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
13934: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.334 brouard 13935: /* It might not be a good idea to mix dummies and quantitative */
13936: /* 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 *\/ */
13937: 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 */
13938: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
13939: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
13940: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
13941: * (V5 is quanti) V4 and V3 are dummies
13942: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
13943: * l=1 l=2
13944: * k=1 1 1 0 0
13945: * k=2 2 1 1 0
13946: * k=3 [1] [2] 0 1
13947: * k=4 2 2 1 1
13948: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
13949: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
13950: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
13951: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
13952: */
13953: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
13954: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
13955: /* We give up with the combinations!! */
13956: 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 */
13957:
13958: if(Dummy[modelresult[nres][j]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to j in resultline */
1.337 ! brouard 13959: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][j]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
! 13960: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][j]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
! 13961: fprintf(ficrest,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][j]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
1.334 brouard 13962: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
13963: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
13964: }else{
13965: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
13966: }
13967: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13968: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
13969: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
13970: /* For each selected (single) quantitative value */
1.337 ! brouard 13971: printf(" V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
! 13972: fprintf(ficlog," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
! 13973: fprintf(ficrest," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
1.334 brouard 13974: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
13975: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
13976: }else{
13977: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
13978: }
13979: }else{
13980: 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 */
13981: 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 */
13982: exit(1);
13983: }
1.335 brouard 13984: } /* End loop for each variable in the resultline */
1.334 brouard 13985: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
13986: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
13987: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
13988: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
13989: /* } */
1.208 brouard 13990: fprintf(ficrest,"******\n");
1.227 brouard 13991: fprintf(ficlog,"******\n");
13992: printf("******\n");
1.208 brouard 13993:
13994: fprintf(ficresstdeij,"\n#****** ");
13995: fprintf(ficrescveij,"\n#****** ");
1.337 ! brouard 13996: /* It could have been: for(j=1;j<=cptcoveff;j++) {printf("V=%d=%lg",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);} */
! 13997: /* But it won't be sorted and depends on how the resultline is ordered */
1.225 brouard 13998: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 13999: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14000: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14001: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14002: }
14003: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value, TvarsQind gives the position of a quantitative in model equation */
1.337 ! brouard 14004: fprintf(ficresstdeij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
! 14005: fprintf(ficrescveij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 14006: }
1.208 brouard 14007: fprintf(ficresstdeij,"******\n");
14008: fprintf(ficrescveij,"******\n");
14009:
14010: fprintf(ficresvij,"\n#****** ");
1.238 brouard 14011: /* pstamp(ficresvij); */
1.225 brouard 14012: for(j=1;j<=cptcoveff;j++)
1.335 brouard 14013: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14014: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 14015: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 14016: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
1.337 ! brouard 14017: fprintf(ficresvij," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 14018: }
1.208 brouard 14019: fprintf(ficresvij,"******\n");
14020:
14021: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14022: oldm=oldms;savm=savms;
1.235 brouard 14023: printf(" cvevsij ");
14024: fprintf(ficlog, " cvevsij ");
14025: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 14026: printf(" end cvevsij \n ");
14027: fprintf(ficlog, " end cvevsij \n ");
14028:
14029: /*
14030: */
14031: /* goto endfree; */
14032:
14033: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14034: pstamp(ficrest);
14035:
1.269 brouard 14036: epj=vector(1,nlstate+1);
1.208 brouard 14037: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 14038: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
14039: cptcod= 0; /* To be deleted */
14040: printf("varevsij vpopbased=%d \n",vpopbased);
14041: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 14042: 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 14043: 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 ");
14044: if(vpopbased==1)
14045: 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);
14046: else
1.288 brouard 14047: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.335 brouard 14048: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 14049: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
14050: fprintf(ficrest,"\n");
14051: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 14052: printf("Computing age specific forward period (stable) prevalences in each health state \n");
14053: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 14054: for(age=bage; age <=fage ;age++){
1.235 brouard 14055: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 14056: if (vpopbased==1) {
14057: if(mobilav ==0){
14058: for(i=1; i<=nlstate;i++)
14059: prlim[i][i]=probs[(int)age][i][k];
14060: }else{ /* mobilav */
14061: for(i=1; i<=nlstate;i++)
14062: prlim[i][i]=mobaverage[(int)age][i][k];
14063: }
14064: }
1.219 brouard 14065:
1.227 brouard 14066: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
14067: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
14068: /* printf(" age %4.0f ",age); */
14069: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
14070: for(i=1, epj[j]=0.;i <=nlstate;i++) {
14071: epj[j] += prlim[i][i]*eij[i][j][(int)age];
14072: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
14073: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
14074: }
14075: epj[nlstate+1] +=epj[j];
14076: }
14077: /* printf(" age %4.0f \n",age); */
1.219 brouard 14078:
1.227 brouard 14079: for(i=1, vepp=0.;i <=nlstate;i++)
14080: for(j=1;j <=nlstate;j++)
14081: vepp += vareij[i][j][(int)age];
14082: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
14083: for(j=1;j <=nlstate;j++){
14084: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
14085: }
14086: fprintf(ficrest,"\n");
14087: }
1.208 brouard 14088: } /* End vpopbased */
1.269 brouard 14089: free_vector(epj,1,nlstate+1);
1.208 brouard 14090: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
14091: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 14092: printf("done selection\n");fflush(stdout);
14093: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 14094:
1.335 brouard 14095: } /* End k selection or end covariate selection for nres */
1.227 brouard 14096:
14097: printf("done State-specific expectancies\n");fflush(stdout);
14098: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
14099:
1.335 brouard 14100: /* variance-covariance of forward period prevalence */
1.269 brouard 14101: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14102:
1.227 brouard 14103:
1.290 brouard 14104: free_vector(weight,firstobs,lastobs);
1.330 brouard 14105: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 14106: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 14107: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
14108: free_matrix(anint,1,maxwav,firstobs,lastobs);
14109: free_matrix(mint,1,maxwav,firstobs,lastobs);
14110: free_ivector(cod,firstobs,lastobs);
1.227 brouard 14111: free_ivector(tab,1,NCOVMAX);
14112: fclose(ficresstdeij);
14113: fclose(ficrescveij);
14114: fclose(ficresvij);
14115: fclose(ficrest);
14116: fclose(ficpar);
14117:
14118:
1.126 brouard 14119: /*---------- End : free ----------------*/
1.219 brouard 14120: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 14121: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
14122: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 14123: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
14124: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 14125: } /* mle==-3 arrives here for freeing */
1.227 brouard 14126: /* endfree:*/
14127: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
14128: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
14129: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 14130: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
14131: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
14132: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
14133: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 14134: free_matrix(matcov,1,npar,1,npar);
14135: free_matrix(hess,1,npar,1,npar);
14136: /*free_vector(delti,1,npar);*/
14137: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14138: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 14139: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 14140: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14141:
14142: free_ivector(ncodemax,1,NCOVMAX);
14143: free_ivector(ncodemaxwundef,1,NCOVMAX);
14144: free_ivector(Dummy,-1,NCOVMAX);
14145: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 14146: free_ivector(DummyV,1,NCOVMAX);
14147: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 14148: free_ivector(Typevar,-1,NCOVMAX);
14149: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 14150: free_ivector(TvarsQ,1,NCOVMAX);
14151: free_ivector(TvarsQind,1,NCOVMAX);
14152: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 14153: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 14154: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 14155: free_ivector(TvarFD,1,NCOVMAX);
14156: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 14157: free_ivector(TvarF,1,NCOVMAX);
14158: free_ivector(TvarFind,1,NCOVMAX);
14159: free_ivector(TvarV,1,NCOVMAX);
14160: free_ivector(TvarVind,1,NCOVMAX);
14161: free_ivector(TvarA,1,NCOVMAX);
14162: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 14163: free_ivector(TvarFQ,1,NCOVMAX);
14164: free_ivector(TvarFQind,1,NCOVMAX);
14165: free_ivector(TvarVD,1,NCOVMAX);
14166: free_ivector(TvarVDind,1,NCOVMAX);
14167: free_ivector(TvarVQ,1,NCOVMAX);
14168: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 14169: free_ivector(Tvarsel,1,NCOVMAX);
14170: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 14171: free_ivector(Tposprod,1,NCOVMAX);
14172: free_ivector(Tprod,1,NCOVMAX);
14173: free_ivector(Tvaraff,1,NCOVMAX);
14174: free_ivector(invalidvarcomb,1,ncovcombmax);
14175: free_ivector(Tage,1,NCOVMAX);
14176: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 14177: free_ivector(TmodelInvind,1,NCOVMAX);
14178: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 14179:
14180: free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
14181:
1.227 brouard 14182: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
14183: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 14184: fflush(fichtm);
14185: fflush(ficgp);
14186:
1.227 brouard 14187:
1.126 brouard 14188: if((nberr >0) || (nbwarn>0)){
1.216 brouard 14189: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
14190: 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 14191: }else{
14192: printf("End of Imach\n");
14193: fprintf(ficlog,"End of Imach\n");
14194: }
14195: printf("See log file on %s\n",filelog);
14196: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 14197: /*(void) gettimeofday(&end_time,&tzp);*/
14198: rend_time = time(NULL);
14199: end_time = *localtime(&rend_time);
14200: /* tml = *localtime(&end_time.tm_sec); */
14201: strcpy(strtend,asctime(&end_time));
1.126 brouard 14202: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
14203: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 14204: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 14205:
1.157 brouard 14206: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
14207: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
14208: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 14209: /* printf("Total time was %d uSec.\n", total_usecs);*/
14210: /* if(fileappend(fichtm,optionfilehtm)){ */
14211: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14212: fclose(fichtm);
14213: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14214: fclose(fichtmcov);
14215: fclose(ficgp);
14216: fclose(ficlog);
14217: /*------ End -----------*/
1.227 brouard 14218:
1.281 brouard 14219:
14220: /* Executes gnuplot */
1.227 brouard 14221:
14222: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 14223: #ifdef WIN32
1.227 brouard 14224: if (_chdir(pathcd) != 0)
14225: printf("Can't move to directory %s!\n",path);
14226: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 14227: #else
1.227 brouard 14228: if(chdir(pathcd) != 0)
14229: printf("Can't move to directory %s!\n", path);
14230: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 14231: #endif
1.126 brouard 14232: printf("Current directory %s!\n",pathcd);
14233: /*strcat(plotcmd,CHARSEPARATOR);*/
14234: sprintf(plotcmd,"gnuplot");
1.157 brouard 14235: #ifdef _WIN32
1.126 brouard 14236: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
14237: #endif
14238: if(!stat(plotcmd,&info)){
1.158 brouard 14239: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14240: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 14241: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 14242: }else
14243: strcpy(pplotcmd,plotcmd);
1.157 brouard 14244: #ifdef __unix
1.126 brouard 14245: strcpy(plotcmd,GNUPLOTPROGRAM);
14246: if(!stat(plotcmd,&info)){
1.158 brouard 14247: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14248: }else
14249: strcpy(pplotcmd,plotcmd);
14250: #endif
14251: }else
14252: strcpy(pplotcmd,plotcmd);
14253:
14254: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 14255: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 14256: strcpy(pplotcmd,plotcmd);
1.227 brouard 14257:
1.126 brouard 14258: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 14259: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 14260: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 14261: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 14262: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 14263: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 14264: strcpy(plotcmd,pplotcmd);
14265: }
1.126 brouard 14266: }
1.158 brouard 14267: printf(" Successful, please wait...");
1.126 brouard 14268: while (z[0] != 'q') {
14269: /* chdir(path); */
1.154 brouard 14270: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 14271: scanf("%s",z);
14272: /* if (z[0] == 'c') system("./imach"); */
14273: if (z[0] == 'e') {
1.158 brouard 14274: #ifdef __APPLE__
1.152 brouard 14275: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 14276: #elif __linux
14277: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 14278: #else
1.152 brouard 14279: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 14280: #endif
14281: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
14282: system(pplotcmd);
1.126 brouard 14283: }
14284: else if (z[0] == 'g') system(plotcmd);
14285: else if (z[0] == 'q') exit(0);
14286: }
1.227 brouard 14287: end:
1.126 brouard 14288: while (z[0] != 'q') {
1.195 brouard 14289: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 14290: scanf("%s",z);
14291: }
1.283 brouard 14292: printf("End\n");
1.282 brouard 14293: exit(0);
1.126 brouard 14294: }
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