Annotation of imach/src/imach.c, revision 1.344
1.344 ! brouard 1: /* $Id: imach.c,v 1.343 2022/09/14 14:22:16 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.344 ! brouard 4: Revision 1.343 2022/09/14 14:22:16 brouard
! 5: Summary: version 0.99r39
! 6:
! 7: * imach.c (Module): Version 0.99r39 with colored dummy covariates
! 8: (fixed or time varying), using new last columns of
! 9: ILK_parameter.txt file.
! 10:
1.343 brouard 11: Revision 1.342 2022/09/11 19:54:09 brouard
12: Summary: 0.99r38
13:
14: * imach.c (Module): Adding timevarying products of any kinds,
15: should work before shifting cotvar from ncovcol+nqv columns in
16: order to have a correspondance between the column of cotvar and
17: the id of column.
18: (Module): Some cleaning and adding covariates in ILK.txt
19:
1.342 brouard 20: Revision 1.341 2022/09/11 07:58:42 brouard
21: Summary: Version 0.99r38
22:
23: After adding change in cotvar.
24:
1.341 brouard 25: Revision 1.340 2022/09/11 07:53:11 brouard
26: Summary: Version imach 0.99r37
27:
28: * imach.c (Module): Adding timevarying products of any kinds,
29: should work before shifting cotvar from ncovcol+nqv columns in
30: order to have a correspondance between the column of cotvar and
31: the id of column.
32:
1.340 brouard 33: Revision 1.339 2022/09/09 17:55:22 brouard
34: Summary: version 0.99r37
35:
36: * imach.c (Module): Many improvements for fixing products of fixed
37: timevarying as well as fixed * fixed, and test with quantitative
38: covariate.
39:
1.339 brouard 40: Revision 1.338 2022/09/04 17:40:33 brouard
41: Summary: 0.99r36
42:
43: * imach.c (Module): Now the easy runs i.e. without result or
44: model=1+age only did not work. The defautl combination should be 1
45: and not 0 because everything hasn't been tranformed yet.
46:
1.338 brouard 47: Revision 1.337 2022/09/02 14:26:02 brouard
48: Summary: version 0.99r35
49:
50: * src/imach.c: Version 0.99r35 because it outputs same results with
51: 1+age+V1+V1*age for females and 1+age for females only
52: (education=1 noweight)
53:
1.337 brouard 54: Revision 1.336 2022/08/31 09:52:36 brouard
55: *** empty log message ***
56:
1.336 brouard 57: Revision 1.335 2022/08/31 08:23:16 brouard
58: Summary: improvements...
59:
1.335 brouard 60: Revision 1.334 2022/08/25 09:08:41 brouard
61: Summary: In progress for quantitative
62:
1.334 brouard 63: Revision 1.333 2022/08/21 09:10:30 brouard
64: * src/imach.c (Module): Version 0.99r33 A lot of changes in
65: reassigning covariates: my first idea was that people will always
66: use the first covariate V1 into the model but in fact they are
67: producing data with many covariates and can use an equation model
68: with some of the covariate; it means that in a model V2+V3 instead
69: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
70: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
71: the equation model is restricted to two variables only (V2, V3)
72: and the combination for V2 should be codtabm(k,1) instead of
73: (codtabm(k,2), and the code should be
74: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
75: made. All of these should be simplified once a day like we did in
76: hpxij() for example by using precov[nres] which is computed in
77: decoderesult for each nres of each resultline. Loop should be done
78: on the equation model globally by distinguishing only product with
79: age (which are changing with age) and no more on type of
80: covariates, single dummies, single covariates.
81:
1.333 brouard 82: Revision 1.332 2022/08/21 09:06:25 brouard
83: Summary: Version 0.99r33
84:
85: * src/imach.c (Module): Version 0.99r33 A lot of changes in
86: reassigning covariates: my first idea was that people will always
87: use the first covariate V1 into the model but in fact they are
88: producing data with many covariates and can use an equation model
89: with some of the covariate; it means that in a model V2+V3 instead
90: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
91: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
92: the equation model is restricted to two variables only (V2, V3)
93: and the combination for V2 should be codtabm(k,1) instead of
94: (codtabm(k,2), and the code should be
95: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
96: made. All of these should be simplified once a day like we did in
97: hpxij() for example by using precov[nres] which is computed in
98: decoderesult for each nres of each resultline. Loop should be done
99: on the equation model globally by distinguishing only product with
100: age (which are changing with age) and no more on type of
101: covariates, single dummies, single covariates.
102:
1.332 brouard 103: Revision 1.331 2022/08/07 05:40:09 brouard
104: *** empty log message ***
105:
1.331 brouard 106: Revision 1.330 2022/08/06 07:18:25 brouard
107: Summary: last 0.99r31
108:
109: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
110:
1.330 brouard 111: Revision 1.329 2022/08/03 17:29:54 brouard
112: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
113:
1.329 brouard 114: Revision 1.328 2022/07/27 17:40:48 brouard
115: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
116:
1.328 brouard 117: Revision 1.327 2022/07/27 14:47:35 brouard
118: Summary: Still a problem for one-step probabilities in case of quantitative variables
119:
1.327 brouard 120: Revision 1.326 2022/07/26 17:33:55 brouard
121: Summary: some test with nres=1
122:
1.326 brouard 123: Revision 1.325 2022/07/25 14:27:23 brouard
124: Summary: r30
125:
126: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
127: coredumped, revealed by Feiuno, thank you.
128:
1.325 brouard 129: Revision 1.324 2022/07/23 17:44:26 brouard
130: *** empty log message ***
131:
1.324 brouard 132: Revision 1.323 2022/07/22 12:30:08 brouard
133: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
134:
1.323 brouard 135: Revision 1.322 2022/07/22 12:27:48 brouard
136: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
137:
1.322 brouard 138: Revision 1.321 2022/07/22 12:04:24 brouard
139: Summary: r28
140:
141: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
142:
1.321 brouard 143: Revision 1.320 2022/06/02 05:10:11 brouard
144: *** empty log message ***
145:
1.320 brouard 146: Revision 1.319 2022/06/02 04:45:11 brouard
147: * imach.c (Module): Adding the Wald tests from the log to the main
148: htm for better display of the maximum likelihood estimators.
149:
1.319 brouard 150: Revision 1.318 2022/05/24 08:10:59 brouard
151: * imach.c (Module): Some attempts to find a bug of wrong estimates
152: of confidencce intervals with product in the equation modelC
153:
1.318 brouard 154: Revision 1.317 2022/05/15 15:06:23 brouard
155: * imach.c (Module): Some minor improvements
156:
1.317 brouard 157: Revision 1.316 2022/05/11 15:11:31 brouard
158: Summary: r27
159:
1.316 brouard 160: Revision 1.315 2022/05/11 15:06:32 brouard
161: *** empty log message ***
162:
1.315 brouard 163: Revision 1.314 2022/04/13 17:43:09 brouard
164: * imach.c (Module): Adding link to text data files
165:
1.314 brouard 166: Revision 1.313 2022/04/11 15:57:42 brouard
167: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
168:
1.313 brouard 169: Revision 1.312 2022/04/05 21:24:39 brouard
170: *** empty log message ***
171:
1.312 brouard 172: Revision 1.311 2022/04/05 21:03:51 brouard
173: Summary: Fixed quantitative covariates
174:
175: Fixed covariates (dummy or quantitative)
176: with missing values have never been allowed but are ERRORS and
177: program quits. Standard deviations of fixed covariates were
178: wrongly computed. Mean and standard deviations of time varying
179: covariates are still not computed.
180:
1.311 brouard 181: Revision 1.310 2022/03/17 08:45:53 brouard
182: Summary: 99r25
183:
184: Improving detection of errors: result lines should be compatible with
185: the model.
186:
1.310 brouard 187: Revision 1.309 2021/05/20 12:39:14 brouard
188: Summary: Version 0.99r24
189:
1.309 brouard 190: Revision 1.308 2021/03/31 13:11:57 brouard
191: Summary: Version 0.99r23
192:
193:
194: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
195:
1.308 brouard 196: Revision 1.307 2021/03/08 18:11:32 brouard
197: Summary: 0.99r22 fixed bug on result:
198:
1.307 brouard 199: Revision 1.306 2021/02/20 15:44:02 brouard
200: Summary: Version 0.99r21
201:
202: * imach.c (Module): Fix bug on quitting after result lines!
203: (Module): Version 0.99r21
204:
1.306 brouard 205: Revision 1.305 2021/02/20 15:28:30 brouard
206: * imach.c (Module): Fix bug on quitting after result lines!
207:
1.305 brouard 208: Revision 1.304 2021/02/12 11:34:20 brouard
209: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
210:
1.304 brouard 211: Revision 1.303 2021/02/11 19:50:15 brouard
212: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
213:
1.303 brouard 214: Revision 1.302 2020/02/22 21:00:05 brouard
215: * (Module): imach.c Update mle=-3 (for computing Life expectancy
216: and life table from the data without any state)
217:
1.302 brouard 218: Revision 1.301 2019/06/04 13:51:20 brouard
219: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
220:
1.301 brouard 221: Revision 1.300 2019/05/22 19:09:45 brouard
222: Summary: version 0.99r19 of May 2019
223:
1.300 brouard 224: Revision 1.299 2019/05/22 18:37:08 brouard
225: Summary: Cleaned 0.99r19
226:
1.299 brouard 227: Revision 1.298 2019/05/22 18:19:56 brouard
228: *** empty log message ***
229:
1.298 brouard 230: Revision 1.297 2019/05/22 17:56:10 brouard
231: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
232:
1.297 brouard 233: Revision 1.296 2019/05/20 13:03:18 brouard
234: Summary: Projection syntax simplified
235:
236:
237: We can now start projections, forward or backward, from the mean date
238: of inteviews up to or down to a number of years of projection:
239: prevforecast=1 yearsfproj=15.3 mobil_average=0
240: or
241: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
242: or
243: prevbackcast=1 yearsbproj=12.3 mobil_average=1
244: or
245: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
246:
1.296 brouard 247: Revision 1.295 2019/05/18 09:52:50 brouard
248: Summary: doxygen tex bug
249:
1.295 brouard 250: Revision 1.294 2019/05/16 14:54:33 brouard
251: Summary: There was some wrong lines added
252:
1.294 brouard 253: Revision 1.293 2019/05/09 15:17:34 brouard
254: *** empty log message ***
255:
1.293 brouard 256: Revision 1.292 2019/05/09 14:17:20 brouard
257: Summary: Some updates
258:
1.292 brouard 259: Revision 1.291 2019/05/09 13:44:18 brouard
260: Summary: Before ncovmax
261:
1.291 brouard 262: Revision 1.290 2019/05/09 13:39:37 brouard
263: Summary: 0.99r18 unlimited number of individuals
264:
265: 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.
266:
1.290 brouard 267: Revision 1.289 2018/12/13 09:16:26 brouard
268: Summary: Bug for young ages (<-30) will be in r17
269:
1.289 brouard 270: Revision 1.288 2018/05/02 20:58:27 brouard
271: Summary: Some bugs fixed
272:
1.288 brouard 273: Revision 1.287 2018/05/01 17:57:25 brouard
274: Summary: Bug fixed by providing frequencies only for non missing covariates
275:
1.287 brouard 276: Revision 1.286 2018/04/27 14:27:04 brouard
277: Summary: some minor bugs
278:
1.286 brouard 279: Revision 1.285 2018/04/21 21:02:16 brouard
280: Summary: Some bugs fixed, valgrind tested
281:
1.285 brouard 282: Revision 1.284 2018/04/20 05:22:13 brouard
283: Summary: Computing mean and stdeviation of fixed quantitative variables
284:
1.284 brouard 285: Revision 1.283 2018/04/19 14:49:16 brouard
286: Summary: Some minor bugs fixed
287:
1.283 brouard 288: Revision 1.282 2018/02/27 22:50:02 brouard
289: *** empty log message ***
290:
1.282 brouard 291: Revision 1.281 2018/02/27 19:25:23 brouard
292: Summary: Adding second argument for quitting
293:
1.281 brouard 294: Revision 1.280 2018/02/21 07:58:13 brouard
295: Summary: 0.99r15
296:
297: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
298:
1.280 brouard 299: Revision 1.279 2017/07/20 13:35:01 brouard
300: Summary: temporary working
301:
1.279 brouard 302: Revision 1.278 2017/07/19 14:09:02 brouard
303: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
304:
1.278 brouard 305: Revision 1.277 2017/07/17 08:53:49 brouard
306: Summary: BOM files can be read now
307:
1.277 brouard 308: Revision 1.276 2017/06/30 15:48:31 brouard
309: Summary: Graphs improvements
310:
1.276 brouard 311: Revision 1.275 2017/06/30 13:39:33 brouard
312: Summary: Saito's color
313:
1.275 brouard 314: Revision 1.274 2017/06/29 09:47:08 brouard
315: Summary: Version 0.99r14
316:
1.274 brouard 317: Revision 1.273 2017/06/27 11:06:02 brouard
318: Summary: More documentation on projections
319:
1.273 brouard 320: Revision 1.272 2017/06/27 10:22:40 brouard
321: Summary: Color of backprojection changed from 6 to 5(yellow)
322:
1.272 brouard 323: Revision 1.271 2017/06/27 10:17:50 brouard
324: Summary: Some bug with rint
325:
1.271 brouard 326: Revision 1.270 2017/05/24 05:45:29 brouard
327: *** empty log message ***
328:
1.270 brouard 329: Revision 1.269 2017/05/23 08:39:25 brouard
330: Summary: Code into subroutine, cleanings
331:
1.269 brouard 332: Revision 1.268 2017/05/18 20:09:32 brouard
333: Summary: backprojection and confidence intervals of backprevalence
334:
1.268 brouard 335: Revision 1.267 2017/05/13 10:25:05 brouard
336: Summary: temporary save for backprojection
337:
1.267 brouard 338: Revision 1.266 2017/05/13 07:26:12 brouard
339: Summary: Version 0.99r13 (improvements and bugs fixed)
340:
1.266 brouard 341: Revision 1.265 2017/04/26 16:22:11 brouard
342: Summary: imach 0.99r13 Some bugs fixed
343:
1.265 brouard 344: Revision 1.264 2017/04/26 06:01:29 brouard
345: Summary: Labels in graphs
346:
1.264 brouard 347: Revision 1.263 2017/04/24 15:23:15 brouard
348: Summary: to save
349:
1.263 brouard 350: Revision 1.262 2017/04/18 16:48:12 brouard
351: *** empty log message ***
352:
1.262 brouard 353: Revision 1.261 2017/04/05 10:14:09 brouard
354: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
355:
1.261 brouard 356: Revision 1.260 2017/04/04 17:46:59 brouard
357: Summary: Gnuplot indexations fixed (humm)
358:
1.260 brouard 359: Revision 1.259 2017/04/04 13:01:16 brouard
360: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
361:
1.259 brouard 362: Revision 1.258 2017/04/03 10:17:47 brouard
363: Summary: Version 0.99r12
364:
365: Some cleanings, conformed with updated documentation.
366:
1.258 brouard 367: Revision 1.257 2017/03/29 16:53:30 brouard
368: Summary: Temp
369:
1.257 brouard 370: Revision 1.256 2017/03/27 05:50:23 brouard
371: Summary: Temporary
372:
1.256 brouard 373: Revision 1.255 2017/03/08 16:02:28 brouard
374: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
375:
1.255 brouard 376: Revision 1.254 2017/03/08 07:13:00 brouard
377: Summary: Fixing data parameter line
378:
1.254 brouard 379: Revision 1.253 2016/12/15 11:59:41 brouard
380: Summary: 0.99 in progress
381:
1.253 brouard 382: Revision 1.252 2016/09/15 21:15:37 brouard
383: *** empty log message ***
384:
1.252 brouard 385: Revision 1.251 2016/09/15 15:01:13 brouard
386: Summary: not working
387:
1.251 brouard 388: Revision 1.250 2016/09/08 16:07:27 brouard
389: Summary: continue
390:
1.250 brouard 391: Revision 1.249 2016/09/07 17:14:18 brouard
392: Summary: Starting values from frequencies
393:
1.249 brouard 394: Revision 1.248 2016/09/07 14:10:18 brouard
395: *** empty log message ***
396:
1.248 brouard 397: Revision 1.247 2016/09/02 11:11:21 brouard
398: *** empty log message ***
399:
1.247 brouard 400: Revision 1.246 2016/09/02 08:49:22 brouard
401: *** empty log message ***
402:
1.246 brouard 403: Revision 1.245 2016/09/02 07:25:01 brouard
404: *** empty log message ***
405:
1.245 brouard 406: Revision 1.244 2016/09/02 07:17:34 brouard
407: *** empty log message ***
408:
1.244 brouard 409: Revision 1.243 2016/09/02 06:45:35 brouard
410: *** empty log message ***
411:
1.243 brouard 412: Revision 1.242 2016/08/30 15:01:20 brouard
413: Summary: Fixing a lots
414:
1.242 brouard 415: Revision 1.241 2016/08/29 17:17:25 brouard
416: Summary: gnuplot problem in Back projection to fix
417:
1.241 brouard 418: Revision 1.240 2016/08/29 07:53:18 brouard
419: Summary: Better
420:
1.240 brouard 421: Revision 1.239 2016/08/26 15:51:03 brouard
422: Summary: Improvement in Powell output in order to copy and paste
423:
424: Author:
425:
1.239 brouard 426: Revision 1.238 2016/08/26 14:23:35 brouard
427: Summary: Starting tests of 0.99
428:
1.238 brouard 429: Revision 1.237 2016/08/26 09:20:19 brouard
430: Summary: to valgrind
431:
1.237 brouard 432: Revision 1.236 2016/08/25 10:50:18 brouard
433: *** empty log message ***
434:
1.236 brouard 435: Revision 1.235 2016/08/25 06:59:23 brouard
436: *** empty log message ***
437:
1.235 brouard 438: Revision 1.234 2016/08/23 16:51:20 brouard
439: *** empty log message ***
440:
1.234 brouard 441: Revision 1.233 2016/08/23 07:40:50 brouard
442: Summary: not working
443:
1.233 brouard 444: Revision 1.232 2016/08/22 14:20:21 brouard
445: Summary: not working
446:
1.232 brouard 447: Revision 1.231 2016/08/22 07:17:15 brouard
448: Summary: not working
449:
1.231 brouard 450: Revision 1.230 2016/08/22 06:55:53 brouard
451: Summary: Not working
452:
1.230 brouard 453: Revision 1.229 2016/07/23 09:45:53 brouard
454: Summary: Completing for func too
455:
1.229 brouard 456: Revision 1.228 2016/07/22 17:45:30 brouard
457: Summary: Fixing some arrays, still debugging
458:
1.227 brouard 459: Revision 1.226 2016/07/12 18:42:34 brouard
460: Summary: temp
461:
1.226 brouard 462: Revision 1.225 2016/07/12 08:40:03 brouard
463: Summary: saving but not running
464:
1.225 brouard 465: Revision 1.224 2016/07/01 13:16:01 brouard
466: Summary: Fixes
467:
1.224 brouard 468: Revision 1.223 2016/02/19 09:23:35 brouard
469: Summary: temporary
470:
1.223 brouard 471: Revision 1.222 2016/02/17 08:14:50 brouard
472: Summary: Probably last 0.98 stable version 0.98r6
473:
1.222 brouard 474: Revision 1.221 2016/02/15 23:35:36 brouard
475: Summary: minor bug
476:
1.220 brouard 477: Revision 1.219 2016/02/15 00:48:12 brouard
478: *** empty log message ***
479:
1.219 brouard 480: Revision 1.218 2016/02/12 11:29:23 brouard
481: Summary: 0.99 Back projections
482:
1.218 brouard 483: Revision 1.217 2015/12/23 17:18:31 brouard
484: Summary: Experimental backcast
485:
1.217 brouard 486: Revision 1.216 2015/12/18 17:32:11 brouard
487: Summary: 0.98r4 Warning and status=-2
488:
489: Version 0.98r4 is now:
490: - displaying an error when status is -1, date of interview unknown and date of death known;
491: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
492: Older changes concerning s=-2, dating from 2005 have been supersed.
493:
1.216 brouard 494: Revision 1.215 2015/12/16 08:52:24 brouard
495: Summary: 0.98r4 working
496:
1.215 brouard 497: Revision 1.214 2015/12/16 06:57:54 brouard
498: Summary: temporary not working
499:
1.214 brouard 500: Revision 1.213 2015/12/11 18:22:17 brouard
501: Summary: 0.98r4
502:
1.213 brouard 503: Revision 1.212 2015/11/21 12:47:24 brouard
504: Summary: minor typo
505:
1.212 brouard 506: Revision 1.211 2015/11/21 12:41:11 brouard
507: Summary: 0.98r3 with some graph of projected cross-sectional
508:
509: Author: Nicolas Brouard
510:
1.211 brouard 511: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 512: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 513: Summary: Adding ftolpl parameter
514: Author: N Brouard
515:
516: We had difficulties to get smoothed confidence intervals. It was due
517: to the period prevalence which wasn't computed accurately. The inner
518: parameter ftolpl is now an outer parameter of the .imach parameter
519: file after estepm. If ftolpl is small 1.e-4 and estepm too,
520: computation are long.
521:
1.209 brouard 522: Revision 1.208 2015/11/17 14:31:57 brouard
523: Summary: temporary
524:
1.208 brouard 525: Revision 1.207 2015/10/27 17:36:57 brouard
526: *** empty log message ***
527:
1.207 brouard 528: Revision 1.206 2015/10/24 07:14:11 brouard
529: *** empty log message ***
530:
1.206 brouard 531: Revision 1.205 2015/10/23 15:50:53 brouard
532: Summary: 0.98r3 some clarification for graphs on likelihood contributions
533:
1.205 brouard 534: Revision 1.204 2015/10/01 16:20:26 brouard
535: Summary: Some new graphs of contribution to likelihood
536:
1.204 brouard 537: Revision 1.203 2015/09/30 17:45:14 brouard
538: Summary: looking at better estimation of the hessian
539:
540: Also a better criteria for convergence to the period prevalence And
541: therefore adding the number of years needed to converge. (The
542: prevalence in any alive state shold sum to one
543:
1.203 brouard 544: Revision 1.202 2015/09/22 19:45:16 brouard
545: Summary: Adding some overall graph on contribution to likelihood. Might change
546:
1.202 brouard 547: Revision 1.201 2015/09/15 17:34:58 brouard
548: Summary: 0.98r0
549:
550: - Some new graphs like suvival functions
551: - Some bugs fixed like model=1+age+V2.
552:
1.201 brouard 553: Revision 1.200 2015/09/09 16:53:55 brouard
554: Summary: Big bug thanks to Flavia
555:
556: Even model=1+age+V2. did not work anymore
557:
1.200 brouard 558: Revision 1.199 2015/09/07 14:09:23 brouard
559: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
560:
1.199 brouard 561: Revision 1.198 2015/09/03 07:14:39 brouard
562: Summary: 0.98q5 Flavia
563:
1.198 brouard 564: Revision 1.197 2015/09/01 18:24:39 brouard
565: *** empty log message ***
566:
1.197 brouard 567: Revision 1.196 2015/08/18 23:17:52 brouard
568: Summary: 0.98q5
569:
1.196 brouard 570: Revision 1.195 2015/08/18 16:28:39 brouard
571: Summary: Adding a hack for testing purpose
572:
573: After reading the title, ftol and model lines, if the comment line has
574: a q, starting with #q, the answer at the end of the run is quit. It
575: permits to run test files in batch with ctest. The former workaround was
576: $ echo q | imach foo.imach
577:
1.195 brouard 578: Revision 1.194 2015/08/18 13:32:00 brouard
579: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
580:
1.194 brouard 581: Revision 1.193 2015/08/04 07:17:42 brouard
582: Summary: 0.98q4
583:
1.193 brouard 584: Revision 1.192 2015/07/16 16:49:02 brouard
585: Summary: Fixing some outputs
586:
1.192 brouard 587: Revision 1.191 2015/07/14 10:00:33 brouard
588: Summary: Some fixes
589:
1.191 brouard 590: Revision 1.190 2015/05/05 08:51:13 brouard
591: Summary: Adding digits in output parameters (7 digits instead of 6)
592:
593: Fix 1+age+.
594:
1.190 brouard 595: Revision 1.189 2015/04/30 14:45:16 brouard
596: Summary: 0.98q2
597:
1.189 brouard 598: Revision 1.188 2015/04/30 08:27:53 brouard
599: *** empty log message ***
600:
1.188 brouard 601: Revision 1.187 2015/04/29 09:11:15 brouard
602: *** empty log message ***
603:
1.187 brouard 604: Revision 1.186 2015/04/23 12:01:52 brouard
605: Summary: V1*age is working now, version 0.98q1
606:
607: Some codes had been disabled in order to simplify and Vn*age was
608: working in the optimization phase, ie, giving correct MLE parameters,
609: but, as usual, outputs were not correct and program core dumped.
610:
1.186 brouard 611: Revision 1.185 2015/03/11 13:26:42 brouard
612: Summary: Inclusion of compile and links command line for Intel Compiler
613:
1.185 brouard 614: Revision 1.184 2015/03/11 11:52:39 brouard
615: Summary: Back from Windows 8. Intel Compiler
616:
1.184 brouard 617: Revision 1.183 2015/03/10 20:34:32 brouard
618: Summary: 0.98q0, trying with directest, mnbrak fixed
619:
620: We use directest instead of original Powell test; probably no
621: incidence on the results, but better justifications;
622: We fixed Numerical Recipes mnbrak routine which was wrong and gave
623: wrong results.
624:
1.183 brouard 625: Revision 1.182 2015/02/12 08:19:57 brouard
626: Summary: Trying to keep directest which seems simpler and more general
627: Author: Nicolas Brouard
628:
1.182 brouard 629: Revision 1.181 2015/02/11 23:22:24 brouard
630: Summary: Comments on Powell added
631:
632: Author:
633:
1.181 brouard 634: Revision 1.180 2015/02/11 17:33:45 brouard
635: Summary: Finishing move from main to function (hpijx and prevalence_limit)
636:
1.180 brouard 637: Revision 1.179 2015/01/04 09:57:06 brouard
638: Summary: back to OS/X
639:
1.179 brouard 640: Revision 1.178 2015/01/04 09:35:48 brouard
641: *** empty log message ***
642:
1.178 brouard 643: Revision 1.177 2015/01/03 18:40:56 brouard
644: Summary: Still testing ilc32 on OSX
645:
1.177 brouard 646: Revision 1.176 2015/01/03 16:45:04 brouard
647: *** empty log message ***
648:
1.176 brouard 649: Revision 1.175 2015/01/03 16:33:42 brouard
650: *** empty log message ***
651:
1.175 brouard 652: Revision 1.174 2015/01/03 16:15:49 brouard
653: Summary: Still in cross-compilation
654:
1.174 brouard 655: Revision 1.173 2015/01/03 12:06:26 brouard
656: Summary: trying to detect cross-compilation
657:
1.173 brouard 658: Revision 1.172 2014/12/27 12:07:47 brouard
659: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
660:
1.172 brouard 661: Revision 1.171 2014/12/23 13:26:59 brouard
662: Summary: Back from Visual C
663:
664: Still problem with utsname.h on Windows
665:
1.171 brouard 666: Revision 1.170 2014/12/23 11:17:12 brouard
667: Summary: Cleaning some \%% back to %%
668:
669: The escape was mandatory for a specific compiler (which one?), but too many warnings.
670:
1.170 brouard 671: Revision 1.169 2014/12/22 23:08:31 brouard
672: Summary: 0.98p
673:
674: Outputs some informations on compiler used, OS etc. Testing on different platforms.
675:
1.169 brouard 676: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 677: Summary: update
1.169 brouard 678:
1.168 brouard 679: Revision 1.167 2014/12/22 13:50:56 brouard
680: Summary: Testing uname and compiler version and if compiled 32 or 64
681:
682: Testing on Linux 64
683:
1.167 brouard 684: Revision 1.166 2014/12/22 11:40:47 brouard
685: *** empty log message ***
686:
1.166 brouard 687: Revision 1.165 2014/12/16 11:20:36 brouard
688: Summary: After compiling on Visual C
689:
690: * imach.c (Module): Merging 1.61 to 1.162
691:
1.165 brouard 692: Revision 1.164 2014/12/16 10:52:11 brouard
693: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
694:
695: * imach.c (Module): Merging 1.61 to 1.162
696:
1.164 brouard 697: Revision 1.163 2014/12/16 10:30:11 brouard
698: * imach.c (Module): Merging 1.61 to 1.162
699:
1.163 brouard 700: Revision 1.162 2014/09/25 11:43:39 brouard
701: Summary: temporary backup 0.99!
702:
1.162 brouard 703: Revision 1.1 2014/09/16 11:06:58 brouard
704: Summary: With some code (wrong) for nlopt
705:
706: Author:
707:
708: Revision 1.161 2014/09/15 20:41:41 brouard
709: Summary: Problem with macro SQR on Intel compiler
710:
1.161 brouard 711: Revision 1.160 2014/09/02 09:24:05 brouard
712: *** empty log message ***
713:
1.160 brouard 714: Revision 1.159 2014/09/01 10:34:10 brouard
715: Summary: WIN32
716: Author: Brouard
717:
1.159 brouard 718: Revision 1.158 2014/08/27 17:11:51 brouard
719: *** empty log message ***
720:
1.158 brouard 721: Revision 1.157 2014/08/27 16:26:55 brouard
722: Summary: Preparing windows Visual studio version
723: Author: Brouard
724:
725: In order to compile on Visual studio, time.h is now correct and time_t
726: and tm struct should be used. difftime should be used but sometimes I
727: just make the differences in raw time format (time(&now).
728: Trying to suppress #ifdef LINUX
729: Add xdg-open for __linux in order to open default browser.
730:
1.157 brouard 731: Revision 1.156 2014/08/25 20:10:10 brouard
732: *** empty log message ***
733:
1.156 brouard 734: Revision 1.155 2014/08/25 18:32:34 brouard
735: Summary: New compile, minor changes
736: Author: Brouard
737:
1.155 brouard 738: Revision 1.154 2014/06/20 17:32:08 brouard
739: Summary: Outputs now all graphs of convergence to period prevalence
740:
1.154 brouard 741: Revision 1.153 2014/06/20 16:45:46 brouard
742: Summary: If 3 live state, convergence to period prevalence on same graph
743: Author: Brouard
744:
1.153 brouard 745: Revision 1.152 2014/06/18 17:54:09 brouard
746: Summary: open browser, use gnuplot on same dir than imach if not found in the path
747:
1.152 brouard 748: Revision 1.151 2014/06/18 16:43:30 brouard
749: *** empty log message ***
750:
1.151 brouard 751: Revision 1.150 2014/06/18 16:42:35 brouard
752: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
753: Author: brouard
754:
1.150 brouard 755: Revision 1.149 2014/06/18 15:51:14 brouard
756: Summary: Some fixes in parameter files errors
757: Author: Nicolas Brouard
758:
1.149 brouard 759: Revision 1.148 2014/06/17 17:38:48 brouard
760: Summary: Nothing new
761: Author: Brouard
762:
763: Just a new packaging for OS/X version 0.98nS
764:
1.148 brouard 765: Revision 1.147 2014/06/16 10:33:11 brouard
766: *** empty log message ***
767:
1.147 brouard 768: Revision 1.146 2014/06/16 10:20:28 brouard
769: Summary: Merge
770: Author: Brouard
771:
772: Merge, before building revised version.
773:
1.146 brouard 774: Revision 1.145 2014/06/10 21:23:15 brouard
775: Summary: Debugging with valgrind
776: Author: Nicolas Brouard
777:
778: Lot of changes in order to output the results with some covariates
779: After the Edimburgh REVES conference 2014, it seems mandatory to
780: improve the code.
781: No more memory valgrind error but a lot has to be done in order to
782: continue the work of splitting the code into subroutines.
783: Also, decodemodel has been improved. Tricode is still not
784: optimal. nbcode should be improved. Documentation has been added in
785: the source code.
786:
1.144 brouard 787: Revision 1.143 2014/01/26 09:45:38 brouard
788: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
789:
790: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
791: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
792:
1.143 brouard 793: Revision 1.142 2014/01/26 03:57:36 brouard
794: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
795:
796: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
797:
1.142 brouard 798: Revision 1.141 2014/01/26 02:42:01 brouard
799: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
800:
1.141 brouard 801: Revision 1.140 2011/09/02 10:37:54 brouard
802: Summary: times.h is ok with mingw32 now.
803:
1.140 brouard 804: Revision 1.139 2010/06/14 07:50:17 brouard
805: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
806: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
807:
1.139 brouard 808: Revision 1.138 2010/04/30 18:19:40 brouard
809: *** empty log message ***
810:
1.138 brouard 811: Revision 1.137 2010/04/29 18:11:38 brouard
812: (Module): Checking covariates for more complex models
813: than V1+V2. A lot of change to be done. Unstable.
814:
1.137 brouard 815: Revision 1.136 2010/04/26 20:30:53 brouard
816: (Module): merging some libgsl code. Fixing computation
817: of likelione (using inter/intrapolation if mle = 0) in order to
818: get same likelihood as if mle=1.
819: Some cleaning of code and comments added.
820:
1.136 brouard 821: Revision 1.135 2009/10/29 15:33:14 brouard
822: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
823:
1.135 brouard 824: Revision 1.134 2009/10/29 13:18:53 brouard
825: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
826:
1.134 brouard 827: Revision 1.133 2009/07/06 10:21:25 brouard
828: just nforces
829:
1.133 brouard 830: Revision 1.132 2009/07/06 08:22:05 brouard
831: Many tings
832:
1.132 brouard 833: Revision 1.131 2009/06/20 16:22:47 brouard
834: Some dimensions resccaled
835:
1.131 brouard 836: Revision 1.130 2009/05/26 06:44:34 brouard
837: (Module): Max Covariate is now set to 20 instead of 8. A
838: lot of cleaning with variables initialized to 0. Trying to make
839: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
840:
1.130 brouard 841: Revision 1.129 2007/08/31 13:49:27 lievre
842: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
843:
1.129 lievre 844: Revision 1.128 2006/06/30 13:02:05 brouard
845: (Module): Clarifications on computing e.j
846:
1.128 brouard 847: Revision 1.127 2006/04/28 18:11:50 brouard
848: (Module): Yes the sum of survivors was wrong since
849: imach-114 because nhstepm was no more computed in the age
850: loop. Now we define nhstepma in the age loop.
851: (Module): In order to speed up (in case of numerous covariates) we
852: compute health expectancies (without variances) in a first step
853: and then all the health expectancies with variances or standard
854: deviation (needs data from the Hessian matrices) which slows the
855: computation.
856: In the future we should be able to stop the program is only health
857: expectancies and graph are needed without standard deviations.
858:
1.127 brouard 859: Revision 1.126 2006/04/28 17:23:28 brouard
860: (Module): Yes the sum of survivors was wrong since
861: imach-114 because nhstepm was no more computed in the age
862: loop. Now we define nhstepma in the age loop.
863: Version 0.98h
864:
1.126 brouard 865: Revision 1.125 2006/04/04 15:20:31 lievre
866: Errors in calculation of health expectancies. Age was not initialized.
867: Forecasting file added.
868:
869: Revision 1.124 2006/03/22 17:13:53 lievre
870: Parameters are printed with %lf instead of %f (more numbers after the comma).
871: The log-likelihood is printed in the log file
872:
873: Revision 1.123 2006/03/20 10:52:43 brouard
874: * imach.c (Module): <title> changed, corresponds to .htm file
875: name. <head> headers where missing.
876:
877: * imach.c (Module): Weights can have a decimal point as for
878: English (a comma might work with a correct LC_NUMERIC environment,
879: otherwise the weight is truncated).
880: Modification of warning when the covariates values are not 0 or
881: 1.
882: Version 0.98g
883:
884: Revision 1.122 2006/03/20 09:45:41 brouard
885: (Module): Weights can have a decimal point as for
886: English (a comma might work with a correct LC_NUMERIC environment,
887: otherwise the weight is truncated).
888: Modification of warning when the covariates values are not 0 or
889: 1.
890: Version 0.98g
891:
892: Revision 1.121 2006/03/16 17:45:01 lievre
893: * imach.c (Module): Comments concerning covariates added
894:
895: * imach.c (Module): refinements in the computation of lli if
896: status=-2 in order to have more reliable computation if stepm is
897: not 1 month. Version 0.98f
898:
899: Revision 1.120 2006/03/16 15:10:38 lievre
900: (Module): refinements in the computation of lli if
901: status=-2 in order to have more reliable computation if stepm is
902: not 1 month. Version 0.98f
903:
904: Revision 1.119 2006/03/15 17:42:26 brouard
905: (Module): Bug if status = -2, the loglikelihood was
906: computed as likelihood omitting the logarithm. Version O.98e
907:
908: Revision 1.118 2006/03/14 18:20:07 brouard
909: (Module): varevsij Comments added explaining the second
910: table of variances if popbased=1 .
911: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
912: (Module): Function pstamp added
913: (Module): Version 0.98d
914:
915: Revision 1.117 2006/03/14 17:16:22 brouard
916: (Module): varevsij Comments added explaining the second
917: table of variances if popbased=1 .
918: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
919: (Module): Function pstamp added
920: (Module): Version 0.98d
921:
922: Revision 1.116 2006/03/06 10:29:27 brouard
923: (Module): Variance-covariance wrong links and
924: varian-covariance of ej. is needed (Saito).
925:
926: Revision 1.115 2006/02/27 12:17:45 brouard
927: (Module): One freematrix added in mlikeli! 0.98c
928:
929: Revision 1.114 2006/02/26 12:57:58 brouard
930: (Module): Some improvements in processing parameter
931: filename with strsep.
932:
933: Revision 1.113 2006/02/24 14:20:24 brouard
934: (Module): Memory leaks checks with valgrind and:
935: datafile was not closed, some imatrix were not freed and on matrix
936: allocation too.
937:
938: Revision 1.112 2006/01/30 09:55:26 brouard
939: (Module): Back to gnuplot.exe instead of wgnuplot.exe
940:
941: Revision 1.111 2006/01/25 20:38:18 brouard
942: (Module): Lots of cleaning and bugs added (Gompertz)
943: (Module): Comments can be added in data file. Missing date values
944: can be a simple dot '.'.
945:
946: Revision 1.110 2006/01/25 00:51:50 brouard
947: (Module): Lots of cleaning and bugs added (Gompertz)
948:
949: Revision 1.109 2006/01/24 19:37:15 brouard
950: (Module): Comments (lines starting with a #) are allowed in data.
951:
952: Revision 1.108 2006/01/19 18:05:42 lievre
953: Gnuplot problem appeared...
954: To be fixed
955:
956: Revision 1.107 2006/01/19 16:20:37 brouard
957: Test existence of gnuplot in imach path
958:
959: Revision 1.106 2006/01/19 13:24:36 brouard
960: Some cleaning and links added in html output
961:
962: Revision 1.105 2006/01/05 20:23:19 lievre
963: *** empty log message ***
964:
965: Revision 1.104 2005/09/30 16:11:43 lievre
966: (Module): sump fixed, loop imx fixed, and simplifications.
967: (Module): If the status is missing at the last wave but we know
968: that the person is alive, then we can code his/her status as -2
969: (instead of missing=-1 in earlier versions) and his/her
970: contributions to the likelihood is 1 - Prob of dying from last
971: health status (= 1-p13= p11+p12 in the easiest case of somebody in
972: the healthy state at last known wave). Version is 0.98
973:
974: Revision 1.103 2005/09/30 15:54:49 lievre
975: (Module): sump fixed, loop imx fixed, and simplifications.
976:
977: Revision 1.102 2004/09/15 17:31:30 brouard
978: Add the possibility to read data file including tab characters.
979:
980: Revision 1.101 2004/09/15 10:38:38 brouard
981: Fix on curr_time
982:
983: Revision 1.100 2004/07/12 18:29:06 brouard
984: Add version for Mac OS X. Just define UNIX in Makefile
985:
986: Revision 1.99 2004/06/05 08:57:40 brouard
987: *** empty log message ***
988:
989: Revision 1.98 2004/05/16 15:05:56 brouard
990: New version 0.97 . First attempt to estimate force of mortality
991: directly from the data i.e. without the need of knowing the health
992: state at each age, but using a Gompertz model: log u =a + b*age .
993: This is the basic analysis of mortality and should be done before any
994: other analysis, in order to test if the mortality estimated from the
995: cross-longitudinal survey is different from the mortality estimated
996: from other sources like vital statistic data.
997:
998: The same imach parameter file can be used but the option for mle should be -3.
999:
1.324 brouard 1000: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 1001: former routines in order to include the new code within the former code.
1002:
1003: The output is very simple: only an estimate of the intercept and of
1004: the slope with 95% confident intervals.
1005:
1006: Current limitations:
1007: A) Even if you enter covariates, i.e. with the
1008: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
1009: B) There is no computation of Life Expectancy nor Life Table.
1010:
1011: Revision 1.97 2004/02/20 13:25:42 lievre
1012: Version 0.96d. Population forecasting command line is (temporarily)
1013: suppressed.
1014:
1015: Revision 1.96 2003/07/15 15:38:55 brouard
1016: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
1017: rewritten within the same printf. Workaround: many printfs.
1018:
1019: Revision 1.95 2003/07/08 07:54:34 brouard
1020: * imach.c (Repository):
1021: (Repository): Using imachwizard code to output a more meaningful covariance
1022: matrix (cov(a12,c31) instead of numbers.
1023:
1024: Revision 1.94 2003/06/27 13:00:02 brouard
1025: Just cleaning
1026:
1027: Revision 1.93 2003/06/25 16:33:55 brouard
1028: (Module): On windows (cygwin) function asctime_r doesn't
1029: exist so I changed back to asctime which exists.
1030: (Module): Version 0.96b
1031:
1032: Revision 1.92 2003/06/25 16:30:45 brouard
1033: (Module): On windows (cygwin) function asctime_r doesn't
1034: exist so I changed back to asctime which exists.
1035:
1036: Revision 1.91 2003/06/25 15:30:29 brouard
1037: * imach.c (Repository): Duplicated warning errors corrected.
1038: (Repository): Elapsed time after each iteration is now output. It
1039: helps to forecast when convergence will be reached. Elapsed time
1040: is stamped in powell. We created a new html file for the graphs
1041: concerning matrix of covariance. It has extension -cov.htm.
1042:
1043: Revision 1.90 2003/06/24 12:34:15 brouard
1044: (Module): Some bugs corrected for windows. Also, when
1045: mle=-1 a template is output in file "or"mypar.txt with the design
1046: of the covariance matrix to be input.
1047:
1048: Revision 1.89 2003/06/24 12:30:52 brouard
1049: (Module): Some bugs corrected for windows. Also, when
1050: mle=-1 a template is output in file "or"mypar.txt with the design
1051: of the covariance matrix to be input.
1052:
1053: Revision 1.88 2003/06/23 17:54:56 brouard
1054: * 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.
1055:
1056: Revision 1.87 2003/06/18 12:26:01 brouard
1057: Version 0.96
1058:
1059: Revision 1.86 2003/06/17 20:04:08 brouard
1060: (Module): Change position of html and gnuplot routines and added
1061: routine fileappend.
1062:
1063: Revision 1.85 2003/06/17 13:12:43 brouard
1064: * imach.c (Repository): Check when date of death was earlier that
1065: current date of interview. It may happen when the death was just
1066: prior to the death. In this case, dh was negative and likelihood
1067: was wrong (infinity). We still send an "Error" but patch by
1068: assuming that the date of death was just one stepm after the
1069: interview.
1070: (Repository): Because some people have very long ID (first column)
1071: we changed int to long in num[] and we added a new lvector for
1072: memory allocation. But we also truncated to 8 characters (left
1073: truncation)
1074: (Repository): No more line truncation errors.
1075:
1076: Revision 1.84 2003/06/13 21:44:43 brouard
1077: * imach.c (Repository): Replace "freqsummary" at a correct
1078: place. It differs from routine "prevalence" which may be called
1079: many times. Probs is memory consuming and must be used with
1080: parcimony.
1081: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1082:
1083: Revision 1.83 2003/06/10 13:39:11 lievre
1084: *** empty log message ***
1085:
1086: Revision 1.82 2003/06/05 15:57:20 brouard
1087: Add log in imach.c and fullversion number is now printed.
1088:
1089: */
1090: /*
1091: Interpolated Markov Chain
1092:
1093: Short summary of the programme:
1094:
1.227 brouard 1095: This program computes Healthy Life Expectancies or State-specific
1096: (if states aren't health statuses) Expectancies from
1097: cross-longitudinal data. Cross-longitudinal data consist in:
1098:
1099: -1- a first survey ("cross") where individuals from different ages
1100: are interviewed on their health status or degree of disability (in
1101: the case of a health survey which is our main interest)
1102:
1103: -2- at least a second wave of interviews ("longitudinal") which
1104: measure each change (if any) in individual health status. Health
1105: expectancies are computed from the time spent in each health state
1106: according to a model. More health states you consider, more time is
1107: necessary to reach the Maximum Likelihood of the parameters involved
1108: in the model. The simplest model is the multinomial logistic model
1109: where pij is the probability to be observed in state j at the second
1110: wave conditional to be observed in state i at the first
1111: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1112: etc , where 'age' is age and 'sex' is a covariate. If you want to
1113: have a more complex model than "constant and age", you should modify
1114: the program where the markup *Covariates have to be included here
1115: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1116: convergence.
1117:
1118: The advantage of this computer programme, compared to a simple
1119: multinomial logistic model, is clear when the delay between waves is not
1120: identical for each individual. Also, if a individual missed an
1121: intermediate interview, the information is lost, but taken into
1122: account using an interpolation or extrapolation.
1123:
1124: hPijx is the probability to be observed in state i at age x+h
1125: conditional to the observed state i at age x. The delay 'h' can be
1126: split into an exact number (nh*stepm) of unobserved intermediate
1127: states. This elementary transition (by month, quarter,
1128: semester or year) is modelled as a multinomial logistic. The hPx
1129: matrix is simply the matrix product of nh*stepm elementary matrices
1130: and the contribution of each individual to the likelihood is simply
1131: hPijx.
1132:
1133: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1134: of the life expectancies. It also computes the period (stable) prevalence.
1135:
1136: Back prevalence and projections:
1.227 brouard 1137:
1138: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1139: double agemaxpar, double ftolpl, int *ncvyearp, double
1140: dateprev1,double dateprev2, int firstpass, int lastpass, int
1141: mobilavproj)
1142:
1143: Computes the back prevalence limit for any combination of
1144: covariate values k at any age between ageminpar and agemaxpar and
1145: returns it in **bprlim. In the loops,
1146:
1147: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1148: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1149:
1150: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1151: Computes for any combination of covariates k and any age between bage and fage
1152: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1153: oldm=oldms;savm=savms;
1.227 brouard 1154:
1.267 brouard 1155: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1156: Computes the transition matrix starting at age 'age' over
1157: 'nhstepm*hstepm*stepm' months (i.e. until
1158: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1159: nhstepm*hstepm matrices.
1160:
1161: Returns p3mat[i][j][h] after calling
1162: p3mat[i][j][h]=matprod2(newm,
1163: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1164: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1165: oldm);
1.226 brouard 1166:
1167: Important routines
1168:
1169: - func (or funcone), computes logit (pij) distinguishing
1170: o fixed variables (single or product dummies or quantitative);
1171: o varying variables by:
1172: (1) wave (single, product dummies, quantitative),
1173: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1174: % fixed dummy (treated) or quantitative (not done because time-consuming);
1175: % varying dummy (not done) or quantitative (not done);
1176: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1177: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1178: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1179: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1180: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1181:
1.226 brouard 1182:
1183:
1.324 brouard 1184: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1185: Institut national d'études démographiques, Paris.
1.126 brouard 1186: This software have been partly granted by Euro-REVES, a concerted action
1187: from the European Union.
1188: It is copyrighted identically to a GNU software product, ie programme and
1189: software can be distributed freely for non commercial use. Latest version
1190: can be accessed at http://euroreves.ined.fr/imach .
1191:
1192: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1193: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1194:
1195: **********************************************************************/
1196: /*
1197: main
1198: read parameterfile
1199: read datafile
1200: concatwav
1201: freqsummary
1202: if (mle >= 1)
1203: mlikeli
1204: print results files
1205: if mle==1
1206: computes hessian
1207: read end of parameter file: agemin, agemax, bage, fage, estepm
1208: begin-prev-date,...
1209: open gnuplot file
1210: open html file
1.145 brouard 1211: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1212: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1213: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1214: freexexit2 possible for memory heap.
1215:
1216: h Pij x | pij_nom ficrestpij
1217: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1218: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1219: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1220:
1221: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1222: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1223: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1224: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1225: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1226:
1.126 brouard 1227: forecasting if prevfcast==1 prevforecast call prevalence()
1228: health expectancies
1229: Variance-covariance of DFLE
1230: prevalence()
1231: movingaverage()
1232: varevsij()
1233: if popbased==1 varevsij(,popbased)
1234: total life expectancies
1235: Variance of period (stable) prevalence
1236: end
1237: */
1238:
1.187 brouard 1239: /* #define DEBUG */
1240: /* #define DEBUGBRENT */
1.203 brouard 1241: /* #define DEBUGLINMIN */
1242: /* #define DEBUGHESS */
1243: #define DEBUGHESSIJ
1.224 brouard 1244: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1245: #define POWELL /* Instead of NLOPT */
1.224 brouard 1246: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1247: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1248: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1249: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1250:
1251: #include <math.h>
1252: #include <stdio.h>
1253: #include <stdlib.h>
1254: #include <string.h>
1.226 brouard 1255: #include <ctype.h>
1.159 brouard 1256:
1257: #ifdef _WIN32
1258: #include <io.h>
1.172 brouard 1259: #include <windows.h>
1260: #include <tchar.h>
1.159 brouard 1261: #else
1.126 brouard 1262: #include <unistd.h>
1.159 brouard 1263: #endif
1.126 brouard 1264:
1265: #include <limits.h>
1266: #include <sys/types.h>
1.171 brouard 1267:
1268: #if defined(__GNUC__)
1269: #include <sys/utsname.h> /* Doesn't work on Windows */
1270: #endif
1271:
1.126 brouard 1272: #include <sys/stat.h>
1273: #include <errno.h>
1.159 brouard 1274: /* extern int errno; */
1.126 brouard 1275:
1.157 brouard 1276: /* #ifdef LINUX */
1277: /* #include <time.h> */
1278: /* #include "timeval.h" */
1279: /* #else */
1280: /* #include <sys/time.h> */
1281: /* #endif */
1282:
1.126 brouard 1283: #include <time.h>
1284:
1.136 brouard 1285: #ifdef GSL
1286: #include <gsl/gsl_errno.h>
1287: #include <gsl/gsl_multimin.h>
1288: #endif
1289:
1.167 brouard 1290:
1.162 brouard 1291: #ifdef NLOPT
1292: #include <nlopt.h>
1293: typedef struct {
1294: double (* function)(double [] );
1295: } myfunc_data ;
1296: #endif
1297:
1.126 brouard 1298: /* #include <libintl.h> */
1299: /* #define _(String) gettext (String) */
1300:
1.251 brouard 1301: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1302:
1303: #define GNUPLOTPROGRAM "gnuplot"
1.343 brouard 1304: #define GNUPLOTVERSION 5.1
1305: double gnuplotversion=GNUPLOTVERSION;
1.126 brouard 1306: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1307: #define FILENAMELENGTH 256
1.126 brouard 1308:
1309: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1310: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1311:
1.144 brouard 1312: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1313: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1314:
1315: #define NINTERVMAX 8
1.144 brouard 1316: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1317: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1318: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1319: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1320: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1321: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1322: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1323: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1324: /* #define AGESUP 130 */
1.288 brouard 1325: /* #define AGESUP 150 */
1326: #define AGESUP 200
1.268 brouard 1327: #define AGEINF 0
1.218 brouard 1328: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1329: #define AGEBASE 40
1.194 brouard 1330: #define AGEOVERFLOW 1.e20
1.164 brouard 1331: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1332: #ifdef _WIN32
1333: #define DIRSEPARATOR '\\'
1334: #define CHARSEPARATOR "\\"
1335: #define ODIRSEPARATOR '/'
1336: #else
1.126 brouard 1337: #define DIRSEPARATOR '/'
1338: #define CHARSEPARATOR "/"
1339: #define ODIRSEPARATOR '\\'
1340: #endif
1341:
1.344 ! brouard 1342: /* $Id: imach.c,v 1.343 2022/09/14 14:22:16 brouard Exp $ */
1.126 brouard 1343: /* $State: Exp $ */
1.196 brouard 1344: #include "version.h"
1345: char version[]=__IMACH_VERSION__;
1.337 brouard 1346: char copyright[]="September 2022,INED-EUROREVES-Institut de longevite-Japan Society for the Promotion of Science (Grant-in-Aid for Scientific Research 25293121), Intel Software 2015-2020, Nihon University 2021-202, INED 2000-2022";
1.344 ! brouard 1347: char fullversion[]="$Revision: 1.343 $ $Date: 2022/09/14 14:22:16 $";
1.126 brouard 1348: char strstart[80];
1349: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1350: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.342 brouard 1351: int debugILK=0; /* debugILK is set by a #d in a comment line */
1.187 brouard 1352: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1353: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1354: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1.335 brouard 1355: 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 1356: 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 1357: int cptcovs=0; /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1358: int cptcovsnq=0; /**< cptcovsnq number of SIMPLE covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1359: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1360: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.335 brouard 1361: 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 1362: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1363: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.339 brouard 1364: int ncovvt=0; /* Total number of effective (wave) varying covariates (dummy or quantitative or products [without age]) in the model */
1.232 brouard 1365: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1366: int nsd=0; /**< Total number of single dummy variables (output) */
1367: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1368: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1369: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1370: int ntveff=0; /**< ntveff number of effective time varying variables */
1371: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1372: int cptcov=0; /* Working variable */
1.334 brouard 1373: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs+1 declared globally ;*/
1.290 brouard 1374: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1375: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1376: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1377: int nlstate=2; /* Number of live states */
1378: int ndeath=1; /* Number of dead states */
1.130 brouard 1379: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.339 brouard 1380: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable*/
1381: int ncovcolt=0; /* ncovcolt=ncovcol+nqv+ntv+nqtv; total of covariates in the data, not in the model equation*/
1.126 brouard 1382: int popbased=0;
1383:
1384: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1385: int maxwav=0; /* Maxim number of waves */
1386: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1387: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1388: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1389: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1390: int mle=1, weightopt=0;
1.126 brouard 1391: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1392: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1393: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1394: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1395: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1396: int selected(int kvar); /* Is covariate kvar selected for printing results */
1397:
1.130 brouard 1398: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1399: double **matprod2(); /* test */
1.126 brouard 1400: double **oldm, **newm, **savm; /* Working pointers to matrices */
1401: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1402: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1403:
1.136 brouard 1404: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1405: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1406: FILE *ficlog, *ficrespow;
1.130 brouard 1407: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1408: double fretone; /* Only one call to likelihood */
1.130 brouard 1409: long ipmx=0; /* Number of contributions */
1.126 brouard 1410: double sw; /* Sum of weights */
1411: char filerespow[FILENAMELENGTH];
1412: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1413: FILE *ficresilk;
1414: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1415: FILE *ficresprobmorprev;
1416: FILE *fichtm, *fichtmcov; /* Html File */
1417: FILE *ficreseij;
1418: char filerese[FILENAMELENGTH];
1419: FILE *ficresstdeij;
1420: char fileresstde[FILENAMELENGTH];
1421: FILE *ficrescveij;
1422: char filerescve[FILENAMELENGTH];
1423: FILE *ficresvij;
1424: char fileresv[FILENAMELENGTH];
1.269 brouard 1425:
1.126 brouard 1426: char title[MAXLINE];
1.234 brouard 1427: char model[MAXLINE]; /**< The model line */
1.217 brouard 1428: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1429: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1430: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1431: char command[FILENAMELENGTH];
1432: int outcmd=0;
1433:
1.217 brouard 1434: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1435: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1436: char filelog[FILENAMELENGTH]; /* Log file */
1437: char filerest[FILENAMELENGTH];
1438: char fileregp[FILENAMELENGTH];
1439: char popfile[FILENAMELENGTH];
1440:
1441: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1442:
1.157 brouard 1443: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1444: /* struct timezone tzp; */
1445: /* extern int gettimeofday(); */
1446: struct tm tml, *gmtime(), *localtime();
1447:
1448: extern time_t time();
1449:
1450: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1451: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1452: struct tm tm;
1453:
1.126 brouard 1454: char strcurr[80], strfor[80];
1455:
1456: char *endptr;
1457: long lval;
1458: double dval;
1459:
1460: #define NR_END 1
1461: #define FREE_ARG char*
1462: #define FTOL 1.0e-10
1463:
1464: #define NRANSI
1.240 brouard 1465: #define ITMAX 200
1466: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1467:
1468: #define TOL 2.0e-4
1469:
1470: #define CGOLD 0.3819660
1471: #define ZEPS 1.0e-10
1472: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1473:
1474: #define GOLD 1.618034
1475: #define GLIMIT 100.0
1476: #define TINY 1.0e-20
1477:
1478: static double maxarg1,maxarg2;
1479: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1480: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1481:
1482: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1483: #define rint(a) floor(a+0.5)
1.166 brouard 1484: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1485: #define mytinydouble 1.0e-16
1.166 brouard 1486: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1487: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1488: /* static double dsqrarg; */
1489: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1490: static double sqrarg;
1491: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1492: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1493: int agegomp= AGEGOMP;
1494:
1495: int imx;
1496: int stepm=1;
1497: /* Stepm, step in month: minimum step interpolation*/
1498:
1499: int estepm;
1500: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1501:
1502: int m,nb;
1503: long *num;
1.197 brouard 1504: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1505: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1506: covariate for which somebody answered excluding
1507: undefined. Usually 2: 0 and 1. */
1508: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1509: covariate for which somebody answered including
1510: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1511: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1512: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1513: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1514: 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 1515: double *ageexmed,*agecens;
1516: double dateintmean=0;
1.296 brouard 1517: double anprojd, mprojd, jprojd; /* For eventual projections */
1518: double anprojf, mprojf, jprojf;
1.126 brouard 1519:
1.296 brouard 1520: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1521: double anbackf, mbackf, jbackf;
1522: double jintmean,mintmean,aintmean;
1.126 brouard 1523: double *weight;
1524: int **s; /* Status */
1.141 brouard 1525: double *agedc;
1.145 brouard 1526: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1527: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1528: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1529: double **coqvar; /* Fixed quantitative covariate nqv */
1.341 brouard 1530: double ***cotvar; /* Time varying covariate start at ncovcol + nqv + (1 to ntv) */
1.225 brouard 1531: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1532: double idx;
1533: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1534: /* Some documentation */
1535: /* Design original data
1536: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1537: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1538: * ntv=3 nqtv=1
1.330 brouard 1539: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1540: * For time varying covariate, quanti or dummies
1541: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1.341 brouard 1542: * cotvar[wav][ncovcol+nqv+ iv(1 to nqtv)][i]= [(1 to nqtv)][i]=(V12) quanti
1.319 brouard 1543: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1544: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1545: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1546: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1547: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1548: * k= 1 2 3 4 5 6 7 8 9 10 11
1549: */
1550: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1551: /* 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
1552: # States 1=Coresidence, 2 Living alone, 3 Institution
1553: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1554: */
1.343 brouard 1555: /* V5+V4+ V3+V4*V3 +V5*age+V2 +V1*V2+V1*age+V1 */
1556: /* kmodel 1 2 3 4 5 6 7 8 9 */
1.319 brouard 1557: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1558: /* fixed or varying), 1 for age product, 2 for*/
1559: /* product */
1560: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1561: /*(single or product without age), 2 dummy*/
1562: /* with age product, 3 quant with age product*/
1563: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1564: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 brouard 1565: /*TnsdVar[Tvar] 1 2 3 */
1.337 brouard 1566: /*Tvaraff[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.319 brouard 1567: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1.338 brouard 1568: /*TvarsDind[nsd] 2 3 9 */ /* position K of single dummy cova */
1.319 brouard 1569: /* nsq 1 2 */ /* Counting single quantit tv */
1570: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1571: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1572: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1573: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1574: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1575: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1576: /* 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 1577: /* 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 1578: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1579: /* Type */
1580: /* V 1 2 3 4 5 */
1581: /* F F V V V */
1582: /* D Q D D Q */
1583: /* */
1584: int *TvarsD;
1.330 brouard 1585: int *TnsdVar;
1.234 brouard 1586: int *TvarsDind;
1587: int *TvarsQ;
1588: int *TvarsQind;
1589:
1.318 brouard 1590: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1591: int nresult=0;
1.258 brouard 1592: int parameterline=0; /* # of the parameter (type) line */
1.334 brouard 1593: int TKresult[MAXRESULTLINESPONE]; /* TKresult[nres]=k for each resultline nres give the corresponding combination of dummies */
1594: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model corresponds to the k3 position in the resultline */
1595: int modelresult[MAXRESULTLINESPONE][NCOVMAX];/* modelresult[k3]=k1: k1th position in the model corresponds to the k3 position in the resultline */
1596: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1597: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1598: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.334 brouard 1599: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 1600: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1601: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1602: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1603:
1604: /* 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
1605: # States 1=Coresidence, 2 Living alone, 3 Institution
1606: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1607: */
1.234 brouard 1608: /* 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 1609: 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 */
1610: 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 */
1611: 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 */
1612: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1613: 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 */
1614: 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 1615: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1616: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1617: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1618: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1619: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1620: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1621: 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 */
1622: int *TvarVQind; /* TvarVQind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple time varying quantitative variable */
1.339 brouard 1623: int *TvarVV; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1624: int *TvarVVind; /* We count ncovvt time varying covariates (single or products without age) and put their name into TvarVV */
1625: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
1626: /* model V1+V3+age*V1+age*V3+V1*V3 */
1627: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
1628: /* TvarVV={3,1,3}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1629: /* TvarVVind={2,5,5}, for V3 and then the product V1*V3 is decomposed into V1 and V3 */
1.230 brouard 1630: int *Tvarsel; /**< Selected covariates for output */
1631: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1632: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1633: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1634: 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 1635: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1636: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1637: int *Tage;
1.227 brouard 1638: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1639: 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 1640: 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*/
1641: 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 1642: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1643: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1644: int **Tvard;
1.330 brouard 1645: int **Tvardk;
1.227 brouard 1646: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1647: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1648: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1649: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1650: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1651: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1652: double *lsurv, *lpop, *tpop;
1653:
1.231 brouard 1654: #define FD 1; /* Fixed dummy covariate */
1655: #define FQ 2; /* Fixed quantitative covariate */
1656: #define FP 3; /* Fixed product covariate */
1657: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1658: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1659: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1660: #define VD 10; /* Varying dummy covariate */
1661: #define VQ 11; /* Varying quantitative covariate */
1662: #define VP 12; /* Varying product covariate */
1663: #define VPDD 13; /* Varying product dummy*dummy covariate */
1664: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1665: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1666: #define APFD 16; /* Age product * fixed dummy covariate */
1667: #define APFQ 17; /* Age product * fixed quantitative covariate */
1668: #define APVD 18; /* Age product * varying dummy covariate */
1669: #define APVQ 19; /* Age product * varying quantitative covariate */
1670:
1671: #define FTYPE 1; /* Fixed covariate */
1672: #define VTYPE 2; /* Varying covariate (loop in wave) */
1673: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1674:
1675: struct kmodel{
1676: int maintype; /* main type */
1677: int subtype; /* subtype */
1678: };
1679: struct kmodel modell[NCOVMAX];
1680:
1.143 brouard 1681: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1682: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1683:
1684: /**************** split *************************/
1685: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1686: {
1687: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1688: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1689: */
1690: char *ss; /* pointer */
1.186 brouard 1691: int l1=0, l2=0; /* length counters */
1.126 brouard 1692:
1693: l1 = strlen(path ); /* length of path */
1694: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1695: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1696: if ( ss == NULL ) { /* no directory, so determine current directory */
1697: strcpy( name, path ); /* we got the fullname name because no directory */
1698: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1699: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1700: /* get current working directory */
1701: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1702: #ifdef WIN32
1703: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1704: #else
1705: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1706: #endif
1.126 brouard 1707: return( GLOCK_ERROR_GETCWD );
1708: }
1709: /* got dirc from getcwd*/
1710: printf(" DIRC = %s \n",dirc);
1.205 brouard 1711: } else { /* strip directory from path */
1.126 brouard 1712: ss++; /* after this, the filename */
1713: l2 = strlen( ss ); /* length of filename */
1714: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1715: strcpy( name, ss ); /* save file name */
1716: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1717: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1718: printf(" DIRC2 = %s \n",dirc);
1719: }
1720: /* We add a separator at the end of dirc if not exists */
1721: l1 = strlen( dirc ); /* length of directory */
1722: if( dirc[l1-1] != DIRSEPARATOR ){
1723: dirc[l1] = DIRSEPARATOR;
1724: dirc[l1+1] = 0;
1725: printf(" DIRC3 = %s \n",dirc);
1726: }
1727: ss = strrchr( name, '.' ); /* find last / */
1728: if (ss >0){
1729: ss++;
1730: strcpy(ext,ss); /* save extension */
1731: l1= strlen( name);
1732: l2= strlen(ss)+1;
1733: strncpy( finame, name, l1-l2);
1734: finame[l1-l2]= 0;
1735: }
1736:
1737: return( 0 ); /* we're done */
1738: }
1739:
1740:
1741: /******************************************/
1742:
1743: void replace_back_to_slash(char *s, char*t)
1744: {
1745: int i;
1746: int lg=0;
1747: i=0;
1748: lg=strlen(t);
1749: for(i=0; i<= lg; i++) {
1750: (s[i] = t[i]);
1751: if (t[i]== '\\') s[i]='/';
1752: }
1753: }
1754:
1.132 brouard 1755: char *trimbb(char *out, char *in)
1.137 brouard 1756: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1757: char *s;
1758: s=out;
1759: while (*in != '\0'){
1.137 brouard 1760: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1761: in++;
1762: }
1763: *out++ = *in++;
1764: }
1765: *out='\0';
1766: return s;
1767: }
1768:
1.187 brouard 1769: /* char *substrchaine(char *out, char *in, char *chain) */
1770: /* { */
1771: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1772: /* char *s, *t; */
1773: /* t=in;s=out; */
1774: /* while ((*in != *chain) && (*in != '\0')){ */
1775: /* *out++ = *in++; */
1776: /* } */
1777:
1778: /* /\* *in matches *chain *\/ */
1779: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1780: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1781: /* } */
1782: /* in--; chain--; */
1783: /* while ( (*in != '\0')){ */
1784: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1785: /* *out++ = *in++; */
1786: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1787: /* } */
1788: /* *out='\0'; */
1789: /* out=s; */
1790: /* return out; */
1791: /* } */
1792: char *substrchaine(char *out, char *in, char *chain)
1793: {
1794: /* Substract chain 'chain' from 'in', return and output 'out' */
1795: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1796:
1797: char *strloc;
1798:
1799: strcpy (out, in);
1800: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1801: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1802: if(strloc != NULL){
1803: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1804: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1805: /* strcpy (strloc, strloc +strlen(chain));*/
1806: }
1807: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1808: return out;
1809: }
1810:
1811:
1.145 brouard 1812: char *cutl(char *blocc, char *alocc, char *in, char occ)
1813: {
1.187 brouard 1814: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1815: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1816: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1817: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1818: */
1.160 brouard 1819: char *s, *t;
1.145 brouard 1820: t=in;s=in;
1821: while ((*in != occ) && (*in != '\0')){
1822: *alocc++ = *in++;
1823: }
1824: if( *in == occ){
1825: *(alocc)='\0';
1826: s=++in;
1827: }
1828:
1829: if (s == t) {/* occ not found */
1830: *(alocc-(in-s))='\0';
1831: in=s;
1832: }
1833: while ( *in != '\0'){
1834: *blocc++ = *in++;
1835: }
1836:
1837: *blocc='\0';
1838: return t;
1839: }
1.137 brouard 1840: char *cutv(char *blocc, char *alocc, char *in, char occ)
1841: {
1.187 brouard 1842: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1843: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1844: gives blocc="abcdef2ghi" and alocc="j".
1845: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1846: */
1847: char *s, *t;
1848: t=in;s=in;
1849: while (*in != '\0'){
1850: while( *in == occ){
1851: *blocc++ = *in++;
1852: s=in;
1853: }
1854: *blocc++ = *in++;
1855: }
1856: if (s == t) /* occ not found */
1857: *(blocc-(in-s))='\0';
1858: else
1859: *(blocc-(in-s)-1)='\0';
1860: in=s;
1861: while ( *in != '\0'){
1862: *alocc++ = *in++;
1863: }
1864:
1865: *alocc='\0';
1866: return s;
1867: }
1868:
1.126 brouard 1869: int nbocc(char *s, char occ)
1870: {
1871: int i,j=0;
1872: int lg=20;
1873: i=0;
1874: lg=strlen(s);
1875: for(i=0; i<= lg; i++) {
1.234 brouard 1876: if (s[i] == occ ) j++;
1.126 brouard 1877: }
1878: return j;
1879: }
1880:
1.137 brouard 1881: /* void cutv(char *u,char *v, char*t, char occ) */
1882: /* { */
1883: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1884: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1885: /* gives u="abcdef2ghi" and v="j" *\/ */
1886: /* int i,lg,j,p=0; */
1887: /* i=0; */
1888: /* lg=strlen(t); */
1889: /* for(j=0; j<=lg-1; j++) { */
1890: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1891: /* } */
1.126 brouard 1892:
1.137 brouard 1893: /* for(j=0; j<p; j++) { */
1894: /* (u[j] = t[j]); */
1895: /* } */
1896: /* u[p]='\0'; */
1.126 brouard 1897:
1.137 brouard 1898: /* for(j=0; j<= lg; j++) { */
1899: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1900: /* } */
1901: /* } */
1.126 brouard 1902:
1.160 brouard 1903: #ifdef _WIN32
1904: char * strsep(char **pp, const char *delim)
1905: {
1906: char *p, *q;
1907:
1908: if ((p = *pp) == NULL)
1909: return 0;
1910: if ((q = strpbrk (p, delim)) != NULL)
1911: {
1912: *pp = q + 1;
1913: *q = '\0';
1914: }
1915: else
1916: *pp = 0;
1917: return p;
1918: }
1919: #endif
1920:
1.126 brouard 1921: /********************** nrerror ********************/
1922:
1923: void nrerror(char error_text[])
1924: {
1925: fprintf(stderr,"ERREUR ...\n");
1926: fprintf(stderr,"%s\n",error_text);
1927: exit(EXIT_FAILURE);
1928: }
1929: /*********************** vector *******************/
1930: double *vector(int nl, int nh)
1931: {
1932: double *v;
1933: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1934: if (!v) nrerror("allocation failure in vector");
1935: return v-nl+NR_END;
1936: }
1937:
1938: /************************ free vector ******************/
1939: void free_vector(double*v, int nl, int nh)
1940: {
1941: free((FREE_ARG)(v+nl-NR_END));
1942: }
1943:
1944: /************************ivector *******************************/
1945: int *ivector(long nl,long nh)
1946: {
1947: int *v;
1948: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1949: if (!v) nrerror("allocation failure in ivector");
1950: return v-nl+NR_END;
1951: }
1952:
1953: /******************free ivector **************************/
1954: void free_ivector(int *v, long nl, long nh)
1955: {
1956: free((FREE_ARG)(v+nl-NR_END));
1957: }
1958:
1959: /************************lvector *******************************/
1960: long *lvector(long nl,long nh)
1961: {
1962: long *v;
1963: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1964: if (!v) nrerror("allocation failure in ivector");
1965: return v-nl+NR_END;
1966: }
1967:
1968: /******************free lvector **************************/
1969: void free_lvector(long *v, long nl, long nh)
1970: {
1971: free((FREE_ARG)(v+nl-NR_END));
1972: }
1973:
1974: /******************* imatrix *******************************/
1975: int **imatrix(long nrl, long nrh, long ncl, long nch)
1976: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1977: {
1978: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1979: int **m;
1980:
1981: /* allocate pointers to rows */
1982: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1983: if (!m) nrerror("allocation failure 1 in matrix()");
1984: m += NR_END;
1985: m -= nrl;
1986:
1987:
1988: /* allocate rows and set pointers to them */
1989: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1990: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1991: m[nrl] += NR_END;
1992: m[nrl] -= ncl;
1993:
1994: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1995:
1996: /* return pointer to array of pointers to rows */
1997: return m;
1998: }
1999:
2000: /****************** free_imatrix *************************/
2001: void free_imatrix(m,nrl,nrh,ncl,nch)
2002: int **m;
2003: long nch,ncl,nrh,nrl;
2004: /* free an int matrix allocated by imatrix() */
2005: {
2006: free((FREE_ARG) (m[nrl]+ncl-NR_END));
2007: free((FREE_ARG) (m+nrl-NR_END));
2008: }
2009:
2010: /******************* matrix *******************************/
2011: double **matrix(long nrl, long nrh, long ncl, long nch)
2012: {
2013: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
2014: double **m;
2015:
2016: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2017: if (!m) nrerror("allocation failure 1 in matrix()");
2018: m += NR_END;
2019: m -= nrl;
2020:
2021: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2022: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2023: m[nrl] += NR_END;
2024: m[nrl] -= ncl;
2025:
2026: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2027: return m;
1.145 brouard 2028: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
2029: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
2030: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 2031: */
2032: }
2033:
2034: /*************************free matrix ************************/
2035: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
2036: {
2037: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2038: free((FREE_ARG)(m+nrl-NR_END));
2039: }
2040:
2041: /******************* ma3x *******************************/
2042: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
2043: {
2044: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
2045: double ***m;
2046:
2047: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
2048: if (!m) nrerror("allocation failure 1 in matrix()");
2049: m += NR_END;
2050: m -= nrl;
2051:
2052: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
2053: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
2054: m[nrl] += NR_END;
2055: m[nrl] -= ncl;
2056:
2057: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
2058:
2059: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
2060: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
2061: m[nrl][ncl] += NR_END;
2062: m[nrl][ncl] -= nll;
2063: for (j=ncl+1; j<=nch; j++)
2064: m[nrl][j]=m[nrl][j-1]+nlay;
2065:
2066: for (i=nrl+1; i<=nrh; i++) {
2067: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
2068: for (j=ncl+1; j<=nch; j++)
2069: m[i][j]=m[i][j-1]+nlay;
2070: }
2071: return m;
2072: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
2073: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
2074: */
2075: }
2076:
2077: /*************************free ma3x ************************/
2078: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
2079: {
2080: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
2081: free((FREE_ARG)(m[nrl]+ncl-NR_END));
2082: free((FREE_ARG)(m+nrl-NR_END));
2083: }
2084:
2085: /*************** function subdirf ***********/
2086: char *subdirf(char fileres[])
2087: {
2088: /* Caution optionfilefiname is hidden */
2089: strcpy(tmpout,optionfilefiname);
2090: strcat(tmpout,"/"); /* Add to the right */
2091: strcat(tmpout,fileres);
2092: return tmpout;
2093: }
2094:
2095: /*************** function subdirf2 ***********/
2096: char *subdirf2(char fileres[], char *preop)
2097: {
1.314 brouard 2098: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2099: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2100: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2101: /* Caution optionfilefiname is hidden */
2102: strcpy(tmpout,optionfilefiname);
2103: strcat(tmpout,"/");
2104: strcat(tmpout,preop);
2105: strcat(tmpout,fileres);
2106: return tmpout;
2107: }
2108:
2109: /*************** function subdirf3 ***********/
2110: char *subdirf3(char fileres[], char *preop, char *preop2)
2111: {
2112:
2113: /* Caution optionfilefiname is hidden */
2114: strcpy(tmpout,optionfilefiname);
2115: strcat(tmpout,"/");
2116: strcat(tmpout,preop);
2117: strcat(tmpout,preop2);
2118: strcat(tmpout,fileres);
2119: return tmpout;
2120: }
1.213 brouard 2121:
2122: /*************** function subdirfext ***********/
2123: char *subdirfext(char fileres[], char *preop, char *postop)
2124: {
2125:
2126: strcpy(tmpout,preop);
2127: strcat(tmpout,fileres);
2128: strcat(tmpout,postop);
2129: return tmpout;
2130: }
1.126 brouard 2131:
1.213 brouard 2132: /*************** function subdirfext3 ***********/
2133: char *subdirfext3(char fileres[], char *preop, char *postop)
2134: {
2135:
2136: /* Caution optionfilefiname is hidden */
2137: strcpy(tmpout,optionfilefiname);
2138: strcat(tmpout,"/");
2139: strcat(tmpout,preop);
2140: strcat(tmpout,fileres);
2141: strcat(tmpout,postop);
2142: return tmpout;
2143: }
2144:
1.162 brouard 2145: char *asc_diff_time(long time_sec, char ascdiff[])
2146: {
2147: long sec_left, days, hours, minutes;
2148: days = (time_sec) / (60*60*24);
2149: sec_left = (time_sec) % (60*60*24);
2150: hours = (sec_left) / (60*60) ;
2151: sec_left = (sec_left) %(60*60);
2152: minutes = (sec_left) /60;
2153: sec_left = (sec_left) % (60);
2154: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2155: return ascdiff;
2156: }
2157:
1.126 brouard 2158: /***************** f1dim *************************/
2159: extern int ncom;
2160: extern double *pcom,*xicom;
2161: extern double (*nrfunc)(double []);
2162:
2163: double f1dim(double x)
2164: {
2165: int j;
2166: double f;
2167: double *xt;
2168:
2169: xt=vector(1,ncom);
2170: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2171: f=(*nrfunc)(xt);
2172: free_vector(xt,1,ncom);
2173: return f;
2174: }
2175:
2176: /*****************brent *************************/
2177: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2178: {
2179: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2180: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2181: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2182: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2183: * returned function value.
2184: */
1.126 brouard 2185: int iter;
2186: double a,b,d,etemp;
1.159 brouard 2187: double fu=0,fv,fw,fx;
1.164 brouard 2188: double ftemp=0.;
1.126 brouard 2189: double p,q,r,tol1,tol2,u,v,w,x,xm;
2190: double e=0.0;
2191:
2192: a=(ax < cx ? ax : cx);
2193: b=(ax > cx ? ax : cx);
2194: x=w=v=bx;
2195: fw=fv=fx=(*f)(x);
2196: for (iter=1;iter<=ITMAX;iter++) {
2197: xm=0.5*(a+b);
2198: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2199: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2200: printf(".");fflush(stdout);
2201: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2202: #ifdef DEBUGBRENT
1.126 brouard 2203: 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);
2204: 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);
2205: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2206: #endif
2207: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2208: *xmin=x;
2209: return fx;
2210: }
2211: ftemp=fu;
2212: if (fabs(e) > tol1) {
2213: r=(x-w)*(fx-fv);
2214: q=(x-v)*(fx-fw);
2215: p=(x-v)*q-(x-w)*r;
2216: q=2.0*(q-r);
2217: if (q > 0.0) p = -p;
2218: q=fabs(q);
2219: etemp=e;
2220: e=d;
2221: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2222: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2223: else {
1.224 brouard 2224: d=p/q;
2225: u=x+d;
2226: if (u-a < tol2 || b-u < tol2)
2227: d=SIGN(tol1,xm-x);
1.126 brouard 2228: }
2229: } else {
2230: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2231: }
2232: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2233: fu=(*f)(u);
2234: if (fu <= fx) {
2235: if (u >= x) a=x; else b=x;
2236: SHFT(v,w,x,u)
1.183 brouard 2237: SHFT(fv,fw,fx,fu)
2238: } else {
2239: if (u < x) a=u; else b=u;
2240: if (fu <= fw || w == x) {
1.224 brouard 2241: v=w;
2242: w=u;
2243: fv=fw;
2244: fw=fu;
1.183 brouard 2245: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2246: v=u;
2247: fv=fu;
1.183 brouard 2248: }
2249: }
1.126 brouard 2250: }
2251: nrerror("Too many iterations in brent");
2252: *xmin=x;
2253: return fx;
2254: }
2255:
2256: /****************** mnbrak ***********************/
2257:
2258: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2259: double (*func)(double))
1.183 brouard 2260: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2261: the downhill direction (defined by the function as evaluated at the initial points) and returns
2262: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2263: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2264: */
1.126 brouard 2265: double ulim,u,r,q, dum;
2266: double fu;
1.187 brouard 2267:
2268: double scale=10.;
2269: int iterscale=0;
2270:
2271: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2272: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2273:
2274:
2275: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2276: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2277: /* *bx = *ax - (*ax - *bx)/scale; */
2278: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2279: /* } */
2280:
1.126 brouard 2281: if (*fb > *fa) {
2282: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2283: SHFT(dum,*fb,*fa,dum)
2284: }
1.126 brouard 2285: *cx=(*bx)+GOLD*(*bx-*ax);
2286: *fc=(*func)(*cx);
1.183 brouard 2287: #ifdef DEBUG
1.224 brouard 2288: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2289: 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 2290: #endif
1.224 brouard 2291: 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 2292: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2293: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2294: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2295: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2296: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2297: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2298: fu=(*func)(u);
1.163 brouard 2299: #ifdef DEBUG
2300: /* f(x)=A(x-u)**2+f(u) */
2301: double A, fparabu;
2302: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2303: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2304: 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);
2305: 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 2306: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2307: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2308: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2309: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2310: #endif
1.184 brouard 2311: #ifdef MNBRAKORIGINAL
1.183 brouard 2312: #else
1.191 brouard 2313: /* if (fu > *fc) { */
2314: /* #ifdef DEBUG */
2315: /* printf("mnbrak4 fu > fc \n"); */
2316: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2317: /* #endif */
2318: /* /\* 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 *\\/ *\/ */
2319: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2320: /* dum=u; /\* Shifting c and u *\/ */
2321: /* u = *cx; */
2322: /* *cx = dum; */
2323: /* dum = fu; */
2324: /* fu = *fc; */
2325: /* *fc =dum; */
2326: /* } else { /\* end *\/ */
2327: /* #ifdef DEBUG */
2328: /* printf("mnbrak3 fu < fc \n"); */
2329: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2330: /* #endif */
2331: /* dum=u; /\* Shifting c and u *\/ */
2332: /* u = *cx; */
2333: /* *cx = dum; */
2334: /* dum = fu; */
2335: /* fu = *fc; */
2336: /* *fc =dum; */
2337: /* } */
1.224 brouard 2338: #ifdef DEBUGMNBRAK
2339: double A, fparabu;
2340: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2341: fparabu= *fa - A*(*ax-u)*(*ax-u);
2342: 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);
2343: 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 2344: #endif
1.191 brouard 2345: dum=u; /* Shifting c and u */
2346: u = *cx;
2347: *cx = dum;
2348: dum = fu;
2349: fu = *fc;
2350: *fc =dum;
1.183 brouard 2351: #endif
1.162 brouard 2352: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2353: #ifdef DEBUG
1.224 brouard 2354: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2355: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2356: #endif
1.126 brouard 2357: fu=(*func)(u);
2358: if (fu < *fc) {
1.183 brouard 2359: #ifdef DEBUG
1.224 brouard 2360: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2361: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2362: #endif
2363: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2364: SHFT(*fb,*fc,fu,(*func)(u))
2365: #ifdef DEBUG
2366: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2367: #endif
2368: }
1.162 brouard 2369: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2370: #ifdef DEBUG
1.224 brouard 2371: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2372: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2373: #endif
1.126 brouard 2374: u=ulim;
2375: fu=(*func)(u);
1.183 brouard 2376: } else { /* u could be left to b (if r > q parabola has a maximum) */
2377: #ifdef DEBUG
1.224 brouard 2378: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2379: 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 2380: #endif
1.126 brouard 2381: u=(*cx)+GOLD*(*cx-*bx);
2382: fu=(*func)(u);
1.224 brouard 2383: #ifdef DEBUG
2384: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2385: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2386: #endif
1.183 brouard 2387: } /* end tests */
1.126 brouard 2388: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2389: SHFT(*fa,*fb,*fc,fu)
2390: #ifdef DEBUG
1.224 brouard 2391: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2392: 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 2393: #endif
2394: } /* 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 2395: }
2396:
2397: /*************** linmin ************************/
1.162 brouard 2398: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2399: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2400: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2401: the value of func at the returned location p . This is actually all accomplished by calling the
2402: routines mnbrak and brent .*/
1.126 brouard 2403: int ncom;
2404: double *pcom,*xicom;
2405: double (*nrfunc)(double []);
2406:
1.224 brouard 2407: #ifdef LINMINORIGINAL
1.126 brouard 2408: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2409: #else
2410: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2411: #endif
1.126 brouard 2412: {
2413: double brent(double ax, double bx, double cx,
2414: double (*f)(double), double tol, double *xmin);
2415: double f1dim(double x);
2416: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2417: double *fc, double (*func)(double));
2418: int j;
2419: double xx,xmin,bx,ax;
2420: double fx,fb,fa;
1.187 brouard 2421:
1.203 brouard 2422: #ifdef LINMINORIGINAL
2423: #else
2424: double scale=10., axs, xxs; /* Scale added for infinity */
2425: #endif
2426:
1.126 brouard 2427: ncom=n;
2428: pcom=vector(1,n);
2429: xicom=vector(1,n);
2430: nrfunc=func;
2431: for (j=1;j<=n;j++) {
2432: pcom[j]=p[j];
1.202 brouard 2433: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2434: }
1.187 brouard 2435:
1.203 brouard 2436: #ifdef LINMINORIGINAL
2437: xx=1.;
2438: #else
2439: axs=0.0;
2440: xxs=1.;
2441: do{
2442: xx= xxs;
2443: #endif
1.187 brouard 2444: ax=0.;
2445: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2446: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2447: /* 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)) */
2448: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2449: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2450: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2451: /* 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 2452: #ifdef LINMINORIGINAL
2453: #else
2454: if (fx != fx){
1.224 brouard 2455: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2456: printf("|");
2457: fprintf(ficlog,"|");
1.203 brouard 2458: #ifdef DEBUGLINMIN
1.224 brouard 2459: 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 2460: #endif
2461: }
1.224 brouard 2462: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2463: #endif
2464:
1.191 brouard 2465: #ifdef DEBUGLINMIN
2466: 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 2467: 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 2468: #endif
1.224 brouard 2469: #ifdef LINMINORIGINAL
2470: #else
1.317 brouard 2471: if(fb == fx){ /* Flat function in the direction */
2472: xmin=xx;
1.224 brouard 2473: *flat=1;
1.317 brouard 2474: }else{
1.224 brouard 2475: *flat=0;
2476: #endif
2477: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2478: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2479: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2480: /* fmin = f(p[j] + xmin * xi[j]) */
2481: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2482: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2483: #ifdef DEBUG
1.224 brouard 2484: 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);
2485: 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);
2486: #endif
2487: #ifdef LINMINORIGINAL
2488: #else
2489: }
1.126 brouard 2490: #endif
1.191 brouard 2491: #ifdef DEBUGLINMIN
2492: printf("linmin end ");
1.202 brouard 2493: fprintf(ficlog,"linmin end ");
1.191 brouard 2494: #endif
1.126 brouard 2495: for (j=1;j<=n;j++) {
1.203 brouard 2496: #ifdef LINMINORIGINAL
2497: xi[j] *= xmin;
2498: #else
2499: #ifdef DEBUGLINMIN
2500: if(xxs <1.0)
2501: printf(" before xi[%d]=%12.8f", j,xi[j]);
2502: #endif
2503: 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) */
2504: #ifdef DEBUGLINMIN
2505: if(xxs <1.0)
2506: 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 );
2507: #endif
2508: #endif
1.187 brouard 2509: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2510: }
1.191 brouard 2511: #ifdef DEBUGLINMIN
1.203 brouard 2512: printf("\n");
1.191 brouard 2513: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2514: 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 2515: for (j=1;j<=n;j++) {
1.202 brouard 2516: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2517: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2518: if(j % ncovmodel == 0){
1.191 brouard 2519: printf("\n");
1.202 brouard 2520: fprintf(ficlog,"\n");
2521: }
1.191 brouard 2522: }
1.203 brouard 2523: #else
1.191 brouard 2524: #endif
1.126 brouard 2525: free_vector(xicom,1,n);
2526: free_vector(pcom,1,n);
2527: }
2528:
2529:
2530: /*************** powell ************************/
1.162 brouard 2531: /*
1.317 brouard 2532: Minimization of a function func of n variables. Input consists in an initial starting point
2533: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2534: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2535: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2536: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2537: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2538: */
1.224 brouard 2539: #ifdef LINMINORIGINAL
2540: #else
2541: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2542: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2543: #endif
1.126 brouard 2544: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2545: double (*func)(double []))
2546: {
1.224 brouard 2547: #ifdef LINMINORIGINAL
2548: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2549: double (*func)(double []));
1.224 brouard 2550: #else
1.241 brouard 2551: void linmin(double p[], double xi[], int n, double *fret,
2552: double (*func)(double []),int *flat);
1.224 brouard 2553: #endif
1.239 brouard 2554: int i,ibig,j,jk,k;
1.126 brouard 2555: double del,t,*pt,*ptt,*xit;
1.181 brouard 2556: double directest;
1.126 brouard 2557: double fp,fptt;
2558: double *xits;
2559: int niterf, itmp;
2560:
2561: pt=vector(1,n);
2562: ptt=vector(1,n);
2563: xit=vector(1,n);
2564: xits=vector(1,n);
2565: *fret=(*func)(p);
2566: for (j=1;j<=n;j++) pt[j]=p[j];
1.338 brouard 2567: rcurr_time = time(NULL);
2568: fp=(*fret); /* Initialisation */
1.126 brouard 2569: for (*iter=1;;++(*iter)) {
2570: ibig=0;
2571: del=0.0;
1.157 brouard 2572: rlast_time=rcurr_time;
2573: /* (void) gettimeofday(&curr_time,&tzp); */
2574: rcurr_time = time(NULL);
2575: curr_time = *localtime(&rcurr_time);
1.337 brouard 2576: /* 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); */
2577: /* 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); */
2578: 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);
2579: 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 2580: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2581: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2582: for (i=1;i<=n;i++) {
1.126 brouard 2583: fprintf(ficrespow," %.12lf", p[i]);
2584: }
1.239 brouard 2585: fprintf(ficrespow,"\n");fflush(ficrespow);
2586: printf("\n#model= 1 + age ");
2587: fprintf(ficlog,"\n#model= 1 + age ");
2588: if(nagesqr==1){
1.241 brouard 2589: printf(" + age*age ");
2590: fprintf(ficlog," + age*age ");
1.239 brouard 2591: }
2592: for(j=1;j <=ncovmodel-2;j++){
2593: if(Typevar[j]==0) {
2594: printf(" + V%d ",Tvar[j]);
2595: fprintf(ficlog," + V%d ",Tvar[j]);
2596: }else if(Typevar[j]==1) {
2597: printf(" + V%d*age ",Tvar[j]);
2598: fprintf(ficlog," + V%d*age ",Tvar[j]);
2599: }else if(Typevar[j]==2) {
2600: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2601: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2602: }
2603: }
1.126 brouard 2604: printf("\n");
1.239 brouard 2605: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2606: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2607: fprintf(ficlog,"\n");
1.239 brouard 2608: for(i=1,jk=1; i <=nlstate; i++){
2609: for(k=1; k <=(nlstate+ndeath); k++){
2610: if (k != i) {
2611: printf("%d%d ",i,k);
2612: fprintf(ficlog,"%d%d ",i,k);
2613: for(j=1; j <=ncovmodel; j++){
2614: printf("%12.7f ",p[jk]);
2615: fprintf(ficlog,"%12.7f ",p[jk]);
2616: jk++;
2617: }
2618: printf("\n");
2619: fprintf(ficlog,"\n");
2620: }
2621: }
2622: }
1.241 brouard 2623: if(*iter <=3 && *iter >1){
1.157 brouard 2624: tml = *localtime(&rcurr_time);
2625: strcpy(strcurr,asctime(&tml));
2626: rforecast_time=rcurr_time;
1.126 brouard 2627: itmp = strlen(strcurr);
2628: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2629: strcurr[itmp-1]='\0';
1.162 brouard 2630: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2631: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2632: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2633: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2634: forecast_time = *localtime(&rforecast_time);
2635: strcpy(strfor,asctime(&forecast_time));
2636: itmp = strlen(strfor);
2637: if(strfor[itmp-1]=='\n')
2638: strfor[itmp-1]='\0';
2639: 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);
2640: 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 2641: }
2642: }
1.187 brouard 2643: for (i=1;i<=n;i++) { /* For each direction i */
2644: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2645: fptt=(*fret);
2646: #ifdef DEBUG
1.203 brouard 2647: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2648: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2649: #endif
1.203 brouard 2650: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2651: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2652: #ifdef LINMINORIGINAL
1.188 brouard 2653: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2654: #else
2655: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2656: flatdir[i]=flat; /* Function is vanishing in that direction i */
2657: #endif
2658: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2659: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2660: /* because that direction will be replaced unless the gain del is small */
2661: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2662: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2663: /* with the new direction. */
2664: del=fabs(fptt-(*fret));
2665: ibig=i;
1.126 brouard 2666: }
2667: #ifdef DEBUG
2668: printf("%d %.12e",i,(*fret));
2669: fprintf(ficlog,"%d %.12e",i,(*fret));
2670: for (j=1;j<=n;j++) {
1.224 brouard 2671: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2672: printf(" x(%d)=%.12e",j,xit[j]);
2673: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2674: }
2675: for(j=1;j<=n;j++) {
1.225 brouard 2676: printf(" p(%d)=%.12e",j,p[j]);
2677: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2678: }
2679: printf("\n");
2680: fprintf(ficlog,"\n");
2681: #endif
1.187 brouard 2682: } /* end loop on each direction i */
2683: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2684: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2685: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2686: for(j=1;j<=n;j++) {
2687: if(flatdir[j] >0){
2688: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2689: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2690: }
1.319 brouard 2691: /* printf("\n"); */
2692: /* fprintf(ficlog,"\n"); */
2693: }
1.243 brouard 2694: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2695: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2696: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2697: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2698: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2699: /* decreased of more than 3.84 */
2700: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2701: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2702: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2703:
1.188 brouard 2704: /* Starting the program with initial values given by a former maximization will simply change */
2705: /* the scales of the directions and the directions, because the are reset to canonical directions */
2706: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2707: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2708: #ifdef DEBUG
2709: int k[2],l;
2710: k[0]=1;
2711: k[1]=-1;
2712: printf("Max: %.12e",(*func)(p));
2713: fprintf(ficlog,"Max: %.12e",(*func)(p));
2714: for (j=1;j<=n;j++) {
2715: printf(" %.12e",p[j]);
2716: fprintf(ficlog," %.12e",p[j]);
2717: }
2718: printf("\n");
2719: fprintf(ficlog,"\n");
2720: for(l=0;l<=1;l++) {
2721: for (j=1;j<=n;j++) {
2722: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2723: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2724: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2725: }
2726: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2727: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2728: }
2729: #endif
2730:
2731: free_vector(xit,1,n);
2732: free_vector(xits,1,n);
2733: free_vector(ptt,1,n);
2734: free_vector(pt,1,n);
2735: return;
1.192 brouard 2736: } /* enough precision */
1.240 brouard 2737: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2738: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2739: ptt[j]=2.0*p[j]-pt[j];
2740: xit[j]=p[j]-pt[j];
2741: pt[j]=p[j];
2742: }
1.181 brouard 2743: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2744: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2745: if (*iter <=4) {
1.225 brouard 2746: #else
2747: #endif
1.224 brouard 2748: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2749: #else
1.161 brouard 2750: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2751: #endif
1.162 brouard 2752: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2753: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2754: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2755: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2756: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2757: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2758: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2759: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2760: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2761: /* Even if f3 <f1, directest can be negative and t >0 */
2762: /* mu² and del² are equal when f3=f1 */
2763: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2764: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2765: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2766: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2767: #ifdef NRCORIGINAL
2768: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2769: #else
2770: 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 2771: t= t- del*SQR(fp-fptt);
1.183 brouard 2772: #endif
1.202 brouard 2773: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2774: #ifdef DEBUG
1.181 brouard 2775: 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);
2776: 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 2777: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2778: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2779: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2780: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2781: 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);
2782: 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);
2783: #endif
1.183 brouard 2784: #ifdef POWELLORIGINAL
2785: if (t < 0.0) { /* Then we use it for new direction */
2786: #else
1.182 brouard 2787: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2788: 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 2789: 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 2790: 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 2791: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2792: }
1.181 brouard 2793: if (directest < 0.0) { /* Then we use it for new direction */
2794: #endif
1.191 brouard 2795: #ifdef DEBUGLINMIN
1.234 brouard 2796: printf("Before linmin in direction P%d-P0\n",n);
2797: for (j=1;j<=n;j++) {
2798: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2799: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2800: if(j % ncovmodel == 0){
2801: printf("\n");
2802: fprintf(ficlog,"\n");
2803: }
2804: }
1.224 brouard 2805: #endif
2806: #ifdef LINMINORIGINAL
1.234 brouard 2807: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2808: #else
1.234 brouard 2809: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2810: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2811: #endif
1.234 brouard 2812:
1.191 brouard 2813: #ifdef DEBUGLINMIN
1.234 brouard 2814: for (j=1;j<=n;j++) {
2815: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2816: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2817: if(j % ncovmodel == 0){
2818: printf("\n");
2819: fprintf(ficlog,"\n");
2820: }
2821: }
1.224 brouard 2822: #endif
1.234 brouard 2823: for (j=1;j<=n;j++) {
2824: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2825: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2826: }
1.224 brouard 2827: #ifdef LINMINORIGINAL
2828: #else
1.234 brouard 2829: for (j=1, flatd=0;j<=n;j++) {
2830: if(flatdir[j]>0)
2831: flatd++;
2832: }
2833: if(flatd >0){
1.255 brouard 2834: printf("%d flat directions: ",flatd);
2835: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2836: for (j=1;j<=n;j++) {
2837: if(flatdir[j]>0){
2838: printf("%d ",j);
2839: fprintf(ficlog,"%d ",j);
2840: }
2841: }
2842: printf("\n");
2843: fprintf(ficlog,"\n");
1.319 brouard 2844: #ifdef FLATSUP
2845: free_vector(xit,1,n);
2846: free_vector(xits,1,n);
2847: free_vector(ptt,1,n);
2848: free_vector(pt,1,n);
2849: return;
2850: #endif
1.234 brouard 2851: }
1.191 brouard 2852: #endif
1.234 brouard 2853: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2854: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2855:
1.126 brouard 2856: #ifdef DEBUG
1.234 brouard 2857: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2858: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2859: for(j=1;j<=n;j++){
2860: printf(" %lf",xit[j]);
2861: fprintf(ficlog," %lf",xit[j]);
2862: }
2863: printf("\n");
2864: fprintf(ficlog,"\n");
1.126 brouard 2865: #endif
1.192 brouard 2866: } /* end of t or directest negative */
1.224 brouard 2867: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2868: #else
1.234 brouard 2869: } /* end if (fptt < fp) */
1.192 brouard 2870: #endif
1.225 brouard 2871: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2872: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2873: #else
1.224 brouard 2874: #endif
1.234 brouard 2875: } /* loop iteration */
1.126 brouard 2876: }
1.234 brouard 2877:
1.126 brouard 2878: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2879:
1.235 brouard 2880: 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 2881: {
1.338 brouard 2882: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij . Nicely done
1.279 brouard 2883: * (and selected quantitative values in nres)
2884: * by left multiplying the unit
2885: * matrix by transitions matrix until convergence is reached with precision ftolpl
2886: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2887: * Wx is row vector: population in state 1, population in state 2, population dead
2888: * or prevalence in state 1, prevalence in state 2, 0
2889: * newm is the matrix after multiplications, its rows are identical at a factor.
2890: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2891: * Output is prlim.
2892: * Initial matrix pimij
2893: */
1.206 brouard 2894: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2895: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2896: /* 0, 0 , 1} */
2897: /*
2898: * and after some iteration: */
2899: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2900: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2901: /* 0, 0 , 1} */
2902: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2903: /* {0.51571254859325999, 0.4842874514067399, */
2904: /* 0.51326036147820708, 0.48673963852179264} */
2905: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2906:
1.332 brouard 2907: int i, ii,j,k, k1;
1.209 brouard 2908: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2909: /* double **matprod2(); */ /* test */
1.218 brouard 2910: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2911: double **newm;
1.209 brouard 2912: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2913: int ncvloop=0;
1.288 brouard 2914: int first=0;
1.169 brouard 2915:
1.209 brouard 2916: min=vector(1,nlstate);
2917: max=vector(1,nlstate);
2918: meandiff=vector(1,nlstate);
2919:
1.218 brouard 2920: /* Starting with matrix unity */
1.126 brouard 2921: for (ii=1;ii<=nlstate+ndeath;ii++)
2922: for (j=1;j<=nlstate+ndeath;j++){
2923: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2924: }
1.169 brouard 2925:
2926: cov[1]=1.;
2927:
2928: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2929: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2930: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2931: ncvloop++;
1.126 brouard 2932: newm=savm;
2933: /* Covariates have to be included here again */
1.138 brouard 2934: cov[2]=agefin;
1.319 brouard 2935: if(nagesqr==1){
2936: cov[3]= agefin*agefin;
2937: }
1.332 brouard 2938: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
2939: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
2940: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
2941: if(Typevar[k1]==1){ /* A product with age */
2942: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
2943: }else{
2944: cov[2+nagesqr+k1]=precov[nres][k1];
2945: }
2946: }/* End of loop on model equation */
2947:
2948: /* Start of old code (replaced by a loop on position in the model equation */
2949: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
2950: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
2951: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
2952: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
2953: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
2954: /* * k 1 2 3 4 5 6 7 8 */
2955: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
2956: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
2957: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
2958: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
2959: /* *nsd=3 (1) (2) (3) */
2960: /* *TvarsD[nsd] [1]=2 1 3 */
2961: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
2962: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
2963: /* *Tage[] [1]=1 [2]=2 [3]=3 */
2964: /* *Tvard[] [1][1]=1 [2][1]=1 */
2965: /* * [1][2]=3 [2][2]=2 */
2966: /* *Tprod[](=k) [1]=1 [2]=8 */
2967: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
2968: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
2969: /* *TvarsDpType */
2970: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
2971: /* * nsd=1 (1) (2) */
2972: /* *TvarsD[nsd] 3 2 */
2973: /* *TnsdVar (3)=1 (2)=2 */
2974: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
2975: /* *Tage[] [1]=2 [2]= 3 */
2976: /* *\/ */
2977: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
2978: /* /\* 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)); *\/ */
2979: /* } */
2980: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
2981: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
2982: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
2983: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
2984: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
2985: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2986: /* /\* 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]); *\/ */
2987: /* } */
2988: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
2989: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
2990: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
2991: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
2992: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
2993: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
2994: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2995: /* } */
2996: /* /\* 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]); *\/ */
2997: /* } */
2998: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
2999: /* /\* 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]); *\/ */
3000: /* if(Dummy[Tvard[k][1]]==0){ */
3001: /* if(Dummy[Tvard[k][2]]==0){ */
3002: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3003: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3004: /* }else{ */
3005: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3006: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
3007: /* } */
3008: /* }else{ */
3009: /* if(Dummy[Tvard[k][2]]==0){ */
3010: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3011: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
3012: /* }else{ */
3013: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3014: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
3015: /* } */
3016: /* } */
3017: /* } /\* End product without age *\/ */
3018: /* ENd of old code */
1.138 brouard 3019: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3020: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3021: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 3022: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3023: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 3024: /* age and covariate values of ij are in 'cov' */
1.142 brouard 3025: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 3026:
1.126 brouard 3027: savm=oldm;
3028: oldm=newm;
1.209 brouard 3029:
3030: for(j=1; j<=nlstate; j++){
3031: max[j]=0.;
3032: min[j]=1.;
3033: }
3034: for(i=1;i<=nlstate;i++){
3035: sumnew=0;
3036: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
3037: for(j=1; j<=nlstate; j++){
3038: prlim[i][j]= newm[i][j]/(1-sumnew);
3039: max[j]=FMAX(max[j],prlim[i][j]);
3040: min[j]=FMIN(min[j],prlim[i][j]);
3041: }
3042: }
3043:
1.126 brouard 3044: maxmax=0.;
1.209 brouard 3045: for(j=1; j<=nlstate; j++){
3046: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
3047: maxmax=FMAX(maxmax,meandiff[j]);
3048: /* 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 3049: } /* j loop */
1.203 brouard 3050: *ncvyear= (int)age- (int)agefin;
1.208 brouard 3051: /* 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 3052: if(maxmax < ftolpl){
1.209 brouard 3053: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
3054: free_vector(min,1,nlstate);
3055: free_vector(max,1,nlstate);
3056: free_vector(meandiff,1,nlstate);
1.126 brouard 3057: return prlim;
3058: }
1.288 brouard 3059: } /* agefin loop */
1.208 brouard 3060: /* After some age loop it doesn't converge */
1.288 brouard 3061: if(!first){
3062: first=1;
3063: 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 3064: 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);
3065: }else if (first >=1 && first <10){
3066: 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);
3067: first++;
3068: }else if (first ==10){
3069: 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);
3070: 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");
3071: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
3072: first++;
1.288 brouard 3073: }
3074:
1.209 brouard 3075: /* 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); */
3076: free_vector(min,1,nlstate);
3077: free_vector(max,1,nlstate);
3078: free_vector(meandiff,1,nlstate);
1.208 brouard 3079:
1.169 brouard 3080: return prlim; /* should not reach here */
1.126 brouard 3081: }
3082:
1.217 brouard 3083:
3084: /**** Back Prevalence limit (stable or period prevalence) ****************/
3085:
1.218 brouard 3086: /* 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) */
3087: /* 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 3088: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 3089: {
1.264 brouard 3090: /* 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 3091: matrix by transitions matrix until convergence is reached with precision ftolpl */
3092: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
3093: /* Wx is row vector: population in state 1, population in state 2, population dead */
3094: /* or prevalence in state 1, prevalence in state 2, 0 */
3095: /* newm is the matrix after multiplications, its rows are identical at a factor */
3096: /* Initial matrix pimij */
3097: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
3098: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
3099: /* 0, 0 , 1} */
3100: /*
3101: * and after some iteration: */
3102: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
3103: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
3104: /* 0, 0 , 1} */
3105: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
3106: /* {0.51571254859325999, 0.4842874514067399, */
3107: /* 0.51326036147820708, 0.48673963852179264} */
3108: /* If we start from prlim again, prlim tends to a constant matrix */
3109:
1.332 brouard 3110: int i, ii,j,k, k1;
1.247 brouard 3111: int first=0;
1.217 brouard 3112: double *min, *max, *meandiff, maxmax,sumnew=0.;
3113: /* double **matprod2(); */ /* test */
3114: double **out, cov[NCOVMAX+1], **bmij();
3115: double **newm;
1.218 brouard 3116: double **dnewm, **doldm, **dsavm; /* for use */
3117: double **oldm, **savm; /* for use */
3118:
1.217 brouard 3119: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
3120: int ncvloop=0;
3121:
3122: min=vector(1,nlstate);
3123: max=vector(1,nlstate);
3124: meandiff=vector(1,nlstate);
3125:
1.266 brouard 3126: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
3127: oldm=oldms; savm=savms;
3128:
3129: /* Starting with matrix unity */
3130: for (ii=1;ii<=nlstate+ndeath;ii++)
3131: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 3132: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3133: }
3134:
3135: cov[1]=1.;
3136:
3137: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3138: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 3139: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 3140: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
3141: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 3142: ncvloop++;
1.218 brouard 3143: newm=savm; /* oldm should be kept from previous iteration or unity at start */
3144: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 3145: /* Covariates have to be included here again */
3146: cov[2]=agefin;
1.319 brouard 3147: if(nagesqr==1){
1.217 brouard 3148: cov[3]= agefin*agefin;;
1.319 brouard 3149: }
1.332 brouard 3150: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3151: if(Typevar[k1]==1){ /* A product with age */
3152: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 3153: }else{
1.332 brouard 3154: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 3155: }
1.332 brouard 3156: }/* End of loop on model equation */
3157:
3158: /* Old code */
3159:
3160: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
3161: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
3162: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
3163: /* /\* 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)); *\/ */
3164: /* } */
3165: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
3166: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3167: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3168: /* /\* /\\* 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])]); *\\/ *\/ */
3169: /* /\* } *\/ */
3170: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3171: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3172: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3173: /* /\* 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]); *\/ */
3174: /* } */
3175: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
3176: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
3177: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
3178: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3179: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3180: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
3181: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3182: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3183: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3184: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3185: /* } */
3186: /* /\* 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]); *\/ */
3187: /* } */
3188: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3189: /* /\* 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]); *\/ */
3190: /* if(Dummy[Tvard[k][1]]==0){ */
3191: /* if(Dummy[Tvard[k][2]]==0){ */
3192: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3193: /* }else{ */
3194: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3195: /* } */
3196: /* }else{ */
3197: /* if(Dummy[Tvard[k][2]]==0){ */
3198: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3199: /* }else{ */
3200: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3201: /* } */
3202: /* } */
3203: /* } */
1.217 brouard 3204:
3205: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3206: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3207: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3208: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3209: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3210: /* ij should be linked to the correct index of cov */
3211: /* age and covariate values ij are in 'cov', but we need to pass
3212: * ij for the observed prevalence at age and status and covariate
3213: * number: prevacurrent[(int)agefin][ii][ij]
3214: */
3215: /* 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 *\/ */
3216: /* 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 *\/ */
3217: 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 3218: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3219: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3220: /* for(i=1; i<=nlstate+ndeath; i++) { */
3221: /* printf("%d newm= ",i); */
3222: /* for(j=1;j<=nlstate+ndeath;j++) { */
3223: /* printf("%f ",newm[i][j]); */
3224: /* } */
3225: /* printf("oldm * "); */
3226: /* for(j=1;j<=nlstate+ndeath;j++) { */
3227: /* printf("%f ",oldm[i][j]); */
3228: /* } */
1.268 brouard 3229: /* printf(" bmmij "); */
1.266 brouard 3230: /* for(j=1;j<=nlstate+ndeath;j++) { */
3231: /* printf("%f ",pmmij[i][j]); */
3232: /* } */
3233: /* printf("\n"); */
3234: /* } */
3235: /* } */
1.217 brouard 3236: savm=oldm;
3237: oldm=newm;
1.266 brouard 3238:
1.217 brouard 3239: for(j=1; j<=nlstate; j++){
3240: max[j]=0.;
3241: min[j]=1.;
3242: }
3243: for(j=1; j<=nlstate; j++){
3244: for(i=1;i<=nlstate;i++){
1.234 brouard 3245: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3246: bprlim[i][j]= newm[i][j];
3247: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3248: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3249: }
3250: }
1.218 brouard 3251:
1.217 brouard 3252: maxmax=0.;
3253: for(i=1; i<=nlstate; i++){
1.318 brouard 3254: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3255: maxmax=FMAX(maxmax,meandiff[i]);
3256: /* 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 3257: } /* i loop */
1.217 brouard 3258: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3259: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3260: if(maxmax < ftolpl){
1.220 brouard 3261: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3262: free_vector(min,1,nlstate);
3263: free_vector(max,1,nlstate);
3264: free_vector(meandiff,1,nlstate);
3265: return bprlim;
3266: }
1.288 brouard 3267: } /* agefin loop */
1.217 brouard 3268: /* After some age loop it doesn't converge */
1.288 brouard 3269: if(!first){
1.247 brouard 3270: first=1;
3271: 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\
3272: 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);
3273: }
3274: 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 3275: 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);
3276: /* 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); */
3277: free_vector(min,1,nlstate);
3278: free_vector(max,1,nlstate);
3279: free_vector(meandiff,1,nlstate);
3280:
3281: return bprlim; /* should not reach here */
3282: }
3283:
1.126 brouard 3284: /*************** transition probabilities ***************/
3285:
3286: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3287: {
1.138 brouard 3288: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3289: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3290: model to the ncovmodel covariates (including constant and age).
3291: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3292: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3293: ncth covariate in the global vector x is given by the formula:
3294: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3295: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3296: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3297: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3298: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3299: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3300: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3301: */
3302: double s1, lnpijopii;
1.126 brouard 3303: /*double t34;*/
1.164 brouard 3304: int i,j, nc, ii, jj;
1.126 brouard 3305:
1.223 brouard 3306: for(i=1; i<= nlstate; i++){
3307: for(j=1; j<i;j++){
3308: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3309: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3310: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3311: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3312: }
3313: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3314: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3315: }
3316: for(j=i+1; j<=nlstate+ndeath;j++){
3317: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3318: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3319: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3320: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3321: }
3322: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3323: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3324: }
3325: }
1.218 brouard 3326:
1.223 brouard 3327: for(i=1; i<= nlstate; i++){
3328: s1=0;
3329: for(j=1; j<i; j++){
1.339 brouard 3330: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3331: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3332: }
3333: for(j=i+1; j<=nlstate+ndeath; j++){
1.339 brouard 3334: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf \n",i,j,ps[i][j],exp(ps[i][j])); */
1.223 brouard 3335: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3336: }
3337: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3338: ps[i][i]=1./(s1+1.);
3339: /* Computing other pijs */
3340: for(j=1; j<i; j++)
1.325 brouard 3341: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3342: for(j=i+1; j<=nlstate+ndeath; j++)
3343: ps[i][j]= exp(ps[i][j])*ps[i][i];
3344: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3345: } /* end i */
1.218 brouard 3346:
1.223 brouard 3347: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3348: for(jj=1; jj<= nlstate+ndeath; jj++){
3349: ps[ii][jj]=0;
3350: ps[ii][ii]=1;
3351: }
3352: }
1.294 brouard 3353:
3354:
1.223 brouard 3355: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3356: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3357: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3358: /* } */
3359: /* printf("\n "); */
3360: /* } */
3361: /* printf("\n ");printf("%lf ",cov[2]);*/
3362: /*
3363: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3364: goto end;*/
1.266 brouard 3365: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3366: }
3367:
1.218 brouard 3368: /*************** backward transition probabilities ***************/
3369:
3370: /* 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 ) */
3371: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3372: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3373: {
1.302 brouard 3374: /* 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 3375: * 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 3376: */
1.218 brouard 3377: int i, ii, j,k;
1.222 brouard 3378:
3379: double **out, **pmij();
3380: double sumnew=0.;
1.218 brouard 3381: double agefin;
1.292 brouard 3382: 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 3383: double **dnewm, **dsavm, **doldm;
3384: double **bbmij;
3385:
1.218 brouard 3386: doldm=ddoldms; /* global pointers */
1.222 brouard 3387: dnewm=ddnewms;
3388: dsavm=ddsavms;
1.318 brouard 3389:
3390: /* Debug */
3391: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3392: agefin=cov[2];
1.268 brouard 3393: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3394: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3395: the observed prevalence (with this covariate ij) at beginning of transition */
3396: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3397:
3398: /* P_x */
1.325 brouard 3399: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3400: /* outputs pmmij which is a stochastic matrix in row */
3401:
3402: /* Diag(w_x) */
1.292 brouard 3403: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3404: sumnew=0.;
1.269 brouard 3405: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3406: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3407: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3408: sumnew+=prevacurrent[(int)agefin][ii][ij];
3409: }
3410: if(sumnew >0.01){ /* At least some value in the prevalence */
3411: for (ii=1;ii<=nlstate+ndeath;ii++){
3412: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3413: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3414: }
3415: }else{
3416: for (ii=1;ii<=nlstate+ndeath;ii++){
3417: for (j=1;j<=nlstate+ndeath;j++)
3418: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3419: }
3420: /* if(sumnew <0.9){ */
3421: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3422: /* } */
3423: }
3424: k3=0.0; /* We put the last diagonal to 0 */
3425: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3426: doldm[ii][ii]= k3;
3427: }
3428: /* End doldm, At the end doldm is diag[(w_i)] */
3429:
1.292 brouard 3430: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3431: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3432:
1.292 brouard 3433: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3434: /* 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 3435: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3436: sumnew=0.;
1.222 brouard 3437: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3438: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3439: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3440: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3441: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3442: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3443: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3444: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3445: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3446: /* }else */
1.268 brouard 3447: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3448: } /*End ii */
3449: } /* 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 */
3450:
1.292 brouard 3451: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3452: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3453: /* end bmij */
1.266 brouard 3454: return ps; /*pointer is unchanged */
1.218 brouard 3455: }
1.217 brouard 3456: /*************** transition probabilities ***************/
3457:
1.218 brouard 3458: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3459: {
3460: /* According to parameters values stored in x and the covariate's values stored in cov,
3461: computes the probability to be observed in state j being in state i by appying the
3462: model to the ncovmodel covariates (including constant and age).
3463: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3464: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3465: ncth covariate in the global vector x is given by the formula:
3466: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3467: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3468: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3469: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3470: Outputs ps[i][j] the probability to be observed in j being in j according to
3471: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3472: */
3473: double s1, lnpijopii;
3474: /*double t34;*/
3475: int i,j, nc, ii, jj;
3476:
1.234 brouard 3477: for(i=1; i<= nlstate; i++){
3478: for(j=1; j<i;j++){
3479: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3480: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3481: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3482: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3483: }
3484: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3485: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3486: }
3487: for(j=i+1; j<=nlstate+ndeath;j++){
3488: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3489: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3490: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3491: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3492: }
3493: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3494: }
3495: }
3496:
3497: for(i=1; i<= nlstate; i++){
3498: s1=0;
3499: for(j=1; j<i; j++){
3500: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3501: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3502: }
3503: for(j=i+1; j<=nlstate+ndeath; j++){
3504: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3505: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3506: }
3507: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3508: ps[i][i]=1./(s1+1.);
3509: /* Computing other pijs */
3510: for(j=1; j<i; j++)
3511: ps[i][j]= exp(ps[i][j])*ps[i][i];
3512: for(j=i+1; j<=nlstate+ndeath; j++)
3513: ps[i][j]= exp(ps[i][j])*ps[i][i];
3514: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3515: } /* end i */
3516:
3517: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3518: for(jj=1; jj<= nlstate+ndeath; jj++){
3519: ps[ii][jj]=0;
3520: ps[ii][ii]=1;
3521: }
3522: }
1.296 brouard 3523: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3524: for(jj=1; jj<= nlstate+ndeath; jj++){
3525: s1=0.;
3526: for(ii=1; ii<= nlstate+ndeath; ii++){
3527: s1+=ps[ii][jj];
3528: }
3529: for(ii=1; ii<= nlstate; ii++){
3530: ps[ii][jj]=ps[ii][jj]/s1;
3531: }
3532: }
3533: /* Transposition */
3534: for(jj=1; jj<= nlstate+ndeath; jj++){
3535: for(ii=jj; ii<= nlstate+ndeath; ii++){
3536: s1=ps[ii][jj];
3537: ps[ii][jj]=ps[jj][ii];
3538: ps[jj][ii]=s1;
3539: }
3540: }
3541: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3542: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3543: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3544: /* } */
3545: /* printf("\n "); */
3546: /* } */
3547: /* printf("\n ");printf("%lf ",cov[2]);*/
3548: /*
3549: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3550: goto end;*/
3551: return ps;
1.217 brouard 3552: }
3553:
3554:
1.126 brouard 3555: /**************** Product of 2 matrices ******************/
3556:
1.145 brouard 3557: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3558: {
3559: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3560: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3561: /* in, b, out are matrice of pointers which should have been initialized
3562: before: only the contents of out is modified. The function returns
3563: a pointer to pointers identical to out */
1.145 brouard 3564: int i, j, k;
1.126 brouard 3565: for(i=nrl; i<= nrh; i++)
1.145 brouard 3566: for(k=ncolol; k<=ncoloh; k++){
3567: out[i][k]=0.;
3568: for(j=ncl; j<=nch; j++)
3569: out[i][k] +=in[i][j]*b[j][k];
3570: }
1.126 brouard 3571: return out;
3572: }
3573:
3574:
3575: /************* Higher Matrix Product ***************/
3576:
1.235 brouard 3577: 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 3578: {
1.336 brouard 3579: /* Already optimized with precov.
3580: 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 3581: 'nhstepm*hstepm*stepm' months (i.e. until
3582: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3583: nhstepm*hstepm matrices.
3584: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3585: (typically every 2 years instead of every month which is too big
3586: for the memory).
3587: Model is determined by parameters x and covariates have to be
3588: included manually here.
3589:
3590: */
3591:
1.330 brouard 3592: int i, j, d, h, k, k1;
1.131 brouard 3593: double **out, cov[NCOVMAX+1];
1.126 brouard 3594: double **newm;
1.187 brouard 3595: double agexact;
1.214 brouard 3596: double agebegin, ageend;
1.126 brouard 3597:
3598: /* Hstepm could be zero and should return the unit matrix */
3599: for (i=1;i<=nlstate+ndeath;i++)
3600: for (j=1;j<=nlstate+ndeath;j++){
3601: oldm[i][j]=(i==j ? 1.0 : 0.0);
3602: po[i][j][0]=(i==j ? 1.0 : 0.0);
3603: }
3604: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3605: for(h=1; h <=nhstepm; h++){
3606: for(d=1; d <=hstepm; d++){
3607: newm=savm;
3608: /* Covariates have to be included here again */
3609: cov[1]=1.;
1.214 brouard 3610: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3611: cov[2]=agexact;
1.319 brouard 3612: if(nagesqr==1){
1.227 brouard 3613: cov[3]= agexact*agexact;
1.319 brouard 3614: }
1.330 brouard 3615: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
3616: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
3617: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.332 brouard 3618: if(Typevar[k1]==1){ /* A product with age */
3619: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
3620: }else{
3621: cov[2+nagesqr+k1]=precov[nres][k1];
3622: }
3623: }/* End of loop on model equation */
3624: /* Old code */
3625: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
3626: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
3627: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
3628: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
3629: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
3630: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3631: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3632: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
3633: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
3634: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
3635: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
3636: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
3637: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3638: /* /\* 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]])); *\/ */
3639: /* 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); */
3640: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3641: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
3642: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
3643: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
3644: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
3645: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
3646: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3647: /* 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]]); */
3648: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3649: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
3650: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
3651: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
3652: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
3653: /* 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]); */
3654: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3655:
3656: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
3657: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
3658: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
3659: /* /\* *\/ */
1.330 brouard 3660: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3661: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3662: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 3663: /* /\*cptcovage=2 1 2 *\/ */
3664: /* /\*Tage[k]= 5 8 *\/ */
3665: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
3666: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3667: /* 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]]); */
3668: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3669: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
3670: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
3671: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
3672: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
3673: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
3674: /* /\* 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); *\/ */
3675: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
3676: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
3677: /* /\* } *\/ */
3678: /* /\* 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]); *\/ */
3679: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
3680: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
3681: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
3682: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
3683: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
3684: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
3685: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
3686: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
3687: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 3688:
1.332 brouard 3689: /* /\* 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])]); *\/ */
3690: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3691: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
3692: /* 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]]); */
3693: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3694:
3695: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
3696: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
3697: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3698: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
3699: /* /\* 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]])]; *\/ */
3700: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
3701: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
3702: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
3703: /* /\* } *\/ */
3704: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
3705: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
3706: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
3707: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3708: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
3709: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
3710: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3711: /* /\* } *\/ */
3712: /* /\* }/\\*end of products quantitative *\\/ *\/ */
3713: /* }/\*end of products *\/ */
3714: /* } /\* End of loop on model equation *\/ */
1.235 brouard 3715: /* for (k=1; k<=cptcovn;k++) */
3716: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3717: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3718: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3719: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3720: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3721:
3722:
1.126 brouard 3723: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3724: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3725: /* right multiplication of oldm by the current matrix */
1.126 brouard 3726: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3727: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3728: /* if((int)age == 70){ */
3729: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3730: /* for(i=1; i<=nlstate+ndeath; i++) { */
3731: /* printf("%d pmmij ",i); */
3732: /* for(j=1;j<=nlstate+ndeath;j++) { */
3733: /* printf("%f ",pmmij[i][j]); */
3734: /* } */
3735: /* printf(" oldm "); */
3736: /* for(j=1;j<=nlstate+ndeath;j++) { */
3737: /* printf("%f ",oldm[i][j]); */
3738: /* } */
3739: /* printf("\n"); */
3740: /* } */
3741: /* } */
1.126 brouard 3742: savm=oldm;
3743: oldm=newm;
3744: }
3745: for(i=1; i<=nlstate+ndeath; i++)
3746: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3747: po[i][j][h]=newm[i][j];
3748: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3749: }
1.128 brouard 3750: /*printf("h=%d ",h);*/
1.126 brouard 3751: } /* end h */
1.267 brouard 3752: /* printf("\n H=%d \n",h); */
1.126 brouard 3753: return po;
3754: }
3755:
1.217 brouard 3756: /************* Higher Back Matrix Product ***************/
1.218 brouard 3757: /* 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 3758: 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 3759: {
1.332 brouard 3760: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
3761: computes the transition matrix starting at age 'age' over
1.217 brouard 3762: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3763: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3764: nhstepm*hstepm matrices.
3765: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3766: (typically every 2 years instead of every month which is too big
1.217 brouard 3767: for the memory).
1.218 brouard 3768: Model is determined by parameters x and covariates have to be
1.266 brouard 3769: included manually here. Then we use a call to bmij(x and cov)
3770: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3771: */
1.217 brouard 3772:
1.332 brouard 3773: int i, j, d, h, k, k1;
1.266 brouard 3774: double **out, cov[NCOVMAX+1], **bmij();
3775: double **newm, ***newmm;
1.217 brouard 3776: double agexact;
3777: double agebegin, ageend;
1.222 brouard 3778: double **oldm, **savm;
1.217 brouard 3779:
1.266 brouard 3780: newmm=po; /* To be saved */
3781: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3782: /* Hstepm could be zero and should return the unit matrix */
3783: for (i=1;i<=nlstate+ndeath;i++)
3784: for (j=1;j<=nlstate+ndeath;j++){
3785: oldm[i][j]=(i==j ? 1.0 : 0.0);
3786: po[i][j][0]=(i==j ? 1.0 : 0.0);
3787: }
3788: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3789: for(h=1; h <=nhstepm; h++){
3790: for(d=1; d <=hstepm; d++){
3791: newm=savm;
3792: /* Covariates have to be included here again */
3793: cov[1]=1.;
1.271 brouard 3794: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3795: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3796: /* Debug */
3797: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3798: cov[2]=agexact;
1.332 brouard 3799: if(nagesqr==1){
1.222 brouard 3800: cov[3]= agexact*agexact;
1.332 brouard 3801: }
3802: /** New code */
3803: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3804: if(Typevar[k1]==1){ /* A product with age */
3805: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 3806: }else{
1.332 brouard 3807: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 3808: }
1.332 brouard 3809: }/* End of loop on model equation */
3810: /** End of new code */
3811: /** This was old code */
3812: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
3813: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3814: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3815: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
3816: /* /\* 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)); *\/ */
3817: /* } */
3818: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3819: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3820: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3821: /* /\* 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]); *\/ */
3822: /* } */
3823: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
3824: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
3825: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3826: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3827: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3828: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3829: /* } */
3830: /* /\* 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]); *\/ */
3831: /* } */
3832: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
3833: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3834: /* if(Dummy[Tvard[k][1]]==0){ */
3835: /* if(Dummy[Tvard[k][2]]==0){ */
3836: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
3837: /* }else{ */
3838: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3839: /* } */
3840: /* }else{ */
3841: /* if(Dummy[Tvard[k][2]]==0){ */
3842: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3843: /* }else{ */
3844: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3845: /* } */
3846: /* } */
3847: /* } */
3848: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
3849: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
3850: /** End of old code */
3851:
1.218 brouard 3852: /* Careful transposed matrix */
1.266 brouard 3853: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3854: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3855: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3856: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3857: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3858: /* if((int)age == 70){ */
3859: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3860: /* for(i=1; i<=nlstate+ndeath; i++) { */
3861: /* printf("%d pmmij ",i); */
3862: /* for(j=1;j<=nlstate+ndeath;j++) { */
3863: /* printf("%f ",pmmij[i][j]); */
3864: /* } */
3865: /* printf(" oldm "); */
3866: /* for(j=1;j<=nlstate+ndeath;j++) { */
3867: /* printf("%f ",oldm[i][j]); */
3868: /* } */
3869: /* printf("\n"); */
3870: /* } */
3871: /* } */
3872: savm=oldm;
3873: oldm=newm;
3874: }
3875: for(i=1; i<=nlstate+ndeath; i++)
3876: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3877: po[i][j][h]=newm[i][j];
1.268 brouard 3878: /* if(h==nhstepm) */
3879: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3880: }
1.268 brouard 3881: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3882: } /* end h */
1.268 brouard 3883: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3884: return po;
3885: }
3886:
3887:
1.162 brouard 3888: #ifdef NLOPT
3889: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3890: double fret;
3891: double *xt;
3892: int j;
3893: myfunc_data *d2 = (myfunc_data *) pd;
3894: /* xt = (p1-1); */
3895: xt=vector(1,n);
3896: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3897:
3898: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3899: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3900: printf("Function = %.12lf ",fret);
3901: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3902: printf("\n");
3903: free_vector(xt,1,n);
3904: return fret;
3905: }
3906: #endif
1.126 brouard 3907:
3908: /*************** log-likelihood *************/
3909: double func( double *x)
3910: {
1.336 brouard 3911: int i, ii, j, k, mi, d, kk, kf=0;
1.226 brouard 3912: int ioffset=0;
1.339 brouard 3913: int ipos=0,iposold=0,ncovv=0;
3914:
1.340 brouard 3915: double cotvarv, cotvarvold;
1.226 brouard 3916: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3917: double **out;
3918: double lli; /* Individual log likelihood */
3919: int s1, s2;
1.228 brouard 3920: 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 3921:
1.226 brouard 3922: double bbh, survp;
3923: double agexact;
1.336 brouard 3924: double agebegin, ageend;
1.226 brouard 3925: /*extern weight */
3926: /* We are differentiating ll according to initial status */
3927: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3928: /*for(i=1;i<imx;i++)
3929: printf(" %d\n",s[4][i]);
3930: */
1.162 brouard 3931:
1.226 brouard 3932: ++countcallfunc;
1.162 brouard 3933:
1.226 brouard 3934: cov[1]=1.;
1.126 brouard 3935:
1.226 brouard 3936: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3937: ioffset=0;
1.226 brouard 3938: if(mle==1){
3939: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3940: /* Computes the values of the ncovmodel covariates of the model
3941: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3942: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3943: to be observed in j being in i according to the model.
3944: */
1.243 brouard 3945: ioffset=2+nagesqr ;
1.233 brouard 3946: /* Fixed */
1.336 brouard 3947: for (kf=1; kf<=ncovf;kf++){ /* For each fixed covariate dummu or quant or prod */
1.319 brouard 3948: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3949: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3950: /* 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 3951: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.336 brouard 3952: 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 3953: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3954: }
1.226 brouard 3955: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3956: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3957: has been calculated etc */
3958: /* For an individual i, wav[i] gives the number of effective waves */
3959: /* We compute the contribution to Likelihood of each effective transition
3960: mw[mi][i] is real wave of the mi th effectve wave */
3961: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3962: s2=s[mw[mi+1][i]][i];
1.341 brouard 3963: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i] because now is moved after nvocol+nqv
1.226 brouard 3964: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3965: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3966: */
1.336 brouard 3967: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
3968: /* Wave varying (but not age varying) */
1.339 brouard 3969: /* 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*\/ */
3970: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? *\/ */
3971: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
3972: /* } */
1.340 brouard 3973: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age )*/
3974: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
3975: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
3976: if(TvarFind[itv]==0){ /* Not a fixed covariate */
1.341 brouard 3977: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* cotvar[wav][ncovcol+nqv+iv][i] */
1.340 brouard 3978: }else{ /* fixed covariate */
3979: cotvarv=covar[Tvar[TvarFind[itv]]][i];
3980: }
1.339 brouard 3981: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 3982: cotvarvold=cotvarv;
3983: }else{ /* A second product */
3984: cotvarv=cotvarv*cotvarvold;
1.339 brouard 3985: }
3986: iposold=ipos;
1.340 brouard 3987: cov[ioffset+ipos]=cotvarv;
1.234 brouard 3988: }
1.339 brouard 3989: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
3990: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3991: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3992: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3993: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3994: /* 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]); */
3995: /* } */
3996: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
3997: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3998: /* /\* 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]); *\/ */
3999: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
4000: /* } */
4001: /* for products of time varying to be done */
1.234 brouard 4002: for (ii=1;ii<=nlstate+ndeath;ii++)
4003: for (j=1;j<=nlstate+ndeath;j++){
4004: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4005: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4006: }
1.336 brouard 4007:
4008: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4009: 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 4010: for(d=0; d<dh[mi][i]; d++){
4011: newm=savm;
4012: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4013: cov[2]=agexact;
4014: if(nagesqr==1)
4015: cov[3]= agexact*agexact; /* Should be changed here */
4016: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 4017: if(!FixedV[Tvar[Tage[kk]]])
4018: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4019: else
1.341 brouard 4020: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.234 brouard 4021: }
4022: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4023: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4024: savm=oldm;
4025: oldm=newm;
4026: } /* end mult */
4027:
4028: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4029: /* But now since version 0.9 we anticipate for bias at large stepm.
4030: * If stepm is larger than one month (smallest stepm) and if the exact delay
4031: * (in months) between two waves is not a multiple of stepm, we rounded to
4032: * the nearest (and in case of equal distance, to the lowest) interval but now
4033: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4034: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4035: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 4036: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4037: * -stepm/2 to stepm/2 .
4038: * For stepm=1 the results are the same as for previous versions of Imach.
4039: * For stepm > 1 the results are less biased than in previous versions.
4040: */
1.234 brouard 4041: s1=s[mw[mi][i]][i];
4042: s2=s[mw[mi+1][i]][i];
4043: bbh=(double)bh[mi][i]/(double)stepm;
4044: /* bias bh is positive if real duration
4045: * is higher than the multiple of stepm and negative otherwise.
4046: */
4047: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
4048: if( s2 > nlstate){
4049: /* i.e. if s2 is a death state and if the date of death is known
4050: then the contribution to the likelihood is the probability to
4051: die between last step unit time and current step unit time,
4052: which is also equal to probability to die before dh
4053: minus probability to die before dh-stepm .
4054: In version up to 0.92 likelihood was computed
4055: as if date of death was unknown. Death was treated as any other
4056: health state: the date of the interview describes the actual state
4057: and not the date of a change in health state. The former idea was
4058: to consider that at each interview the state was recorded
4059: (healthy, disable or death) and IMaCh was corrected; but when we
4060: introduced the exact date of death then we should have modified
4061: the contribution of an exact death to the likelihood. This new
4062: contribution is smaller and very dependent of the step unit
4063: stepm. It is no more the probability to die between last interview
4064: and month of death but the probability to survive from last
4065: interview up to one month before death multiplied by the
4066: probability to die within a month. Thanks to Chris
4067: Jackson for correcting this bug. Former versions increased
4068: mortality artificially. The bad side is that we add another loop
4069: which slows down the processing. The difference can be up to 10%
4070: lower mortality.
4071: */
4072: /* If, at the beginning of the maximization mostly, the
4073: cumulative probability or probability to be dead is
4074: constant (ie = 1) over time d, the difference is equal to
4075: 0. out[s1][3] = savm[s1][3]: probability, being at state
4076: s1 at precedent wave, to be dead a month before current
4077: wave is equal to probability, being at state s1 at
4078: precedent wave, to be dead at mont of the current
4079: wave. Then the observed probability (that this person died)
4080: is null according to current estimated parameter. In fact,
4081: it should be very low but not zero otherwise the log go to
4082: infinity.
4083: */
1.183 brouard 4084: /* #ifdef INFINITYORIGINAL */
4085: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4086: /* #else */
4087: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
4088: /* lli=log(mytinydouble); */
4089: /* else */
4090: /* lli=log(out[s1][s2] - savm[s1][s2]); */
4091: /* #endif */
1.226 brouard 4092: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4093:
1.226 brouard 4094: } else if ( s2==-1 ) { /* alive */
4095: for (j=1,survp=0. ; j<=nlstate; j++)
4096: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4097: /*survp += out[s1][j]; */
4098: lli= log(survp);
4099: }
1.336 brouard 4100: /* else if (s2==-4) { */
4101: /* for (j=3,survp=0. ; j<=nlstate; j++) */
4102: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4103: /* lli= log(survp); */
4104: /* } */
4105: /* else if (s2==-5) { */
4106: /* for (j=1,survp=0. ; j<=2; j++) */
4107: /* survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j]; */
4108: /* lli= log(survp); */
4109: /* } */
1.226 brouard 4110: else{
4111: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4112: /* 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 */
4113: }
4114: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
4115: /*if(lli ==000.0)*/
1.340 brouard 4116: /* printf("num[i], i=%d, bbh= %f lli=%f savm=%f out=%f %d\n",bbh,lli,savm[s1][s2], out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]],i); */
1.226 brouard 4117: ipmx +=1;
4118: sw += weight[i];
4119: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4120: /* if (lli < log(mytinydouble)){ */
4121: /* 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); */
4122: /* 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]); */
4123: /* } */
4124: } /* end of wave */
4125: } /* end of individual */
4126: } else if(mle==2){
4127: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 4128: ioffset=2+nagesqr ;
4129: for (k=1; k<=ncovf;k++)
4130: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 4131: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 4132: for(k=1; k <= ncovv ; k++){
1.341 brouard 4133: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.319 brouard 4134: }
1.226 brouard 4135: for (ii=1;ii<=nlstate+ndeath;ii++)
4136: for (j=1;j<=nlstate+ndeath;j++){
4137: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4138: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4139: }
4140: for(d=0; d<=dh[mi][i]; d++){
4141: newm=savm;
4142: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4143: cov[2]=agexact;
4144: if(nagesqr==1)
4145: cov[3]= agexact*agexact;
4146: for (kk=1; kk<=cptcovage;kk++) {
4147: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4148: }
4149: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4150: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4151: savm=oldm;
4152: oldm=newm;
4153: } /* end mult */
4154:
4155: s1=s[mw[mi][i]][i];
4156: s2=s[mw[mi+1][i]][i];
4157: bbh=(double)bh[mi][i]/(double)stepm;
4158: 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 */
4159: ipmx +=1;
4160: sw += weight[i];
4161: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4162: } /* end of wave */
4163: } /* end of individual */
4164: } else if(mle==3){ /* exponential inter-extrapolation */
4165: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4166: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4167: for(mi=1; mi<= wav[i]-1; mi++){
4168: for (ii=1;ii<=nlstate+ndeath;ii++)
4169: for (j=1;j<=nlstate+ndeath;j++){
4170: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4171: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4172: }
4173: for(d=0; d<dh[mi][i]; d++){
4174: newm=savm;
4175: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4176: cov[2]=agexact;
4177: if(nagesqr==1)
4178: cov[3]= agexact*agexact;
4179: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4180: if(!FixedV[Tvar[Tage[kk]]])
4181: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4182: else
1.341 brouard 4183: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.226 brouard 4184: }
4185: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4186: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4187: savm=oldm;
4188: oldm=newm;
4189: } /* end mult */
4190:
4191: s1=s[mw[mi][i]][i];
4192: s2=s[mw[mi+1][i]][i];
4193: bbh=(double)bh[mi][i]/(double)stepm;
4194: 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 */
4195: ipmx +=1;
4196: sw += weight[i];
4197: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4198: } /* end of wave */
4199: } /* end of individual */
4200: }else if (mle==4){ /* ml=4 no inter-extrapolation */
4201: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4202: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4203: for(mi=1; mi<= wav[i]-1; mi++){
4204: for (ii=1;ii<=nlstate+ndeath;ii++)
4205: for (j=1;j<=nlstate+ndeath;j++){
4206: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4207: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4208: }
4209: for(d=0; d<dh[mi][i]; d++){
4210: newm=savm;
4211: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4212: cov[2]=agexact;
4213: if(nagesqr==1)
4214: cov[3]= agexact*agexact;
4215: for (kk=1; kk<=cptcovage;kk++) {
4216: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4217: }
1.126 brouard 4218:
1.226 brouard 4219: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4220: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4221: savm=oldm;
4222: oldm=newm;
4223: } /* end mult */
4224:
4225: s1=s[mw[mi][i]][i];
4226: s2=s[mw[mi+1][i]][i];
4227: if( s2 > nlstate){
4228: lli=log(out[s1][s2] - savm[s1][s2]);
4229: } else if ( s2==-1 ) { /* alive */
4230: for (j=1,survp=0. ; j<=nlstate; j++)
4231: survp += out[s1][j];
4232: lli= log(survp);
4233: }else{
4234: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4235: }
4236: ipmx +=1;
4237: sw += weight[i];
4238: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.343 brouard 4239: /* printf("num[i]=%09ld, i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",num[i],i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.226 brouard 4240: } /* end of wave */
4241: } /* end of individual */
4242: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4243: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4244: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4245: for(mi=1; mi<= wav[i]-1; mi++){
4246: for (ii=1;ii<=nlstate+ndeath;ii++)
4247: for (j=1;j<=nlstate+ndeath;j++){
4248: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4249: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4250: }
4251: for(d=0; d<dh[mi][i]; d++){
4252: newm=savm;
4253: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4254: cov[2]=agexact;
4255: if(nagesqr==1)
4256: cov[3]= agexact*agexact;
4257: for (kk=1; kk<=cptcovage;kk++) {
1.340 brouard 4258: if(!FixedV[Tvar[Tage[kk]]])
4259: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
4260: else
1.341 brouard 4261: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.226 brouard 4262: }
1.126 brouard 4263:
1.226 brouard 4264: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4265: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4266: savm=oldm;
4267: oldm=newm;
4268: } /* end mult */
4269:
4270: s1=s[mw[mi][i]][i];
4271: s2=s[mw[mi+1][i]][i];
4272: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4273: ipmx +=1;
4274: sw += weight[i];
4275: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4276: /*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]);*/
4277: } /* end of wave */
4278: } /* end of individual */
4279: } /* End of if */
4280: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4281: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4282: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4283: return -l;
1.126 brouard 4284: }
4285:
4286: /*************** log-likelihood *************/
4287: double funcone( double *x)
4288: {
1.228 brouard 4289: /* Same as func but slower because of a lot of printf and if */
1.335 brouard 4290: int i, ii, j, k, mi, d, kk, kf=0;
1.228 brouard 4291: int ioffset=0;
1.339 brouard 4292: int ipos=0,iposold=0,ncovv=0;
4293:
1.340 brouard 4294: double cotvarv, cotvarvold;
1.131 brouard 4295: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4296: double **out;
4297: double lli; /* Individual log likelihood */
4298: double llt;
4299: int s1, s2;
1.228 brouard 4300: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4301:
1.126 brouard 4302: double bbh, survp;
1.187 brouard 4303: double agexact;
1.214 brouard 4304: double agebegin, ageend;
1.126 brouard 4305: /*extern weight */
4306: /* We are differentiating ll according to initial status */
4307: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4308: /*for(i=1;i<imx;i++)
4309: printf(" %d\n",s[4][i]);
4310: */
4311: cov[1]=1.;
4312:
4313: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4314: ioffset=0;
4315: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.336 brouard 4316: /* Computes the values of the ncovmodel covariates of the model
4317: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
4318: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
4319: to be observed in j being in i according to the model.
4320: */
1.243 brouard 4321: /* ioffset=2+nagesqr+cptcovage; */
4322: ioffset=2+nagesqr;
1.232 brouard 4323: /* Fixed */
1.224 brouard 4324: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4325: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.335 brouard 4326: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
1.339 brouard 4327: /* printf("Debug3 TvarFind[%d]=%d",kf, TvarFind[kf]); */
4328: /* printf(" Tvar[TvarFind[kf]]=%d", Tvar[TvarFind[kf]]); */
4329: /* printf(" i=%d covar[Tvar[TvarFind[kf]]][i]=%f\n",i,covar[Tvar[TvarFind[kf]]][i]); */
1.335 brouard 4330: 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 4331: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4332: /* cov[2+6]=covar[Tvar[6]][i]; */
4333: /* cov[2+6]=covar[2][i]; V2 */
4334: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4335: /* cov[2+7]=covar[Tvar[7]][i]; */
4336: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4337: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4338: /* cov[2+9]=covar[Tvar[9]][i]; */
4339: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4340: }
1.336 brouard 4341: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
4342: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
4343: has been calculated etc */
4344: /* For an individual i, wav[i] gives the number of effective waves */
4345: /* We compute the contribution to Likelihood of each effective transition
4346: mw[mi][i] is real wave of the mi th effectve wave */
4347: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
4348: s2=s[mw[mi+1][i]][i];
1.341 brouard 4349: And the iv th varying covariate in the DATA is the cotvar[mw[mi+1][i]][ncovcol+nqv+iv][i]
1.336 brouard 4350: */
4351: /* This part may be useless now because everythin should be in covar */
1.232 brouard 4352: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4353: /* 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?)*\/ */
4354: /* } */
1.231 brouard 4355: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4356: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4357: /* } */
1.225 brouard 4358:
1.233 brouard 4359:
4360: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.339 brouard 4361: /* Wave varying (but not age varying) *//* V1+V3+age*V1+age*V3+V1*V3 with V4 tv and V5 tvq k= 1 to 5 and extra at V(5+1)=6 for V1*V3 */
4362: /* for(k=1; k <= ncovv ; k++){ /\* Varying covariates (single and product but no age )*\/ */
4363: /* /\* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; *\/ */
4364: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i]; */
4365: /* } */
4366:
4367: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
4368: /* model V1+V3+age*V1+age*V3+V1*V3 */
4369: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
4370: /* TvarVV[1]=V3 (first time varying in the model equation, TvarVV[2]=V1 (in V1*V3) TvarVV[3]=3(V3) */
4371: /* We need the position of the time varying or product in the model */
4372: /* TvarVVind={2,5,5}, for V3 at position 2 and then the product V1*V3 is decomposed into V1 and V3 but at same position 5 */
4373: /* TvarVV gives the variable name */
1.340 brouard 4374: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
4375: * k= 1 2 3 4 5 6 7 8 9
4376: * varying 1 2 3 4 5
4377: * ncovv 1 2 3 4 5 6 7 8
1.343 brouard 4378: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
1.340 brouard 4379: * TvarVVind 2 3 7 7 8 8 9 9
4380: * TvarFind[k] 1 0 0 0 0 0 0 0 0
4381: */
4382: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
4383: itv=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
4384: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
4385: if(TvarFind[itv]==0){ /* Not a fixed covariate */
1.341 brouard 4386: cotvarv=cotvar[mw[mi][i]][TvarVV[ncovv]][i]; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.340 brouard 4387: }else{ /* fixed covariate */
4388: cotvarv=covar[Tvar[TvarFind[itv]]][i];
4389: }
1.339 brouard 4390: if(ipos!=iposold){ /* Not a product or first of a product */
1.340 brouard 4391: cotvarvold=cotvarv;
4392: }else{ /* A second product */
4393: cotvarv=cotvarv*cotvarvold;
1.339 brouard 4394: }
4395: iposold=ipos;
1.340 brouard 4396: cov[ioffset+ipos]=cotvarv;
1.339 brouard 4397: /* For products */
4398: }
4399: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates single *\/ */
4400: /* iv=TvarVDind[itv]; /\* iv, position in the model equation of time varying covariate itv *\/ */
4401: /* /\* "V1+V3+age*V1+age*V3+V1*V3" with V3 time varying *\/ */
4402: /* /\* 1 2 3 4 5 *\/ */
4403: /* /\*itv 1 *\/ */
4404: /* /\* TvarVInd[1]= 2 *\/ */
4405: /* /\* iv= Tvar[Tmodelind[itv]]-ncovcol-nqv; /\\* Counting the # varying covariate from 1 to ntveff *\\/ *\/ */
4406: /* /\* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; *\/ */
4407: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; *\/ */
4408: /* /\* k=ioffset-2-nagesqr-cptcovage+itv; /\\* position in simple model *\\/ *\/ */
4409: /* /\* cov[ioffset+iv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; *\/ */
4410: /* cov[ioffset+iv]=cotvar[mw[mi][i]][itv][i]; */
4411: /* /\* printf(" i=%d,mi=%d,itv=%d,TmodelInvind[itv]=%d,cotvar[mw[mi][i]][itv][i]=%f\n", i, mi, itv, TvarVDind[itv],cotvar[mw[mi][i]][itv][i]); *\/ */
4412: /* } */
1.232 brouard 4413: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4414: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4415: /* /\* 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]); *\/ */
4416: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4417: /* } */
1.126 brouard 4418: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4419: for (j=1;j<=nlstate+ndeath;j++){
4420: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4421: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4422: }
1.214 brouard 4423:
4424: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4425: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4426: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4427: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4428: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4429: and mw[mi+1][i]. dh depends on stepm.*/
4430: newm=savm;
1.247 brouard 4431: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4432: cov[2]=agexact;
4433: if(nagesqr==1)
4434: cov[3]= agexact*agexact;
4435: for (kk=1; kk<=cptcovage;kk++) {
4436: if(!FixedV[Tvar[Tage[kk]]])
4437: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4438: else
1.341 brouard 4439: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]*agexact; /* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.242 brouard 4440: }
4441: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4442: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4443: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4444: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4445: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4446: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4447: savm=oldm;
4448: oldm=newm;
1.126 brouard 4449: } /* end mult */
1.336 brouard 4450: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
4451: /* But now since version 0.9 we anticipate for bias at large stepm.
4452: * If stepm is larger than one month (smallest stepm) and if the exact delay
4453: * (in months) between two waves is not a multiple of stepm, we rounded to
4454: * the nearest (and in case of equal distance, to the lowest) interval but now
4455: * we keep into memory the bias bh[mi][i] and also the previous matrix product
4456: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
4457: * probability in order to take into account the bias as a fraction of the way
4458: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
4459: * -stepm/2 to stepm/2 .
4460: * For stepm=1 the results are the same as for previous versions of Imach.
4461: * For stepm > 1 the results are less biased than in previous versions.
4462: */
1.126 brouard 4463: s1=s[mw[mi][i]][i];
4464: s2=s[mw[mi+1][i]][i];
1.217 brouard 4465: /* if(s2==-1){ */
1.268 brouard 4466: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4467: /* /\* exit(1); *\/ */
4468: /* } */
1.126 brouard 4469: bbh=(double)bh[mi][i]/(double)stepm;
4470: /* bias is positive if real duration
4471: * is higher than the multiple of stepm and negative otherwise.
4472: */
4473: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4474: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4475: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4476: for (j=1,survp=0. ; j<=nlstate; j++)
4477: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4478: lli= log(survp);
1.126 brouard 4479: }else if (mle==1){
1.242 brouard 4480: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4481: } else if(mle==2){
1.242 brouard 4482: 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 4483: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4484: 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 4485: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4486: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4487: } else{ /* mle=0 back to 1 */
1.242 brouard 4488: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4489: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4490: } /* End of if */
4491: ipmx +=1;
4492: sw += weight[i];
4493: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.342 brouard 4494: /* Printing covariates values for each contribution for checking */
1.343 brouard 4495: /* printf("num[i]=%09ld, i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",num[i],i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.126 brouard 4496: if(globpr){
1.246 brouard 4497: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4498: %11.6f %11.6f %11.6f ", \
1.242 brouard 4499: 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 4500: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.343 brouard 4501: /* printf("%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\ */
4502: /* %11.6f %11.6f %11.6f ", \ */
4503: /* num[i], agebegin, ageend, i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],weight[i]*gipmx/gsw, */
4504: /* 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2])); */
1.242 brouard 4505: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4506: llt +=ll[k]*gipmx/gsw;
4507: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
1.335 brouard 4508: /* printf(" %10.6f",-ll[k]*gipmx/gsw); */
1.242 brouard 4509: }
1.343 brouard 4510: fprintf(ficresilk," %10.6f ", -llt);
1.335 brouard 4511: /* printf(" %10.6f\n", -llt); */
1.342 brouard 4512: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
1.343 brouard 4513: /* fprintf(ficresilk,"%09ld ", num[i]); */ /* not necessary */
4514: for (kf=1; kf<=ncovf;kf++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
4515: fprintf(ficresilk," %g",covar[Tvar[TvarFind[kf]]][i]);
4516: }
4517: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Varying covariates (single and product but no age) including individual from products */
4518: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate*/
4519: if(ipos!=iposold){ /* Not a product or first of a product */
4520: fprintf(ficresilk," %g",cov[ioffset+ipos]);
4521: /* printf(" %g",cov[ioffset+ipos]); */
4522: }else{
4523: fprintf(ficresilk,"*");
4524: /* printf("*"); */
1.342 brouard 4525: }
1.343 brouard 4526: iposold=ipos;
4527: }
4528: for (kk=1; kk<=cptcovage;kk++) {
4529: if(!FixedV[Tvar[Tage[kk]]]){
4530: fprintf(ficresilk," %g*age",covar[Tvar[Tage[kk]]][i]);
4531: /* printf(" %g*age",covar[Tvar[Tage[kk]]][i]); */
4532: }else{
4533: fprintf(ficresilk," %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
4534: /* printf(" %g*age",cotvar[mw[mi][i]][Tvar[Tage[kk]]][i]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
1.342 brouard 4535: }
1.343 brouard 4536: }
4537: /* printf("\n"); */
1.342 brouard 4538: /* } /\* End debugILK *\/ */
4539: fprintf(ficresilk,"\n");
4540: } /* End if globpr */
1.335 brouard 4541: } /* end of wave */
4542: } /* end of individual */
4543: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
1.232 brouard 4544: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
1.335 brouard 4545: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4546: if(globpr==0){ /* First time we count the contributions and weights */
4547: gipmx=ipmx;
4548: gsw=sw;
4549: }
1.343 brouard 4550: return -l;
1.126 brouard 4551: }
4552:
4553:
4554: /*************** function likelione ***********/
1.292 brouard 4555: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4556: {
4557: /* This routine should help understanding what is done with
4558: the selection of individuals/waves and
4559: to check the exact contribution to the likelihood.
4560: Plotting could be done.
1.342 brouard 4561: */
4562: void pstamp(FILE *ficres);
1.343 brouard 4563: int k, kf, kk, kvar, ncovv, iposold, ipos;
1.126 brouard 4564:
4565: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4566: strcpy(fileresilk,"ILK_");
1.202 brouard 4567: strcat(fileresilk,fileresu);
1.126 brouard 4568: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4569: printf("Problem with resultfile: %s\n", fileresilk);
4570: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4571: }
1.342 brouard 4572: pstamp(ficresilk);fprintf(ficresilk,"# model=1+age+%s\n",model);
1.214 brouard 4573: 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");
4574: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4575: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4576: for(k=1; k<=nlstate; k++)
4577: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
1.342 brouard 4578: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total) ");
4579:
4580: /* if(debugILK){ /\* debugILK is set by a #d in a comment line *\/ */
4581: for(kf=1;kf <= ncovf; kf++){
4582: fprintf(ficresilk,"V%d",Tvar[TvarFind[kf]]);
4583: /* printf("V%d",Tvar[TvarFind[kf]]); */
4584: }
4585: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){
1.343 brouard 4586: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
1.342 brouard 4587: if(ipos!=iposold){ /* Not a product or first of a product */
4588: /* printf(" %d",ipos); */
4589: fprintf(ficresilk," V%d",TvarVV[ncovv]);
4590: }else{
4591: /* printf("*"); */
4592: fprintf(ficresilk,"*");
1.343 brouard 4593: }
1.342 brouard 4594: iposold=ipos;
4595: }
4596: for (kk=1; kk<=cptcovage;kk++) {
4597: if(!FixedV[Tvar[Tage[kk]]]){
4598: /* printf(" %d*age(Fixed)",Tvar[Tage[kk]]); */
4599: fprintf(ficresilk," %d*age(Fixed)",Tvar[Tage[kk]]);
4600: }else{
4601: fprintf(ficresilk," %d*age(Varying)",Tvar[Tage[kk]]);/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
4602: /* printf(" %d*age(Varying)",Tvar[Tage[kk]]);/\* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) *\/ */
4603: }
4604: }
4605: /* } /\* End if debugILK *\/ */
4606: /* printf("\n"); */
4607: fprintf(ficresilk,"\n");
4608: } /* End glogpri */
1.126 brouard 4609:
1.292 brouard 4610: *fretone=(*func)(p);
1.126 brouard 4611: if(*globpri !=0){
4612: fclose(ficresilk);
1.205 brouard 4613: if (mle ==0)
4614: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4615: else if(mle >=1)
4616: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4617: 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 4618: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4619:
1.207 brouard 4620: fprintf(fichtm,"<br>- The function drawn is -2Log(L) in Log scale: by state of origin <a href=\"%s-ori.png\">%s-ori.png</a><br> \
1.343 brouard 4621: <img src=\"%s-ori.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4622: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.343 brouard 4623: <img src=\"%s-dest.png\">\n",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
4624:
4625: for (k=1; k<= nlstate ; k++) {
4626: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j. Dot's sizes are related to corresponding weight: <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br>\n \
4627: <img src=\"%s-p%dj.png\">\n",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4628: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
4629: /* kvar=Tvar[TvarFind[kf]]; */ /* variable */
4630: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j with colored covariate V%d. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \
4631: <img src=\"%s-p%dj-%d.png\">",k,k,Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,Tvar[TvarFind[kf]]);
4632: }
4633: for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /* Loop on the time varying extended covariates (with extension of Vn*Vm */
4634: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
4635: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
4636: /* printf("DebugILK fichtm ncovv=%d, kvar=TvarVV[ncovv]=V%d, ipos=TvarVVind[ncovv]=%d, Dummy[ipos]=%d, Typevar[ipos]=%d\n", ncovv,kvar,ipos,Dummy[ipos],Typevar[ipos]); */
4637: if(ipos!=iposold){ /* Not a product or first of a product */
4638: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
4639: /* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); */
4640: if(Dummy[ipos]==0 && Typevar[ipos]==0){ /* Only if dummy time varying: Dummy(0, 1=quant singor prod without age,2 dummy*age, 3quant*age) Typevar (0 single, 1=*age,2=Vn*vm) */
4641: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j with colored time varying dummy covariate V%d. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \
4642: <img src=\"%s-p%dj-%d.png\">",k,k,kvar,kvar,kvar,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,kvar);
4643: } /* End only for dummies time varying (single?) */
4644: }else{ /* Useless product */
4645: /* printf("*"); */
4646: /* fprintf(ficresilk,"*"); */
4647: }
4648: iposold=ipos;
4649: } /* For each time varying covariate */
4650: } /* End loop on states */
4651:
4652: /* if(debugILK){ */
4653: /* for(kf=1; kf <= ncovf; kf++){ /\* For each simple dummy covariate of the model *\/ */
4654: /* /\* kvar=Tvar[TvarFind[kf]]; *\/ /\* variable *\/ */
4655: /* for (k=1; k<= nlstate ; k++) { */
4656: /* fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j with colored covariate V%. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \ */
4657: /* <img src=\"%s-p%dj-%d.png\">",k,k,Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],Tvar[TvarFind[kf]],subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,Tvar[TvarFind[kf]]); */
4658: /* } */
4659: /* } */
4660: /* for(ncovv=1, iposold=0; ncovv <= ncovvt ; ncovv++){ /\* Loop on the time varying extended covariates (with extension of Vn*Vm *\/ */
4661: /* ipos=TvarVVind[ncovv]; /\* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate *\/ */
4662: /* kvar=TvarVV[ncovv]; /\* TvarVV={3, 1, 3} gives the name of each varying covariate *\/ */
4663: /* /\* printf("DebugILK fichtm ncovv=%d, kvar=TvarVV[ncovv]=V%d, ipos=TvarVVind[ncovv]=%d, Dummy[ipos]=%d, Typevar[ipos]=%d\n", ncovv,kvar,ipos,Dummy[ipos],Typevar[ipos]); *\/ */
4664: /* if(ipos!=iposold){ /\* Not a product or first of a product *\/ */
4665: /* /\* fprintf(ficresilk," V%d",TvarVV[ncovv]); *\/ */
4666: /* /\* printf(" DebugILK fichtm ipos=%d != iposold=%d\n", ipos, iposold); *\/ */
4667: /* if(Dummy[ipos]==0 && Typevar[ipos]==0){ /\* Only if dummy time varying: Dummy(0, 1=quant singor prod without age,2 dummy*age, 3quant*age) Typevar (0 single, 1=*age,2=Vn*vm) *\/ */
4668: /* for (k=1; k<= nlstate ; k++) { */
4669: /* fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j. Same dot size of all points but with a different color for transitions with dummy variable V%d=1 at beginning of transition (keeping former color for V%d=0): <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \ */
4670: /* <img src=\"%s-p%dj-%d.png\">",k,k,kvar,kvar,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,kvar); */
4671: /* } /\* End state *\/ */
4672: /* } /\* End only for dummies time varying (single?) *\/ */
4673: /* }else{ /\* Useless product *\/ */
4674: /* /\* printf("*"); *\/ */
4675: /* /\* fprintf(ficresilk,"*"); *\/ */
4676: /* } */
4677: /* iposold=ipos; */
4678: /* } /\* For each time varying covariate *\/ */
4679: /* }/\* End debugILK *\/ */
1.207 brouard 4680: fflush(fichtm);
1.343 brouard 4681: }/* End globpri */
1.126 brouard 4682: return;
4683: }
4684:
4685:
4686: /*********** Maximum Likelihood Estimation ***************/
4687:
4688: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4689: {
1.319 brouard 4690: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4691: double **xi;
4692: double fret;
4693: double fretone; /* Only one call to likelihood */
4694: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4695:
4696: #ifdef NLOPT
4697: int creturn;
4698: nlopt_opt opt;
4699: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4700: double *lb;
4701: double minf; /* the minimum objective value, upon return */
4702: double * p1; /* Shifted parameters from 0 instead of 1 */
4703: myfunc_data dinst, *d = &dinst;
4704: #endif
4705:
4706:
1.126 brouard 4707: xi=matrix(1,npar,1,npar);
4708: for (i=1;i<=npar;i++)
4709: for (j=1;j<=npar;j++)
4710: xi[i][j]=(i==j ? 1.0 : 0.0);
4711: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4712: strcpy(filerespow,"POW_");
1.126 brouard 4713: strcat(filerespow,fileres);
4714: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4715: printf("Problem with resultfile: %s\n", filerespow);
4716: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4717: }
4718: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4719: for (i=1;i<=nlstate;i++)
4720: for(j=1;j<=nlstate+ndeath;j++)
4721: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4722: fprintf(ficrespow,"\n");
1.162 brouard 4723: #ifdef POWELL
1.319 brouard 4724: #ifdef LINMINORIGINAL
4725: #else /* LINMINORIGINAL */
4726:
4727: flatdir=ivector(1,npar);
4728: for (j=1;j<=npar;j++) flatdir[j]=0;
4729: #endif /*LINMINORIGINAL */
4730:
4731: #ifdef FLATSUP
4732: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4733: /* reorganizing p by suppressing flat directions */
4734: for(i=1, jk=1; i <=nlstate; i++){
4735: for(k=1; k <=(nlstate+ndeath); k++){
4736: if (k != i) {
4737: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4738: if(flatdir[jk]==1){
4739: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4740: }
4741: for(j=1; j <=ncovmodel; j++){
4742: printf("%12.7f ",p[jk]);
4743: jk++;
4744: }
4745: printf("\n");
4746: }
4747: }
4748: }
4749: /* skipping */
4750: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4751: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4752: for(k=1; k <=(nlstate+ndeath); k++){
4753: if (k != i) {
4754: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4755: if(flatdir[jk]==1){
4756: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4757: for(j=1; j <=ncovmodel; jk++,j++){
4758: printf(" p[%d]=%12.7f",jk, p[jk]);
4759: /*q[jjk]=p[jk];*/
4760: }
4761: }else{
4762: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4763: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4764: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4765: /*q[jjk]=p[jk];*/
4766: }
4767: }
4768: printf("\n");
4769: }
4770: fflush(stdout);
4771: }
4772: }
4773: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4774: #else /* FLATSUP */
1.126 brouard 4775: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4776: #endif /* FLATSUP */
4777:
4778: #ifdef LINMINORIGINAL
4779: #else
4780: free_ivector(flatdir,1,npar);
4781: #endif /* LINMINORIGINAL*/
4782: #endif /* POWELL */
1.126 brouard 4783:
1.162 brouard 4784: #ifdef NLOPT
4785: #ifdef NEWUOA
4786: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4787: #else
4788: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4789: #endif
4790: lb=vector(0,npar-1);
4791: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4792: nlopt_set_lower_bounds(opt, lb);
4793: nlopt_set_initial_step1(opt, 0.1);
4794:
4795: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4796: d->function = func;
4797: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4798: nlopt_set_min_objective(opt, myfunc, d);
4799: nlopt_set_xtol_rel(opt, ftol);
4800: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4801: printf("nlopt failed! %d\n",creturn);
4802: }
4803: else {
4804: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4805: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4806: iter=1; /* not equal */
4807: }
4808: nlopt_destroy(opt);
4809: #endif
1.319 brouard 4810: #ifdef FLATSUP
4811: /* npared = npar -flatd/ncovmodel; */
4812: /* xired= matrix(1,npared,1,npared); */
4813: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4814: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4815: /* free_matrix(xire,1,npared,1,npared); */
4816: #else /* FLATSUP */
4817: #endif /* FLATSUP */
1.126 brouard 4818: free_matrix(xi,1,npar,1,npar);
4819: fclose(ficrespow);
1.203 brouard 4820: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4821: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4822: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4823:
4824: }
4825:
4826: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4827: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4828: {
4829: double **a,**y,*x,pd;
1.203 brouard 4830: /* double **hess; */
1.164 brouard 4831: int i, j;
1.126 brouard 4832: int *indx;
4833:
4834: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4835: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4836: void lubksb(double **a, int npar, int *indx, double b[]) ;
4837: void ludcmp(double **a, int npar, int *indx, double *d) ;
4838: double gompertz(double p[]);
1.203 brouard 4839: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4840:
4841: printf("\nCalculation of the hessian matrix. Wait...\n");
4842: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4843: for (i=1;i<=npar;i++){
1.203 brouard 4844: printf("%d-",i);fflush(stdout);
4845: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4846:
4847: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4848:
4849: /* printf(" %f ",p[i]);
4850: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4851: }
4852:
4853: for (i=1;i<=npar;i++) {
4854: for (j=1;j<=npar;j++) {
4855: if (j>i) {
1.203 brouard 4856: printf(".%d-%d",i,j);fflush(stdout);
4857: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4858: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4859:
4860: hess[j][i]=hess[i][j];
4861: /*printf(" %lf ",hess[i][j]);*/
4862: }
4863: }
4864: }
4865: printf("\n");
4866: fprintf(ficlog,"\n");
4867:
4868: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4869: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4870:
4871: a=matrix(1,npar,1,npar);
4872: y=matrix(1,npar,1,npar);
4873: x=vector(1,npar);
4874: indx=ivector(1,npar);
4875: for (i=1;i<=npar;i++)
4876: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4877: ludcmp(a,npar,indx,&pd);
4878:
4879: for (j=1;j<=npar;j++) {
4880: for (i=1;i<=npar;i++) x[i]=0;
4881: x[j]=1;
4882: lubksb(a,npar,indx,x);
4883: for (i=1;i<=npar;i++){
4884: matcov[i][j]=x[i];
4885: }
4886: }
4887:
4888: printf("\n#Hessian matrix#\n");
4889: fprintf(ficlog,"\n#Hessian matrix#\n");
4890: for (i=1;i<=npar;i++) {
4891: for (j=1;j<=npar;j++) {
1.203 brouard 4892: printf("%.6e ",hess[i][j]);
4893: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4894: }
4895: printf("\n");
4896: fprintf(ficlog,"\n");
4897: }
4898:
1.203 brouard 4899: /* printf("\n#Covariance matrix#\n"); */
4900: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4901: /* for (i=1;i<=npar;i++) { */
4902: /* for (j=1;j<=npar;j++) { */
4903: /* printf("%.6e ",matcov[i][j]); */
4904: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4905: /* } */
4906: /* printf("\n"); */
4907: /* fprintf(ficlog,"\n"); */
4908: /* } */
4909:
1.126 brouard 4910: /* Recompute Inverse */
1.203 brouard 4911: /* for (i=1;i<=npar;i++) */
4912: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4913: /* ludcmp(a,npar,indx,&pd); */
4914:
4915: /* printf("\n#Hessian matrix recomputed#\n"); */
4916:
4917: /* for (j=1;j<=npar;j++) { */
4918: /* for (i=1;i<=npar;i++) x[i]=0; */
4919: /* x[j]=1; */
4920: /* lubksb(a,npar,indx,x); */
4921: /* for (i=1;i<=npar;i++){ */
4922: /* y[i][j]=x[i]; */
4923: /* printf("%.3e ",y[i][j]); */
4924: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4925: /* } */
4926: /* printf("\n"); */
4927: /* fprintf(ficlog,"\n"); */
4928: /* } */
4929:
4930: /* Verifying the inverse matrix */
4931: #ifdef DEBUGHESS
4932: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4933:
1.203 brouard 4934: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4935: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4936:
4937: for (j=1;j<=npar;j++) {
4938: for (i=1;i<=npar;i++){
1.203 brouard 4939: printf("%.2f ",y[i][j]);
4940: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4941: }
4942: printf("\n");
4943: fprintf(ficlog,"\n");
4944: }
1.203 brouard 4945: #endif
1.126 brouard 4946:
4947: free_matrix(a,1,npar,1,npar);
4948: free_matrix(y,1,npar,1,npar);
4949: free_vector(x,1,npar);
4950: free_ivector(indx,1,npar);
1.203 brouard 4951: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4952:
4953:
4954: }
4955:
4956: /*************** hessian matrix ****************/
4957: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4958: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4959: int i;
4960: int l=1, lmax=20;
1.203 brouard 4961: double k1,k2, res, fx;
1.132 brouard 4962: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4963: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4964: int k=0,kmax=10;
4965: double l1;
4966:
4967: fx=func(x);
4968: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4969: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4970: l1=pow(10,l);
4971: delts=delt;
4972: for(k=1 ; k <kmax; k=k+1){
4973: delt = delta*(l1*k);
4974: p2[theta]=x[theta] +delt;
1.145 brouard 4975: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4976: p2[theta]=x[theta]-delt;
4977: k2=func(p2)-fx;
4978: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4979: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4980:
1.203 brouard 4981: #ifdef DEBUGHESSII
1.126 brouard 4982: 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);
4983: 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);
4984: #endif
4985: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4986: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4987: k=kmax;
4988: }
4989: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4990: k=kmax; l=lmax*10;
1.126 brouard 4991: }
4992: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4993: delts=delt;
4994: }
1.203 brouard 4995: } /* End loop k */
1.126 brouard 4996: }
4997: delti[theta]=delts;
4998: return res;
4999:
5000: }
5001:
1.203 brouard 5002: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 5003: {
5004: int i;
1.164 brouard 5005: int l=1, lmax=20;
1.126 brouard 5006: double k1,k2,k3,k4,res,fx;
1.132 brouard 5007: double p2[MAXPARM+1];
1.203 brouard 5008: int k, kmax=1;
5009: double v1, v2, cv12, lc1, lc2;
1.208 brouard 5010:
5011: int firstime=0;
1.203 brouard 5012:
1.126 brouard 5013: fx=func(x);
1.203 brouard 5014: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 5015: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 5016: p2[thetai]=x[thetai]+delti[thetai]*k;
5017: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 5018: k1=func(p2)-fx;
5019:
1.203 brouard 5020: p2[thetai]=x[thetai]+delti[thetai]*k;
5021: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 5022: k2=func(p2)-fx;
5023:
1.203 brouard 5024: p2[thetai]=x[thetai]-delti[thetai]*k;
5025: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 5026: k3=func(p2)-fx;
5027:
1.203 brouard 5028: p2[thetai]=x[thetai]-delti[thetai]*k;
5029: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 5030: k4=func(p2)-fx;
1.203 brouard 5031: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
5032: if(k1*k2*k3*k4 <0.){
1.208 brouard 5033: firstime=1;
1.203 brouard 5034: kmax=kmax+10;
1.208 brouard 5035: }
5036: if(kmax >=10 || firstime ==1){
1.246 brouard 5037: 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);
5038: 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 5039: 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);
5040: 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);
5041: }
5042: #ifdef DEBUGHESSIJ
5043: v1=hess[thetai][thetai];
5044: v2=hess[thetaj][thetaj];
5045: cv12=res;
5046: /* Computing eigen value of Hessian matrix */
5047: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5048: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
5049: if ((lc2 <0) || (lc1 <0) ){
5050: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
5051: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
5052: 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);
5053: 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);
5054: }
1.126 brouard 5055: #endif
5056: }
5057: return res;
5058: }
5059:
1.203 brouard 5060: /* Not done yet: Was supposed to fix if not exactly at the maximum */
5061: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
5062: /* { */
5063: /* int i; */
5064: /* int l=1, lmax=20; */
5065: /* double k1,k2,k3,k4,res,fx; */
5066: /* double p2[MAXPARM+1]; */
5067: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
5068: /* int k=0,kmax=10; */
5069: /* double l1; */
5070:
5071: /* fx=func(x); */
5072: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
5073: /* l1=pow(10,l); */
5074: /* delts=delt; */
5075: /* for(k=1 ; k <kmax; k=k+1){ */
5076: /* delt = delti*(l1*k); */
5077: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
5078: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5079: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5080: /* k1=func(p2)-fx; */
5081:
5082: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
5083: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5084: /* k2=func(p2)-fx; */
5085:
5086: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5087: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
5088: /* k3=func(p2)-fx; */
5089:
5090: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
5091: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
5092: /* k4=func(p2)-fx; */
5093: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
5094: /* #ifdef DEBUGHESSIJ */
5095: /* 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); */
5096: /* 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); */
5097: /* #endif */
5098: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
5099: /* k=kmax; */
5100: /* } */
5101: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
5102: /* k=kmax; l=lmax*10; */
5103: /* } */
5104: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
5105: /* delts=delt; */
5106: /* } */
5107: /* } /\* End loop k *\/ */
5108: /* } */
5109: /* delti[theta]=delts; */
5110: /* return res; */
5111: /* } */
5112:
5113:
1.126 brouard 5114: /************** Inverse of matrix **************/
5115: void ludcmp(double **a, int n, int *indx, double *d)
5116: {
5117: int i,imax,j,k;
5118: double big,dum,sum,temp;
5119: double *vv;
5120:
5121: vv=vector(1,n);
5122: *d=1.0;
5123: for (i=1;i<=n;i++) {
5124: big=0.0;
5125: for (j=1;j<=n;j++)
5126: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 5127: if (big == 0.0){
5128: printf(" Singular Hessian matrix at row %d:\n",i);
5129: for (j=1;j<=n;j++) {
5130: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
5131: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
5132: }
5133: fflush(ficlog);
5134: fclose(ficlog);
5135: nrerror("Singular matrix in routine ludcmp");
5136: }
1.126 brouard 5137: vv[i]=1.0/big;
5138: }
5139: for (j=1;j<=n;j++) {
5140: for (i=1;i<j;i++) {
5141: sum=a[i][j];
5142: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
5143: a[i][j]=sum;
5144: }
5145: big=0.0;
5146: for (i=j;i<=n;i++) {
5147: sum=a[i][j];
5148: for (k=1;k<j;k++)
5149: sum -= a[i][k]*a[k][j];
5150: a[i][j]=sum;
5151: if ( (dum=vv[i]*fabs(sum)) >= big) {
5152: big=dum;
5153: imax=i;
5154: }
5155: }
5156: if (j != imax) {
5157: for (k=1;k<=n;k++) {
5158: dum=a[imax][k];
5159: a[imax][k]=a[j][k];
5160: a[j][k]=dum;
5161: }
5162: *d = -(*d);
5163: vv[imax]=vv[j];
5164: }
5165: indx[j]=imax;
5166: if (a[j][j] == 0.0) a[j][j]=TINY;
5167: if (j != n) {
5168: dum=1.0/(a[j][j]);
5169: for (i=j+1;i<=n;i++) a[i][j] *= dum;
5170: }
5171: }
5172: free_vector(vv,1,n); /* Doesn't work */
5173: ;
5174: }
5175:
5176: void lubksb(double **a, int n, int *indx, double b[])
5177: {
5178: int i,ii=0,ip,j;
5179: double sum;
5180:
5181: for (i=1;i<=n;i++) {
5182: ip=indx[i];
5183: sum=b[ip];
5184: b[ip]=b[i];
5185: if (ii)
5186: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
5187: else if (sum) ii=i;
5188: b[i]=sum;
5189: }
5190: for (i=n;i>=1;i--) {
5191: sum=b[i];
5192: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
5193: b[i]=sum/a[i][i];
5194: }
5195: }
5196:
5197: void pstamp(FILE *fichier)
5198: {
1.196 brouard 5199: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 5200: }
5201:
1.297 brouard 5202: void date2dmy(double date,double *day, double *month, double *year){
5203: double yp=0., yp1=0., yp2=0.;
5204:
5205: yp1=modf(date,&yp);/* extracts integral of date in yp and
5206: fractional in yp1 */
5207: *year=yp;
5208: yp2=modf((yp1*12),&yp);
5209: *month=yp;
5210: yp1=modf((yp2*30.5),&yp);
5211: *day=yp;
5212: if(*day==0) *day=1;
5213: if(*month==0) *month=1;
5214: }
5215:
1.253 brouard 5216:
5217:
1.126 brouard 5218: /************ Frequencies ********************/
1.251 brouard 5219: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 5220: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
5221: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 5222: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 5223: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 5224: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 5225: int iind=0, iage=0;
5226: int mi; /* Effective wave */
5227: int first;
5228: double ***freq; /* Frequencies */
1.268 brouard 5229: 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 */
5230: 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 5231: double *meanq, *stdq, *idq;
1.226 brouard 5232: double **meanqt;
5233: double *pp, **prop, *posprop, *pospropt;
5234: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
5235: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
5236: double agebegin, ageend;
5237:
5238: pp=vector(1,nlstate);
1.251 brouard 5239: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5240: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
5241: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
5242: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
5243: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 5244: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 5245: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 5246: meanqt=matrix(1,lastpass,1,nqtveff);
5247: strcpy(fileresp,"P_");
5248: strcat(fileresp,fileresu);
5249: /*strcat(fileresphtm,fileresu);*/
5250: if((ficresp=fopen(fileresp,"w"))==NULL) {
5251: printf("Problem with prevalence resultfile: %s\n", fileresp);
5252: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
5253: exit(0);
5254: }
1.240 brouard 5255:
1.226 brouard 5256: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
5257: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
5258: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5259: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
5260: fflush(ficlog);
5261: exit(70);
5262: }
5263: else{
5264: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 5265: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5266: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5267: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5268: }
1.319 brouard 5269: 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 5270:
1.226 brouard 5271: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
5272: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
5273: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5274: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
5275: fflush(ficlog);
5276: exit(70);
1.240 brouard 5277: } else{
1.226 brouard 5278: 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 5279: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 5280: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 5281: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
5282: }
1.319 brouard 5283: 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 5284:
1.253 brouard 5285: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
5286: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 5287: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 5288: j1=0;
1.126 brouard 5289:
1.227 brouard 5290: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
1.335 brouard 5291: j=cptcoveff; /* Only simple dummy covariates used in the model */
1.330 brouard 5292: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 5293: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 5294:
5295:
1.226 brouard 5296: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
5297: reference=low_education V1=0,V2=0
5298: med_educ V1=1 V2=0,
5299: high_educ V1=0 V2=1
1.330 brouard 5300: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 5301: */
1.249 brouard 5302: dateintsum=0;
5303: k2cpt=0;
5304:
1.253 brouard 5305: if(cptcoveff == 0 )
1.265 brouard 5306: nl=1; /* Constant and age model only */
1.253 brouard 5307: else
5308: nl=2;
1.265 brouard 5309:
5310: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
5311: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.335 brouard 5312: * Loop on j1(1 to 2**cptcoveff) covariate combination
1.265 brouard 5313: * freq[s1][s2][iage] =0.
5314: * Loop on iind
5315: * ++freq[s1][s2][iage] weighted
5316: * end iind
5317: * if covariate and j!0
5318: * headers Variable on one line
5319: * endif cov j!=0
5320: * header of frequency table by age
5321: * Loop on age
5322: * pp[s1]+=freq[s1][s2][iage] weighted
5323: * pos+=freq[s1][s2][iage] weighted
5324: * Loop on s1 initial state
5325: * fprintf(ficresp
5326: * end s1
5327: * end age
5328: * if j!=0 computes starting values
5329: * end compute starting values
5330: * end j1
5331: * end nl
5332: */
1.253 brouard 5333: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
5334: if(nj==1)
5335: j=0; /* First pass for the constant */
1.265 brouard 5336: else{
1.335 brouard 5337: 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 5338: }
1.251 brouard 5339: first=1;
1.332 brouard 5340: 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 5341: posproptt=0.;
1.330 brouard 5342: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 5343: scanf("%d", i);*/
5344: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 5345: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 5346: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 5347: freq[i][s2][m]=0;
1.251 brouard 5348:
5349: for (i=1; i<=nlstate; i++) {
1.240 brouard 5350: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 5351: prop[i][m]=0;
5352: posprop[i]=0;
5353: pospropt[i]=0;
5354: }
1.283 brouard 5355: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 5356: idq[z1]=0.;
5357: meanq[z1]=0.;
5358: stdq[z1]=0.;
1.283 brouard 5359: }
5360: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 5361: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 5362: /* meanqt[m][z1]=0.; */
5363: /* } */
5364: /* } */
1.251 brouard 5365: /* dateintsum=0; */
5366: /* k2cpt=0; */
5367:
1.265 brouard 5368: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 5369: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
5370: bool=1;
5371: if(j !=0){
5372: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.335 brouard 5373: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
5374: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
1.251 brouard 5375: /* if(Tvaraff[z1] ==-20){ */
5376: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
5377: /* }else if(Tvaraff[z1] ==-10){ */
5378: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 5379: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.335 brouard 5380: /* 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); */
5381: if(Tvaraff[z1]<1 || Tvaraff[z1]>=NCOVMAX)
1.338 brouard 5382: printf("Error Tvaraff[z1]=%d<1 or >=%d, cptcoveff=%d model=1+age+%s\n",Tvaraff[z1],NCOVMAX, cptcoveff, model);
1.332 brouard 5383: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 5384: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 5385: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 5386: /* 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", */
5387: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
5388: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 5389: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
5390: } /* Onlyf fixed */
5391: } /* end z1 */
1.335 brouard 5392: } /* cptcoveff > 0 */
1.251 brouard 5393: } /* end any */
5394: }/* end j==0 */
1.265 brouard 5395: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 5396: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 5397: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 5398: m=mw[mi][iind];
5399: if(j!=0){
5400: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.335 brouard 5401: for (z1=1; z1<=cptcoveff; z1++) {
1.251 brouard 5402: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5403: /* iv= Tvar[Tmodelind[z1]]-ncovcol-nqv; /\* Good *\/ */
5404: iv= Tvar[Tmodelind[z1]]; /* Good *//* because cotvar starts now at first at ncovcol+nqv+ntv */
1.332 brouard 5405: 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 5406: value is -1, we don't select. It differs from the
5407: constant and age model which counts them. */
5408: bool=0; /* not selected */
5409: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.334 brouard 5410: /* i1=Tvaraff[z1]; */
5411: /* i2=TnsdVar[i1]; */
5412: /* i3=nbcode[i1][i2]; */
5413: /* i4=covar[i1][iind]; */
5414: /* if(i4 != i3){ */
5415: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) { /* Bug valgrind */
1.251 brouard 5416: bool=0;
5417: }
5418: }
5419: }
5420: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
5421: } /* end j==0 */
5422: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 5423: if(bool==1){ /*Selected */
1.251 brouard 5424: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
5425: and mw[mi+1][iind]. dh depends on stepm. */
5426: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
5427: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
5428: if(m >=firstpass && m <=lastpass){
5429: k2=anint[m][iind]+(mint[m][iind]/12.);
5430: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
5431: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
5432: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
5433: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
5434: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
5435: if (m<lastpass) {
5436: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
5437: /* 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]); */
5438: if(s[m][iind]==-1)
5439: 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.));
5440: 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 5441: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
5442: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 5443: idq[z1]=idq[z1]+weight[iind];
5444: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
5445: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
5446: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 5447: }
1.284 brouard 5448: }
1.251 brouard 5449: /* if((int)agev[m][iind] == 55) */
5450: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5451: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5452: 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 5453: }
1.251 brouard 5454: } /* end if between passes */
5455: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5456: dateintsum=dateintsum+k2; /* on all covariates ?*/
5457: k2cpt++;
5458: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5459: }
1.251 brouard 5460: }else{
5461: bool=1;
5462: }/* end bool 2 */
5463: } /* end m */
1.284 brouard 5464: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5465: /* idq[z1]=idq[z1]+weight[iind]; */
5466: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5467: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5468: /* } */
1.251 brouard 5469: } /* end bool */
5470: } /* end iind = 1 to imx */
1.319 brouard 5471: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5472: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5473:
5474:
5475: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.335 brouard 5476: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5477: pstamp(ficresp);
1.335 brouard 5478: if (cptcoveff>0 && j!=0){
1.265 brouard 5479: pstamp(ficresp);
1.251 brouard 5480: printf( "\n#********** Variable ");
5481: fprintf(ficresp, "\n#********** Variable ");
5482: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5483: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5484: fprintf(ficlog, "\n#********** Variable ");
1.340 brouard 5485: for (z1=1; z1<=cptcoveff; z1++){
1.251 brouard 5486: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 5487: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5488: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5489: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5490: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5491: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5492: }else{
1.330 brouard 5493: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5494: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5495: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5496: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5497: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5498: }
5499: }
5500: printf( "**********\n#");
5501: fprintf(ficresp, "**********\n#");
5502: fprintf(ficresphtm, "**********</h3>\n");
5503: fprintf(ficresphtmfr, "**********</h3>\n");
5504: fprintf(ficlog, "**********\n");
5505: }
1.284 brouard 5506: /*
5507: Printing means of quantitative variables if any
5508: */
5509: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5510: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5511: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5512: if(weightopt==1){
5513: printf(" Weighted mean and standard deviation of");
5514: fprintf(ficlog," Weighted mean and standard deviation of");
5515: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5516: }
1.311 brouard 5517: /* mu = \frac{w x}{\sum w}
5518: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5519: */
5520: 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]));
5521: 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]));
5522: 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 5523: }
5524: /* for (z1=1; z1<= nqtveff; z1++) { */
5525: /* for(m=1;m<=lastpass;m++){ */
5526: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5527: /* } */
5528: /* } */
1.283 brouard 5529:
1.251 brouard 5530: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.335 brouard 5531: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5532: fprintf(ficresp, " Age");
1.335 brouard 5533: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) {
5534: 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]]);
5535: fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5536: }
1.251 brouard 5537: for(i=1; i<=nlstate;i++) {
1.335 brouard 5538: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5539: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5540: }
1.335 brouard 5541: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5542: fprintf(ficresphtm, "\n");
5543:
5544: /* Header of frequency table by age */
5545: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5546: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5547: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5548: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5549: if(s2!=0 && m!=0)
5550: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5551: }
1.226 brouard 5552: }
1.251 brouard 5553: fprintf(ficresphtmfr, "\n");
5554:
5555: /* For each age */
5556: for(iage=iagemin; iage <= iagemax+3; iage++){
5557: fprintf(ficresphtm,"<tr>");
5558: if(iage==iagemax+1){
5559: fprintf(ficlog,"1");
5560: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5561: }else if(iage==iagemax+2){
5562: fprintf(ficlog,"0");
5563: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5564: }else if(iage==iagemax+3){
5565: fprintf(ficlog,"Total");
5566: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5567: }else{
1.240 brouard 5568: if(first==1){
1.251 brouard 5569: first=0;
5570: printf("See log file for details...\n");
5571: }
5572: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5573: fprintf(ficlog,"Age %d", iage);
5574: }
1.265 brouard 5575: for(s1=1; s1 <=nlstate ; s1++){
5576: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5577: pp[s1] += freq[s1][m][iage];
1.251 brouard 5578: }
1.265 brouard 5579: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5580: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5581: pos += freq[s1][m][iage];
5582: if(pp[s1]>=1.e-10){
1.251 brouard 5583: if(first==1){
1.265 brouard 5584: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5585: }
1.265 brouard 5586: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5587: }else{
5588: if(first==1)
1.265 brouard 5589: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5590: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5591: }
5592: }
5593:
1.265 brouard 5594: for(s1=1; s1 <=nlstate ; s1++){
5595: /* posprop[s1]=0; */
5596: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5597: pp[s1] += freq[s1][m][iage];
5598: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5599:
5600: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5601: pos += pp[s1]; /* pos is the total number of transitions until this age */
5602: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5603: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5604: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5605: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5606: }
5607:
5608: /* Writing ficresp */
1.335 brouard 5609: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5610: if( iage <= iagemax){
5611: fprintf(ficresp," %d",iage);
5612: }
5613: }else if( nj==2){
5614: if( iage <= iagemax){
5615: fprintf(ficresp," %d",iage);
1.335 brouard 5616: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 5617: }
1.240 brouard 5618: }
1.265 brouard 5619: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5620: if(pos>=1.e-5){
1.251 brouard 5621: if(first==1)
1.265 brouard 5622: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5623: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5624: }else{
5625: if(first==1)
1.265 brouard 5626: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5627: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5628: }
5629: if( iage <= iagemax){
5630: if(pos>=1.e-5){
1.335 brouard 5631: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5632: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5633: }else if( nj==2){
5634: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5635: }
5636: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5637: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5638: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5639: } else{
1.335 brouard 5640: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5641: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5642: }
1.240 brouard 5643: }
1.265 brouard 5644: pospropt[s1] +=posprop[s1];
5645: } /* end loop s1 */
1.251 brouard 5646: /* pospropt=0.; */
1.265 brouard 5647: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5648: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5649: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5650: if(first==1){
1.265 brouard 5651: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5652: }
1.265 brouard 5653: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5654: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5655: }
1.265 brouard 5656: if(s1!=0 && m!=0)
5657: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5658: }
1.265 brouard 5659: } /* end loop s1 */
1.251 brouard 5660: posproptt=0.;
1.265 brouard 5661: for(s1=1; s1 <=nlstate; s1++){
5662: posproptt += pospropt[s1];
1.251 brouard 5663: }
5664: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5665: fprintf(ficresphtm,"</tr>\n");
1.335 brouard 5666: if((cptcoveff==0 && nj==1)|| nj==2 ) {
1.265 brouard 5667: if(iage <= iagemax)
5668: fprintf(ficresp,"\n");
1.240 brouard 5669: }
1.251 brouard 5670: if(first==1)
5671: printf("Others in log...\n");
5672: fprintf(ficlog,"\n");
5673: } /* end loop age iage */
1.265 brouard 5674:
1.251 brouard 5675: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5676: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5677: if(posproptt < 1.e-5){
1.265 brouard 5678: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5679: }else{
1.265 brouard 5680: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5681: }
1.226 brouard 5682: }
1.251 brouard 5683: fprintf(ficresphtm,"</tr>\n");
5684: fprintf(ficresphtm,"</table>\n");
5685: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5686: if(posproptt < 1.e-5){
1.251 brouard 5687: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5688: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5689: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5690: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5691: invalidvarcomb[j1]=1;
1.226 brouard 5692: }else{
1.338 brouard 5693: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced (or no resultline).</p>",j1);
1.251 brouard 5694: invalidvarcomb[j1]=0;
1.226 brouard 5695: }
1.251 brouard 5696: fprintf(ficresphtmfr,"</table>\n");
5697: fprintf(ficlog,"\n");
5698: if(j!=0){
5699: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5700: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5701: for(k=1; k <=(nlstate+ndeath); k++){
5702: if (k != i) {
1.265 brouard 5703: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5704: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5705: if(j1==1){ /* All dummy covariates to zero */
5706: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5707: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5708: printf("%d%d ",i,k);
5709: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5710: 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]));
5711: 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]));
5712: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5713: }
1.253 brouard 5714: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5715: for(iage=iagemin; iage <= iagemax+3; iage++){
5716: x[iage]= (double)iage;
5717: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5718: /* 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 5719: }
1.268 brouard 5720: /* Some are not finite, but linreg will ignore these ages */
5721: no=0;
1.253 brouard 5722: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5723: pstart[s1]=b;
5724: pstart[s1-1]=a;
1.252 brouard 5725: }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 */
5726: 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]);
5727: 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 5728: 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 5729: printf("%d%d ",i,k);
5730: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5731: 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 5732: }else{ /* Other cases, like quantitative fixed or varying covariates */
5733: ;
5734: }
5735: /* printf("%12.7f )", param[i][jj][k]); */
5736: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5737: s1++;
1.251 brouard 5738: } /* end jj */
5739: } /* end k!= i */
5740: } /* end k */
1.265 brouard 5741: } /* end i, s1 */
1.251 brouard 5742: } /* end j !=0 */
5743: } /* end selected combination of covariate j1 */
5744: if(j==0){ /* We can estimate starting values from the occurences in each case */
5745: printf("#Freqsummary: Starting values for the constants:\n");
5746: fprintf(ficlog,"\n");
1.265 brouard 5747: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5748: for(k=1; k <=(nlstate+ndeath); k++){
5749: if (k != i) {
5750: printf("%d%d ",i,k);
5751: fprintf(ficlog,"%d%d ",i,k);
5752: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5753: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5754: if(jj==1){ /* Age has to be done */
1.265 brouard 5755: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5756: 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]));
5757: 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 5758: }
5759: /* printf("%12.7f )", param[i][jj][k]); */
5760: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5761: s1++;
1.250 brouard 5762: }
1.251 brouard 5763: printf("\n");
5764: fprintf(ficlog,"\n");
1.250 brouard 5765: }
5766: }
1.284 brouard 5767: } /* end of state i */
1.251 brouard 5768: printf("#Freqsummary\n");
5769: fprintf(ficlog,"\n");
1.265 brouard 5770: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5771: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5772: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5773: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5774: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5775: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5776: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5777: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5778: /* } */
5779: }
1.265 brouard 5780: } /* end loop s1 */
1.251 brouard 5781:
5782: printf("\n");
5783: fprintf(ficlog,"\n");
5784: } /* end j=0 */
1.249 brouard 5785: } /* end j */
1.252 brouard 5786:
1.253 brouard 5787: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5788: for(i=1, jk=1; i <=nlstate; i++){
5789: for(j=1; j <=nlstate+ndeath; j++){
5790: if(j!=i){
5791: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5792: printf("%1d%1d",i,j);
5793: fprintf(ficparo,"%1d%1d",i,j);
5794: for(k=1; k<=ncovmodel;k++){
5795: /* printf(" %lf",param[i][j][k]); */
5796: /* fprintf(ficparo," %lf",param[i][j][k]); */
5797: p[jk]=pstart[jk];
5798: printf(" %f ",pstart[jk]);
5799: fprintf(ficparo," %f ",pstart[jk]);
5800: jk++;
5801: }
5802: printf("\n");
5803: fprintf(ficparo,"\n");
5804: }
5805: }
5806: }
5807: } /* end mle=-2 */
1.226 brouard 5808: dateintmean=dateintsum/k2cpt;
1.296 brouard 5809: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5810:
1.226 brouard 5811: fclose(ficresp);
5812: fclose(ficresphtm);
5813: fclose(ficresphtmfr);
1.283 brouard 5814: free_vector(idq,1,nqfveff);
1.226 brouard 5815: free_vector(meanq,1,nqfveff);
1.284 brouard 5816: free_vector(stdq,1,nqfveff);
1.226 brouard 5817: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5818: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5819: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5820: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5821: free_vector(pospropt,1,nlstate);
5822: free_vector(posprop,1,nlstate);
1.251 brouard 5823: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5824: free_vector(pp,1,nlstate);
5825: /* End of freqsummary */
5826: }
1.126 brouard 5827:
1.268 brouard 5828: /* Simple linear regression */
5829: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5830:
5831: /* y=a+bx regression */
5832: double sumx = 0.0; /* sum of x */
5833: double sumx2 = 0.0; /* sum of x**2 */
5834: double sumxy = 0.0; /* sum of x * y */
5835: double sumy = 0.0; /* sum of y */
5836: double sumy2 = 0.0; /* sum of y**2 */
5837: double sume2 = 0.0; /* sum of square or residuals */
5838: double yhat;
5839:
5840: double denom=0;
5841: int i;
5842: int ne=*no;
5843:
5844: for ( i=ifi, ne=0;i<=ila;i++) {
5845: if(!isfinite(x[i]) || !isfinite(y[i])){
5846: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5847: continue;
5848: }
5849: ne=ne+1;
5850: sumx += x[i];
5851: sumx2 += x[i]*x[i];
5852: sumxy += x[i] * y[i];
5853: sumy += y[i];
5854: sumy2 += y[i]*y[i];
5855: denom = (ne * sumx2 - sumx*sumx);
5856: /* 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); */
5857: }
5858:
5859: denom = (ne * sumx2 - sumx*sumx);
5860: if (denom == 0) {
5861: // vertical, slope m is infinity
5862: *b = INFINITY;
5863: *a = 0;
5864: if (r) *r = 0;
5865: return 1;
5866: }
5867:
5868: *b = (ne * sumxy - sumx * sumy) / denom;
5869: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5870: if (r!=NULL) {
5871: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5872: sqrt((sumx2 - sumx*sumx/ne) *
5873: (sumy2 - sumy*sumy/ne));
5874: }
5875: *no=ne;
5876: for ( i=ifi, ne=0;i<=ila;i++) {
5877: if(!isfinite(x[i]) || !isfinite(y[i])){
5878: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5879: continue;
5880: }
5881: ne=ne+1;
5882: yhat = y[i] - *a -*b* x[i];
5883: sume2 += yhat * yhat ;
5884:
5885: denom = (ne * sumx2 - sumx*sumx);
5886: /* 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); */
5887: }
5888: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5889: *sa= *sb * sqrt(sumx2/ne);
5890:
5891: return 0;
5892: }
5893:
1.126 brouard 5894: /************ Prevalence ********************/
1.227 brouard 5895: 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)
5896: {
5897: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5898: in each health status at the date of interview (if between dateprev1 and dateprev2).
5899: We still use firstpass and lastpass as another selection.
5900: */
1.126 brouard 5901:
1.227 brouard 5902: int i, m, jk, j1, bool, z1,j, iv;
5903: int mi; /* Effective wave */
5904: int iage;
5905: double agebegin, ageend;
5906:
5907: double **prop;
5908: double posprop;
5909: double y2; /* in fractional years */
5910: int iagemin, iagemax;
5911: int first; /** to stop verbosity which is redirected to log file */
5912:
5913: iagemin= (int) agemin;
5914: iagemax= (int) agemax;
5915: /*pp=vector(1,nlstate);*/
1.251 brouard 5916: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5917: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5918: j1=0;
1.222 brouard 5919:
1.227 brouard 5920: /*j=cptcoveff;*/
5921: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5922:
1.288 brouard 5923: first=0;
1.335 brouard 5924: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of simple dummy covariates */
1.227 brouard 5925: for (i=1; i<=nlstate; i++)
1.251 brouard 5926: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5927: prop[i][iage]=0.0;
5928: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5929: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5930: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5931:
5932: for (i=1; i<=imx; i++) { /* Each individual */
5933: bool=1;
5934: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5935: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5936: m=mw[mi][i];
5937: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5938: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5939: for (z1=1; z1<=cptcoveff; z1++){
5940: if( Fixed[Tmodelind[z1]]==1){
1.341 brouard 5941: iv= Tvar[Tmodelind[z1]];/* because cotvar starts now at first ncovcol+nqv+ (1 to nqtv) */
1.332 brouard 5942: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 5943: bool=0;
5944: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 5945: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 5946: bool=0;
5947: }
5948: }
5949: if(bool==1){ /* Otherwise we skip that wave/person */
5950: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5951: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5952: if(m >=firstpass && m <=lastpass){
5953: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5954: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5955: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5956: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5957: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5958: 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);
5959: exit(1);
5960: }
5961: if (s[m][i]>0 && s[m][i]<=nlstate) {
5962: /*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]]);*/
5963: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5964: prop[s[m][i]][iagemax+3] += weight[i];
5965: } /* end valid statuses */
5966: } /* end selection of dates */
5967: } /* end selection of waves */
5968: } /* end bool */
5969: } /* end wave */
5970: } /* end individual */
5971: for(i=iagemin; i <= iagemax+3; i++){
5972: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5973: posprop += prop[jk][i];
5974: }
5975:
5976: for(jk=1; jk <=nlstate ; jk++){
5977: if( i <= iagemax){
5978: if(posprop>=1.e-5){
5979: probs[i][jk][j1]= prop[jk][i]/posprop;
5980: } else{
1.288 brouard 5981: if(!first){
5982: first=1;
1.266 brouard 5983: 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]);
5984: }else{
1.288 brouard 5985: 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 5986: }
5987: }
5988: }
5989: }/* end jk */
5990: }/* end i */
1.222 brouard 5991: /*} *//* end i1 */
1.227 brouard 5992: } /* end j1 */
1.222 brouard 5993:
1.227 brouard 5994: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5995: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5996: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5997: } /* End of prevalence */
1.126 brouard 5998:
5999: /************* Waves Concatenation ***************/
6000:
6001: 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)
6002: {
1.298 brouard 6003: /* 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 6004: Death is a valid wave (if date is known).
6005: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
6006: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 6007: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 6008: */
1.126 brouard 6009:
1.224 brouard 6010: int i=0, mi=0, m=0, mli=0;
1.126 brouard 6011: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
6012: double sum=0., jmean=0.;*/
1.224 brouard 6013: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 6014: int j, k=0,jk, ju, jl;
6015: double sum=0.;
6016: first=0;
1.214 brouard 6017: firstwo=0;
1.217 brouard 6018: firsthree=0;
1.218 brouard 6019: firstfour=0;
1.164 brouard 6020: jmin=100000;
1.126 brouard 6021: jmax=-1;
6022: jmean=0.;
1.224 brouard 6023:
6024: /* Treating live states */
1.214 brouard 6025: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 6026: mi=0; /* First valid wave */
1.227 brouard 6027: mli=0; /* Last valid wave */
1.309 brouard 6028: m=firstpass; /* Loop on waves */
6029: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 6030: 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 */
6031: mli=m-1;/* mw[++mi][i]=m-1; */
6032: }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 6033: 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 6034: mli=m;
1.224 brouard 6035: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
6036: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 6037: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 6038: }
1.309 brouard 6039: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 6040: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 6041: break;
1.224 brouard 6042: #else
1.317 brouard 6043: 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 6044: if(firsthree == 0){
1.302 brouard 6045: 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 6046: firsthree=1;
1.317 brouard 6047: }else if(firsthree >=1 && firsthree < 10){
6048: 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);
6049: firsthree++;
6050: }else if(firsthree == 10){
6051: printf("Information, too many Information flags: no more reported to log either\n");
6052: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
6053: firsthree++;
6054: }else{
6055: firsthree++;
1.227 brouard 6056: }
1.309 brouard 6057: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 6058: mli=m;
6059: }
6060: if(s[m][i]==-2){ /* Vital status is really unknown */
6061: nbwarn++;
1.309 brouard 6062: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 6063: 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);
6064: 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);
6065: }
6066: break;
6067: }
6068: break;
1.224 brouard 6069: #endif
1.227 brouard 6070: }/* End m >= lastpass */
1.126 brouard 6071: }/* end while */
1.224 brouard 6072:
1.227 brouard 6073: /* 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 6074: /* After last pass */
1.224 brouard 6075: /* Treating death states */
1.214 brouard 6076: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 6077: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
6078: /* } */
1.126 brouard 6079: mi++; /* Death is another wave */
6080: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 6081: /* Only death is a correct wave */
1.126 brouard 6082: mw[mi][i]=m;
1.257 brouard 6083: } /* else not in a death state */
1.224 brouard 6084: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 6085: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 6086: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 6087: 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 6088: nbwarn++;
6089: if(firstfiv==0){
1.309 brouard 6090: 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 6091: firstfiv=1;
6092: }else{
1.309 brouard 6093: 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 6094: }
1.309 brouard 6095: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
6096: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 6097: nberr++;
6098: if(firstwo==0){
1.309 brouard 6099: 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 6100: firstwo=1;
6101: }
1.309 brouard 6102: 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 6103: }
1.257 brouard 6104: }else{ /* if date of interview is unknown */
1.227 brouard 6105: /* death is known but not confirmed by death status at any wave */
6106: if(firstfour==0){
1.309 brouard 6107: 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 6108: firstfour=1;
6109: }
1.309 brouard 6110: 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 6111: }
1.224 brouard 6112: } /* end if date of death is known */
6113: #endif
1.309 brouard 6114: wav[i]=mi; /* mi should be the last effective wave (or mli), */
6115: /* wav[i]=mw[mi][i]; */
1.126 brouard 6116: if(mi==0){
6117: nbwarn++;
6118: if(first==0){
1.227 brouard 6119: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
6120: first=1;
1.126 brouard 6121: }
6122: if(first==1){
1.227 brouard 6123: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 6124: }
6125: } /* end mi==0 */
6126: } /* End individuals */
1.214 brouard 6127: /* wav and mw are no more changed */
1.223 brouard 6128:
1.317 brouard 6129: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
6130: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
6131:
6132:
1.126 brouard 6133: for(i=1; i<=imx; i++){
6134: for(mi=1; mi<wav[i];mi++){
6135: if (stepm <=0)
1.227 brouard 6136: dh[mi][i]=1;
1.126 brouard 6137: else{
1.260 brouard 6138: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 6139: if (agedc[i] < 2*AGESUP) {
6140: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
6141: if(j==0) j=1; /* Survives at least one month after exam */
6142: else if(j<0){
6143: nberr++;
6144: 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]);
6145: j=1; /* Temporary Dangerous patch */
6146: 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);
6147: 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]);
6148: 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);
6149: }
6150: k=k+1;
6151: if (j >= jmax){
6152: jmax=j;
6153: ijmax=i;
6154: }
6155: if (j <= jmin){
6156: jmin=j;
6157: ijmin=i;
6158: }
6159: sum=sum+j;
6160: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
6161: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
6162: }
6163: }
6164: else{
6165: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 6166: /* 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 6167:
1.227 brouard 6168: k=k+1;
6169: if (j >= jmax) {
6170: jmax=j;
6171: ijmax=i;
6172: }
6173: else if (j <= jmin){
6174: jmin=j;
6175: ijmin=i;
6176: }
6177: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
6178: /*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]);*/
6179: if(j<0){
6180: nberr++;
6181: 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]);
6182: 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]);
6183: }
6184: sum=sum+j;
6185: }
6186: jk= j/stepm;
6187: jl= j -jk*stepm;
6188: ju= j -(jk+1)*stepm;
6189: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
6190: if(jl==0){
6191: dh[mi][i]=jk;
6192: bh[mi][i]=0;
6193: }else{ /* We want a negative bias in order to only have interpolation ie
6194: * to avoid the price of an extra matrix product in likelihood */
6195: dh[mi][i]=jk+1;
6196: bh[mi][i]=ju;
6197: }
6198: }else{
6199: if(jl <= -ju){
6200: dh[mi][i]=jk;
6201: bh[mi][i]=jl; /* bias is positive if real duration
6202: * is higher than the multiple of stepm and negative otherwise.
6203: */
6204: }
6205: else{
6206: dh[mi][i]=jk+1;
6207: bh[mi][i]=ju;
6208: }
6209: if(dh[mi][i]==0){
6210: dh[mi][i]=1; /* At least one step */
6211: bh[mi][i]=ju; /* At least one step */
6212: /* 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);*/
6213: }
6214: } /* end if mle */
1.126 brouard 6215: }
6216: } /* end wave */
6217: }
6218: jmean=sum/k;
6219: 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 6220: 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 6221: }
1.126 brouard 6222:
6223: /*********** Tricode ****************************/
1.220 brouard 6224: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 6225: {
6226: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
6227: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
6228: * Boring subroutine which should only output nbcode[Tvar[j]][k]
6229: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
6230: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
6231: */
1.130 brouard 6232:
1.242 brouard 6233: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
6234: int modmaxcovj=0; /* Modality max of covariates j */
6235: int cptcode=0; /* Modality max of covariates j */
6236: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 6237:
6238:
1.242 brouard 6239: /* cptcoveff=0; */
6240: /* *cptcov=0; */
1.126 brouard 6241:
1.242 brouard 6242: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 6243: for (k=1; k <= maxncov; k++)
6244: for(j=1; j<=2; j++)
6245: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 6246:
1.242 brouard 6247: /* Loop on covariates without age and products and no quantitative variable */
1.335 brouard 6248: 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 6249: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
1.343 brouard 6250: /* printf("Testing k=%d, cptcovt=%d\n",k, cptcovt); */
1.339 brouard 6251: if(Dummy[k]==0 && Typevar[k] !=1 && Typevar[k] != 2){ /* Dummy covariate and not age product nor fixed product */
1.242 brouard 6252: switch(Fixed[k]) {
6253: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 6254: modmaxcovj=0;
6255: modmincovj=0;
1.242 brouard 6256: for (i=1; i<=imx; i++) { /* Loop on individuals: reads the data file to get the maximum value of the modality of this covariate Vj*/
1.339 brouard 6257: /* printf("Waiting for error tricode Tvar[%d]=%d i=%d (int)(covar[Tvar[k]][i]=%d\n",k,Tvar[k], i, (int)(covar[Tvar[k]][i])); */
1.242 brouard 6258: ij=(int)(covar[Tvar[k]][i]);
6259: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
6260: * If product of Vn*Vm, still boolean *:
6261: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
6262: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
6263: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
6264: modality of the nth covariate of individual i. */
6265: if (ij > modmaxcovj)
6266: modmaxcovj=ij;
6267: else if (ij < modmincovj)
6268: modmincovj=ij;
1.287 brouard 6269: if (ij <0 || ij >1 ){
1.311 brouard 6270: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6271: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
6272: fflush(ficlog);
6273: exit(1);
1.287 brouard 6274: }
6275: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 6276: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
6277: exit(1);
6278: }else
6279: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
6280: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
6281: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
6282: /* getting the maximum value of the modality of the covariate
6283: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
6284: female ies 1, then modmaxcovj=1.
6285: */
6286: } /* end for loop on individuals i */
6287: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6288: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
6289: cptcode=modmaxcovj;
6290: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
6291: /*for (i=0; i<=cptcode; i++) {*/
6292: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
6293: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6294: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
6295: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
6296: if( j != -1){
6297: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
6298: covariate for which somebody answered excluding
6299: undefined. Usually 2: 0 and 1. */
6300: }
6301: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
6302: covariate for which somebody answered including
6303: undefined. Usually 3: -1, 0 and 1. */
6304: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
6305: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
6306: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 6307:
1.242 brouard 6308: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
6309: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
6310: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
6311: /* modmincovj=3; modmaxcovj = 7; */
6312: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
6313: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
6314: /* defining two dummy variables: variables V1_1 and V1_2.*/
6315: /* nbcode[Tvar[j]][ij]=k; */
6316: /* nbcode[Tvar[j]][1]=0; */
6317: /* nbcode[Tvar[j]][2]=1; */
6318: /* nbcode[Tvar[j]][3]=2; */
6319: /* To be continued (not working yet). */
6320: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 6321:
6322: /* 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*/
6323: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
6324: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
6325: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
6326: /*, could be restored in the future */
6327: 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 6328: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
6329: break;
6330: }
6331: ij++;
1.287 brouard 6332: 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 6333: cptcode = ij; /* New max modality for covar j */
6334: } /* end of loop on modality i=-1 to 1 or more */
6335: break;
6336: case 1: /* Testing on varying covariate, could be simple and
6337: * should look at waves or product of fixed *
6338: * varying. No time to test -1, assuming 0 and 1 only */
6339: ij=0;
6340: for(i=0; i<=1;i++){
6341: nbcode[Tvar[k]][++ij]=i;
6342: }
6343: break;
6344: default:
6345: break;
6346: } /* end switch */
6347: } /* end dummy test */
1.342 brouard 6348: if(Dummy[k]==1 && Typevar[k] !=1 && Fixed ==0){ /* Fixed Quantitative covariate and not age product */
1.311 brouard 6349: 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 6350: if(Tvar[k]<=0 || Tvar[k]>=NCOVMAX){
6351: printf("Error k=%d \n",k);
6352: exit(1);
6353: }
1.311 brouard 6354: if(isnan(covar[Tvar[k]][i])){
6355: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6356: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6357: fflush(ficlog);
6358: exit(1);
6359: }
6360: }
1.335 brouard 6361: } /* end Quanti */
1.287 brouard 6362: } /* 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 6363:
6364: for (k=-1; k< maxncov; k++) Ndum[k]=0;
6365: /* Look at fixed dummy (single or product) covariates to check empty modalities */
6366: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
6367: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
6368: 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 */
6369: 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 */
6370: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
6371: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
6372:
6373: ij=0;
6374: /* 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 6375: 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 */
6376: /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
1.242 brouard 6377: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
6378: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
1.335 brouard 6379: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy simple and non empty in the model */
6380: /* Typevar[k] =0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
6381: /* 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 6382: /* If product not in single variable we don't print results */
6383: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
1.335 brouard 6384: ++ij;/* V5 + V4 + V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V1, *//* V5 quanti, V2 quanti, V4, V3, V1 dummies */
6385: /* k= 1 2 3 4 5 6 7 8 9 */
6386: /* Tvar[k]= 5 4 3 6 5 2 7 1 1 */
6387: /* ij 1 2 3 */
6388: /* Tvaraff[ij]= 4 3 1 */
6389: /* Tmodelind[ij]=2 3 9 */
6390: /* TmodelInvind[ij]=2 1 1 */
1.242 brouard 6391: 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*/
6392: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
6393: 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 */
6394: if(Fixed[k]!=0)
6395: anyvaryingduminmodel=1;
6396: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
6397: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
6398: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
6399: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
6400: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
6401: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
6402: }
6403: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
6404: /* ij--; */
6405: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.335 brouard 6406: *cptcov=ij; /* cptcov= Number of total real effective simple dummies (fixed or time arying) effective (used as cptcoveff in other functions)
1.242 brouard 6407: * because they can be excluded from the model and real
6408: * if in the model but excluded because missing values, but how to get k from ij?*/
6409: for(j=ij+1; j<= cptcovt; j++){
6410: Tvaraff[j]=0;
6411: Tmodelind[j]=0;
6412: }
6413: for(j=ntveff+1; j<= cptcovt; j++){
6414: TmodelInvind[j]=0;
6415: }
6416: /* To be sorted */
6417: ;
6418: }
1.126 brouard 6419:
1.145 brouard 6420:
1.126 brouard 6421: /*********** Health Expectancies ****************/
6422:
1.235 brouard 6423: 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 6424:
6425: {
6426: /* Health expectancies, no variances */
1.329 brouard 6427: /* cij is the combination in the list of combination of dummy covariates */
6428: /* strstart is a string of time at start of computing */
1.164 brouard 6429: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 6430: int nhstepma, nstepma; /* Decreasing with age */
6431: double age, agelim, hf;
6432: double ***p3mat;
6433: double eip;
6434:
1.238 brouard 6435: /* pstamp(ficreseij); */
1.126 brouard 6436: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
6437: fprintf(ficreseij,"# Age");
6438: for(i=1; i<=nlstate;i++){
6439: for(j=1; j<=nlstate;j++){
6440: fprintf(ficreseij," e%1d%1d ",i,j);
6441: }
6442: fprintf(ficreseij," e%1d. ",i);
6443: }
6444: fprintf(ficreseij,"\n");
6445:
6446:
6447: if(estepm < stepm){
6448: printf ("Problem %d lower than %d\n",estepm, stepm);
6449: }
6450: else hstepm=estepm;
6451: /* We compute the life expectancy from trapezoids spaced every estepm months
6452: * This is mainly to measure the difference between two models: for example
6453: * if stepm=24 months pijx are given only every 2 years and by summing them
6454: * we are calculating an estimate of the Life Expectancy assuming a linear
6455: * progression in between and thus overestimating or underestimating according
6456: * to the curvature of the survival function. If, for the same date, we
6457: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6458: * to compare the new estimate of Life expectancy with the same linear
6459: * hypothesis. A more precise result, taking into account a more precise
6460: * curvature will be obtained if estepm is as small as stepm. */
6461:
6462: /* For example we decided to compute the life expectancy with the smallest unit */
6463: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6464: nhstepm is the number of hstepm from age to agelim
6465: nstepm is the number of stepm from age to agelin.
1.270 brouard 6466: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6467: and note for a fixed period like estepm months */
6468: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6469: survival function given by stepm (the optimization length). Unfortunately it
6470: means that if the survival funtion is printed only each two years of age and if
6471: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6472: results. So we changed our mind and took the option of the best precision.
6473: */
6474: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6475:
6476: agelim=AGESUP;
6477: /* If stepm=6 months */
6478: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6479: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6480:
6481: /* nhstepm age range expressed in number of stepm */
6482: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6483: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6484: /* if (stepm >= YEARM) hstepm=1;*/
6485: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6486: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6487:
6488: for (age=bage; age<=fage; age ++){
6489: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6490: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6491: /* if (stepm >= YEARM) hstepm=1;*/
6492: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6493:
6494: /* If stepm=6 months */
6495: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6496: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 6497: /* 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 6498: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6499:
6500: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6501:
6502: printf("%d|",(int)age);fflush(stdout);
6503: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6504:
6505: /* Computing expectancies */
6506: for(i=1; i<=nlstate;i++)
6507: for(j=1; j<=nlstate;j++)
6508: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6509: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6510:
6511: /* 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]);*/
6512:
6513: }
6514:
6515: fprintf(ficreseij,"%3.0f",age );
6516: for(i=1; i<=nlstate;i++){
6517: eip=0;
6518: for(j=1; j<=nlstate;j++){
6519: eip +=eij[i][j][(int)age];
6520: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6521: }
6522: fprintf(ficreseij,"%9.4f", eip );
6523: }
6524: fprintf(ficreseij,"\n");
6525:
6526: }
6527: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6528: printf("\n");
6529: fprintf(ficlog,"\n");
6530:
6531: }
6532:
1.235 brouard 6533: 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 6534:
6535: {
6536: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6537: to initial status i, ei. .
1.126 brouard 6538: */
1.336 brouard 6539: /* Very time consuming function, but already optimized with precov */
1.126 brouard 6540: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6541: int nhstepma, nstepma; /* Decreasing with age */
6542: double age, agelim, hf;
6543: double ***p3matp, ***p3matm, ***varhe;
6544: double **dnewm,**doldm;
6545: double *xp, *xm;
6546: double **gp, **gm;
6547: double ***gradg, ***trgradg;
6548: int theta;
6549:
6550: double eip, vip;
6551:
6552: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6553: xp=vector(1,npar);
6554: xm=vector(1,npar);
6555: dnewm=matrix(1,nlstate*nlstate,1,npar);
6556: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6557:
6558: pstamp(ficresstdeij);
6559: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6560: fprintf(ficresstdeij,"# Age");
6561: for(i=1; i<=nlstate;i++){
6562: for(j=1; j<=nlstate;j++)
6563: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6564: fprintf(ficresstdeij," e%1d. ",i);
6565: }
6566: fprintf(ficresstdeij,"\n");
6567:
6568: pstamp(ficrescveij);
6569: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6570: fprintf(ficrescveij,"# Age");
6571: for(i=1; i<=nlstate;i++)
6572: for(j=1; j<=nlstate;j++){
6573: cptj= (j-1)*nlstate+i;
6574: for(i2=1; i2<=nlstate;i2++)
6575: for(j2=1; j2<=nlstate;j2++){
6576: cptj2= (j2-1)*nlstate+i2;
6577: if(cptj2 <= cptj)
6578: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6579: }
6580: }
6581: fprintf(ficrescveij,"\n");
6582:
6583: if(estepm < stepm){
6584: printf ("Problem %d lower than %d\n",estepm, stepm);
6585: }
6586: else hstepm=estepm;
6587: /* We compute the life expectancy from trapezoids spaced every estepm months
6588: * This is mainly to measure the difference between two models: for example
6589: * if stepm=24 months pijx are given only every 2 years and by summing them
6590: * we are calculating an estimate of the Life Expectancy assuming a linear
6591: * progression in between and thus overestimating or underestimating according
6592: * to the curvature of the survival function. If, for the same date, we
6593: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6594: * to compare the new estimate of Life expectancy with the same linear
6595: * hypothesis. A more precise result, taking into account a more precise
6596: * curvature will be obtained if estepm is as small as stepm. */
6597:
6598: /* For example we decided to compute the life expectancy with the smallest unit */
6599: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6600: nhstepm is the number of hstepm from age to agelim
6601: nstepm is the number of stepm from age to agelin.
6602: Look at hpijx to understand the reason of that which relies in memory size
6603: and note for a fixed period like estepm months */
6604: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6605: survival function given by stepm (the optimization length). Unfortunately it
6606: means that if the survival funtion is printed only each two years of age and if
6607: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6608: results. So we changed our mind and took the option of the best precision.
6609: */
6610: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6611:
6612: /* If stepm=6 months */
6613: /* nhstepm age range expressed in number of stepm */
6614: agelim=AGESUP;
6615: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6616: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6617: /* if (stepm >= YEARM) hstepm=1;*/
6618: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6619:
6620: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6621: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6622: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6623: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6624: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6625: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6626:
6627: for (age=bage; age<=fage; age ++){
6628: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6629: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6630: /* if (stepm >= YEARM) hstepm=1;*/
6631: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6632:
1.126 brouard 6633: /* If stepm=6 months */
6634: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6635: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6636:
6637: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6638:
1.126 brouard 6639: /* Computing Variances of health expectancies */
6640: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6641: decrease memory allocation */
6642: for(theta=1; theta <=npar; theta++){
6643: for(i=1; i<=npar; i++){
1.222 brouard 6644: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6645: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6646: }
1.235 brouard 6647: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6648: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6649:
1.126 brouard 6650: for(j=1; j<= nlstate; j++){
1.222 brouard 6651: for(i=1; i<=nlstate; i++){
6652: for(h=0; h<=nhstepm-1; h++){
6653: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6654: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6655: }
6656: }
1.126 brouard 6657: }
1.218 brouard 6658:
1.126 brouard 6659: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6660: for(h=0; h<=nhstepm-1; h++){
6661: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6662: }
1.126 brouard 6663: }/* End theta */
6664:
6665:
6666: for(h=0; h<=nhstepm-1; h++)
6667: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6668: for(theta=1; theta <=npar; theta++)
6669: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6670:
1.218 brouard 6671:
1.222 brouard 6672: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6673: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6674: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6675:
1.222 brouard 6676: printf("%d|",(int)age);fflush(stdout);
6677: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6678: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6679: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6680: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6681: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6682: for(ij=1;ij<=nlstate*nlstate;ij++)
6683: for(ji=1;ji<=nlstate*nlstate;ji++)
6684: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6685: }
6686: }
1.320 brouard 6687: /* if((int)age ==50){ */
6688: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6689: /* } */
1.126 brouard 6690: /* Computing expectancies */
1.235 brouard 6691: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6692: for(i=1; i<=nlstate;i++)
6693: for(j=1; j<=nlstate;j++)
1.222 brouard 6694: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6695: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6696:
1.222 brouard 6697: /* 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 6698:
1.222 brouard 6699: }
1.269 brouard 6700:
6701: /* Standard deviation of expectancies ij */
1.126 brouard 6702: fprintf(ficresstdeij,"%3.0f",age );
6703: for(i=1; i<=nlstate;i++){
6704: eip=0.;
6705: vip=0.;
6706: for(j=1; j<=nlstate;j++){
1.222 brouard 6707: eip += eij[i][j][(int)age];
6708: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6709: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6710: 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 6711: }
6712: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6713: }
6714: fprintf(ficresstdeij,"\n");
1.218 brouard 6715:
1.269 brouard 6716: /* Variance of expectancies ij */
1.126 brouard 6717: fprintf(ficrescveij,"%3.0f",age );
6718: for(i=1; i<=nlstate;i++)
6719: for(j=1; j<=nlstate;j++){
1.222 brouard 6720: cptj= (j-1)*nlstate+i;
6721: for(i2=1; i2<=nlstate;i2++)
6722: for(j2=1; j2<=nlstate;j2++){
6723: cptj2= (j2-1)*nlstate+i2;
6724: if(cptj2 <= cptj)
6725: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6726: }
1.126 brouard 6727: }
6728: fprintf(ficrescveij,"\n");
1.218 brouard 6729:
1.126 brouard 6730: }
6731: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6732: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6733: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6734: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6735: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6736: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6737: printf("\n");
6738: fprintf(ficlog,"\n");
1.218 brouard 6739:
1.126 brouard 6740: free_vector(xm,1,npar);
6741: free_vector(xp,1,npar);
6742: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6743: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6744: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6745: }
1.218 brouard 6746:
1.126 brouard 6747: /************ Variance ******************/
1.235 brouard 6748: 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 6749: {
1.279 brouard 6750: /** Variance of health expectancies
6751: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6752: * double **newm;
6753: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6754: */
1.218 brouard 6755:
6756: /* int movingaverage(); */
6757: double **dnewm,**doldm;
6758: double **dnewmp,**doldmp;
6759: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6760: int first=0;
1.218 brouard 6761: int k;
6762: double *xp;
1.279 brouard 6763: double **gp, **gm; /**< for var eij */
6764: double ***gradg, ***trgradg; /**< for var eij */
6765: double **gradgp, **trgradgp; /**< for var p point j */
6766: double *gpp, *gmp; /**< for var p point j */
6767: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6768: double ***p3mat;
6769: double age,agelim, hf;
6770: /* double ***mobaverage; */
6771: int theta;
6772: char digit[4];
6773: char digitp[25];
6774:
6775: char fileresprobmorprev[FILENAMELENGTH];
6776:
6777: if(popbased==1){
6778: if(mobilav!=0)
6779: strcpy(digitp,"-POPULBASED-MOBILAV_");
6780: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6781: }
6782: else
6783: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6784:
1.218 brouard 6785: /* if (mobilav!=0) { */
6786: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6787: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6788: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6789: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6790: /* } */
6791: /* } */
6792:
6793: strcpy(fileresprobmorprev,"PRMORPREV-");
6794: sprintf(digit,"%-d",ij);
6795: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6796: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6797: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6798: strcat(fileresprobmorprev,fileresu);
6799: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6800: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6801: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6802: }
6803: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6804: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6805: pstamp(ficresprobmorprev);
6806: 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 6807: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
1.337 brouard 6808:
6809: /* We use TinvDoQresult[nres][resultmodel[nres][j] we sort according to the equation model and the resultline: it is a choice */
6810: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ /\* To be done*\/ */
6811: /* fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
6812: /* } */
6813: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */ /* To be done*/
1.344 ! brouard 6814: /* fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); */
1.337 brouard 6815: fprintf(ficresprobmorprev," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 6816: }
1.337 brouard 6817: /* for(j=1;j<=cptcoveff;j++) */
6818: /* fprintf(ficresprobmorprev," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]); */
1.238 brouard 6819: fprintf(ficresprobmorprev,"\n");
6820:
1.218 brouard 6821: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6822: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6823: fprintf(ficresprobmorprev," p.%-d SE",j);
6824: for(i=1; i<=nlstate;i++)
6825: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6826: }
6827: fprintf(ficresprobmorprev,"\n");
6828:
6829: fprintf(ficgp,"\n# Routine varevsij");
6830: fprintf(ficgp,"\nunset title \n");
6831: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6832: 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");
6833: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6834:
1.218 brouard 6835: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6836: pstamp(ficresvij);
6837: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6838: if(popbased==1)
6839: 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);
6840: else
6841: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6842: fprintf(ficresvij,"# Age");
6843: for(i=1; i<=nlstate;i++)
6844: for(j=1; j<=nlstate;j++)
6845: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6846: fprintf(ficresvij,"\n");
6847:
6848: xp=vector(1,npar);
6849: dnewm=matrix(1,nlstate,1,npar);
6850: doldm=matrix(1,nlstate,1,nlstate);
6851: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6852: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6853:
6854: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6855: gpp=vector(nlstate+1,nlstate+ndeath);
6856: gmp=vector(nlstate+1,nlstate+ndeath);
6857: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6858:
1.218 brouard 6859: if(estepm < stepm){
6860: printf ("Problem %d lower than %d\n",estepm, stepm);
6861: }
6862: else hstepm=estepm;
6863: /* For example we decided to compute the life expectancy with the smallest unit */
6864: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6865: nhstepm is the number of hstepm from age to agelim
6866: nstepm is the number of stepm from age to agelim.
6867: Look at function hpijx to understand why because of memory size limitations,
6868: we decided (b) to get a life expectancy respecting the most precise curvature of the
6869: survival function given by stepm (the optimization length). Unfortunately it
6870: means that if the survival funtion is printed every two years of age and if
6871: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6872: results. So we changed our mind and took the option of the best precision.
6873: */
6874: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6875: agelim = AGESUP;
6876: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6877: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6878: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6879: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6880: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6881: gp=matrix(0,nhstepm,1,nlstate);
6882: gm=matrix(0,nhstepm,1,nlstate);
6883:
6884:
6885: for(theta=1; theta <=npar; theta++){
6886: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6887: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6888: }
1.279 brouard 6889: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6890: * returns into prlim .
1.288 brouard 6891: */
1.242 brouard 6892: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6893:
6894: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6895: if (popbased==1) {
6896: if(mobilav ==0){
6897: for(i=1; i<=nlstate;i++)
6898: prlim[i][i]=probs[(int)age][i][ij];
6899: }else{ /* mobilav */
6900: for(i=1; i<=nlstate;i++)
6901: prlim[i][i]=mobaverage[(int)age][i][ij];
6902: }
6903: }
1.295 brouard 6904: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6905: */
6906: 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 6907: /**< 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 6908: * at horizon h in state j including mortality.
6909: */
1.218 brouard 6910: for(j=1; j<= nlstate; j++){
6911: for(h=0; h<=nhstepm; h++){
6912: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6913: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6914: }
6915: }
1.279 brouard 6916: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6917: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6918: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6919: */
6920: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6921: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6922: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6923: }
6924:
6925: /* Again with minus shift */
1.218 brouard 6926:
6927: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6928: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6929:
1.242 brouard 6930: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6931:
6932: if (popbased==1) {
6933: if(mobilav ==0){
6934: for(i=1; i<=nlstate;i++)
6935: prlim[i][i]=probs[(int)age][i][ij];
6936: }else{ /* mobilav */
6937: for(i=1; i<=nlstate;i++)
6938: prlim[i][i]=mobaverage[(int)age][i][ij];
6939: }
6940: }
6941:
1.235 brouard 6942: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6943:
6944: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6945: for(h=0; h<=nhstepm; h++){
6946: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6947: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6948: }
6949: }
6950: /* This for computing probability of death (h=1 means
6951: computed over hstepm matrices product = hstepm*stepm months)
6952: as a weighted average of prlim.
6953: */
6954: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6955: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6956: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6957: }
1.279 brouard 6958: /* end shifting computations */
6959:
6960: /**< Computing gradient matrix at horizon h
6961: */
1.218 brouard 6962: for(j=1; j<= nlstate; j++) /* vareij */
6963: for(h=0; h<=nhstepm; h++){
6964: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6965: }
1.279 brouard 6966: /**< Gradient of overall mortality p.3 (or p.j)
6967: */
6968: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6969: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6970: }
6971:
6972: } /* End theta */
1.279 brouard 6973:
6974: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6975: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6976:
6977: for(h=0; h<=nhstepm; h++) /* veij */
6978: for(j=1; j<=nlstate;j++)
6979: for(theta=1; theta <=npar; theta++)
6980: trgradg[h][j][theta]=gradg[h][theta][j];
6981:
6982: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6983: for(theta=1; theta <=npar; theta++)
6984: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6985: /**< as well as its transposed matrix
6986: */
1.218 brouard 6987:
6988: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6989: for(i=1;i<=nlstate;i++)
6990: for(j=1;j<=nlstate;j++)
6991: vareij[i][j][(int)age] =0.;
1.279 brouard 6992:
6993: /* Computing trgradg by matcov by gradg at age and summing over h
6994: * and k (nhstepm) formula 15 of article
6995: * Lievre-Brouard-Heathcote
6996: */
6997:
1.218 brouard 6998: for(h=0;h<=nhstepm;h++){
6999: for(k=0;k<=nhstepm;k++){
7000: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
7001: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
7002: for(i=1;i<=nlstate;i++)
7003: for(j=1;j<=nlstate;j++)
7004: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
7005: }
7006: }
7007:
1.279 brouard 7008: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
7009: * p.j overall mortality formula 49 but computed directly because
7010: * we compute the grad (wix pijx) instead of grad (pijx),even if
7011: * wix is independent of theta.
7012: */
1.218 brouard 7013: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
7014: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
7015: for(j=nlstate+1;j<=nlstate+ndeath;j++)
7016: for(i=nlstate+1;i<=nlstate+ndeath;i++)
7017: varppt[j][i]=doldmp[j][i];
7018: /* end ppptj */
7019: /* x centered again */
7020:
1.242 brouard 7021: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 7022:
7023: if (popbased==1) {
7024: if(mobilav ==0){
7025: for(i=1; i<=nlstate;i++)
7026: prlim[i][i]=probs[(int)age][i][ij];
7027: }else{ /* mobilav */
7028: for(i=1; i<=nlstate;i++)
7029: prlim[i][i]=mobaverage[(int)age][i][ij];
7030: }
7031: }
7032:
7033: /* This for computing probability of death (h=1 means
7034: computed over hstepm (estepm) matrices product = hstepm*stepm months)
7035: as a weighted average of prlim.
7036: */
1.235 brouard 7037: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 7038: for(j=nlstate+1;j<=nlstate+ndeath;j++){
7039: for(i=1,gmp[j]=0.;i<= nlstate; i++)
7040: gmp[j] += prlim[i][i]*p3mat[i][j][1];
7041: }
7042: /* end probability of death */
7043:
7044: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
7045: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
7046: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
7047: for(i=1; i<=nlstate;i++){
7048: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
7049: }
7050: }
7051: fprintf(ficresprobmorprev,"\n");
7052:
7053: fprintf(ficresvij,"%.0f ",age );
7054: for(i=1; i<=nlstate;i++)
7055: for(j=1; j<=nlstate;j++){
7056: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
7057: }
7058: fprintf(ficresvij,"\n");
7059: free_matrix(gp,0,nhstepm,1,nlstate);
7060: free_matrix(gm,0,nhstepm,1,nlstate);
7061: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
7062: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
7063: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
7064: } /* End age */
7065: free_vector(gpp,nlstate+1,nlstate+ndeath);
7066: free_vector(gmp,nlstate+1,nlstate+ndeath);
7067: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
7068: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
7069: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
7070: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
7071: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
7072: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
7073: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
7074: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
7075: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
7076: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
7077: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
7078: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
7079: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
7080: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
7081: 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);
7082: /* 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 7083: */
1.218 brouard 7084: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
7085: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 7086:
1.218 brouard 7087: free_vector(xp,1,npar);
7088: free_matrix(doldm,1,nlstate,1,nlstate);
7089: free_matrix(dnewm,1,nlstate,1,npar);
7090: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7091: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
7092: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
7093: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
7094: fclose(ficresprobmorprev);
7095: fflush(ficgp);
7096: fflush(fichtm);
7097: } /* end varevsij */
1.126 brouard 7098:
7099: /************ Variance of prevlim ******************/
1.269 brouard 7100: 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 7101: {
1.205 brouard 7102: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 7103: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 7104:
1.268 brouard 7105: double **dnewmpar,**doldm;
1.126 brouard 7106: int i, j, nhstepm, hstepm;
7107: double *xp;
7108: double *gp, *gm;
7109: double **gradg, **trgradg;
1.208 brouard 7110: double **mgm, **mgp;
1.126 brouard 7111: double age,agelim;
7112: int theta;
7113:
7114: pstamp(ficresvpl);
1.288 brouard 7115: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 7116: fprintf(ficresvpl,"# Age ");
7117: if(nresult >=1)
7118: fprintf(ficresvpl," Result# ");
1.126 brouard 7119: for(i=1; i<=nlstate;i++)
7120: fprintf(ficresvpl," %1d-%1d",i,i);
7121: fprintf(ficresvpl,"\n");
7122:
7123: xp=vector(1,npar);
1.268 brouard 7124: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 7125: doldm=matrix(1,nlstate,1,nlstate);
7126:
7127: hstepm=1*YEARM; /* Every year of age */
7128: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7129: agelim = AGESUP;
7130: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
7131: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7132: if (stepm >= YEARM) hstepm=1;
7133: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7134: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 7135: mgp=matrix(1,npar,1,nlstate);
7136: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 7137: gp=vector(1,nlstate);
7138: gm=vector(1,nlstate);
7139:
7140: for(theta=1; theta <=npar; theta++){
7141: for(i=1; i<=npar; i++){ /* Computes gradient */
7142: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7143: }
1.288 brouard 7144: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7145: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7146: /* else */
7147: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7148: for(i=1;i<=nlstate;i++){
1.126 brouard 7149: gp[i] = prlim[i][i];
1.208 brouard 7150: mgp[theta][i] = prlim[i][i];
7151: }
1.126 brouard 7152: for(i=1; i<=npar; i++) /* Computes gradient */
7153: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 7154: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
7155: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
7156: /* else */
7157: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 7158: for(i=1;i<=nlstate;i++){
1.126 brouard 7159: gm[i] = prlim[i][i];
1.208 brouard 7160: mgm[theta][i] = prlim[i][i];
7161: }
1.126 brouard 7162: for(i=1;i<=nlstate;i++)
7163: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 7164: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 7165: } /* End theta */
7166:
7167: trgradg =matrix(1,nlstate,1,npar);
7168:
7169: for(j=1; j<=nlstate;j++)
7170: for(theta=1; theta <=npar; theta++)
7171: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 7172: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7173: /* printf("\nmgm mgp %d ",(int)age); */
7174: /* for(j=1; j<=nlstate;j++){ */
7175: /* printf(" %d ",j); */
7176: /* for(theta=1; theta <=npar; theta++) */
7177: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7178: /* printf("\n "); */
7179: /* } */
7180: /* } */
7181: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7182: /* printf("\n gradg %d ",(int)age); */
7183: /* for(j=1; j<=nlstate;j++){ */
7184: /* printf("%d ",j); */
7185: /* for(theta=1; theta <=npar; theta++) */
7186: /* printf("%d %lf ",theta,gradg[theta][j]); */
7187: /* printf("\n "); */
7188: /* } */
7189: /* } */
1.126 brouard 7190:
7191: for(i=1;i<=nlstate;i++)
7192: varpl[i][(int)age] =0.;
1.209 brouard 7193: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 7194: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7195: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7196: }else{
1.268 brouard 7197: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7198: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 7199: }
1.126 brouard 7200: for(i=1;i<=nlstate;i++)
7201: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7202:
7203: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 7204: if(nresult >=1)
7205: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 7206: for(i=1; i<=nlstate;i++){
1.126 brouard 7207: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 7208: /* for(j=1;j<=nlstate;j++) */
7209: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
7210: }
1.126 brouard 7211: fprintf(ficresvpl,"\n");
7212: free_vector(gp,1,nlstate);
7213: free_vector(gm,1,nlstate);
1.208 brouard 7214: free_matrix(mgm,1,npar,1,nlstate);
7215: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 7216: free_matrix(gradg,1,npar,1,nlstate);
7217: free_matrix(trgradg,1,nlstate,1,npar);
7218: } /* End age */
7219:
7220: free_vector(xp,1,npar);
7221: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 7222: free_matrix(dnewmpar,1,nlstate,1,nlstate);
7223:
7224: }
7225:
7226:
7227: /************ Variance of backprevalence limit ******************/
1.269 brouard 7228: 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 7229: {
7230: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
7231: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
7232:
7233: double **dnewmpar,**doldm;
7234: int i, j, nhstepm, hstepm;
7235: double *xp;
7236: double *gp, *gm;
7237: double **gradg, **trgradg;
7238: double **mgm, **mgp;
7239: double age,agelim;
7240: int theta;
7241:
7242: pstamp(ficresvbl);
7243: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
7244: fprintf(ficresvbl,"# Age ");
7245: if(nresult >=1)
7246: fprintf(ficresvbl," Result# ");
7247: for(i=1; i<=nlstate;i++)
7248: fprintf(ficresvbl," %1d-%1d",i,i);
7249: fprintf(ficresvbl,"\n");
7250:
7251: xp=vector(1,npar);
7252: dnewmpar=matrix(1,nlstate,1,npar);
7253: doldm=matrix(1,nlstate,1,nlstate);
7254:
7255: hstepm=1*YEARM; /* Every year of age */
7256: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
7257: agelim = AGEINF;
7258: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
7259: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
7260: if (stepm >= YEARM) hstepm=1;
7261: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
7262: gradg=matrix(1,npar,1,nlstate);
7263: mgp=matrix(1,npar,1,nlstate);
7264: mgm=matrix(1,npar,1,nlstate);
7265: gp=vector(1,nlstate);
7266: gm=vector(1,nlstate);
7267:
7268: for(theta=1; theta <=npar; theta++){
7269: for(i=1; i<=npar; i++){ /* Computes gradient */
7270: xp[i] = x[i] + (i==theta ?delti[theta]:0);
7271: }
7272: if(mobilavproj > 0 )
7273: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7274: else
7275: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7276: for(i=1;i<=nlstate;i++){
7277: gp[i] = bprlim[i][i];
7278: mgp[theta][i] = bprlim[i][i];
7279: }
7280: for(i=1; i<=npar; i++) /* Computes gradient */
7281: xp[i] = x[i] - (i==theta ?delti[theta]:0);
7282: if(mobilavproj > 0 )
7283: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7284: else
7285: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
7286: for(i=1;i<=nlstate;i++){
7287: gm[i] = bprlim[i][i];
7288: mgm[theta][i] = bprlim[i][i];
7289: }
7290: for(i=1;i<=nlstate;i++)
7291: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
7292: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
7293: } /* End theta */
7294:
7295: trgradg =matrix(1,nlstate,1,npar);
7296:
7297: for(j=1; j<=nlstate;j++)
7298: for(theta=1; theta <=npar; theta++)
7299: trgradg[j][theta]=gradg[theta][j];
7300: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7301: /* printf("\nmgm mgp %d ",(int)age); */
7302: /* for(j=1; j<=nlstate;j++){ */
7303: /* printf(" %d ",j); */
7304: /* for(theta=1; theta <=npar; theta++) */
7305: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
7306: /* printf("\n "); */
7307: /* } */
7308: /* } */
7309: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
7310: /* printf("\n gradg %d ",(int)age); */
7311: /* for(j=1; j<=nlstate;j++){ */
7312: /* printf("%d ",j); */
7313: /* for(theta=1; theta <=npar; theta++) */
7314: /* printf("%d %lf ",theta,gradg[theta][j]); */
7315: /* printf("\n "); */
7316: /* } */
7317: /* } */
7318:
7319: for(i=1;i<=nlstate;i++)
7320: varbpl[i][(int)age] =0.;
7321: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
7322: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7323: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7324: }else{
7325: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
7326: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
7327: }
7328: for(i=1;i<=nlstate;i++)
7329: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
7330:
7331: fprintf(ficresvbl,"%.0f ",age );
7332: if(nresult >=1)
7333: fprintf(ficresvbl,"%d ",nres );
7334: for(i=1; i<=nlstate;i++)
7335: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
7336: fprintf(ficresvbl,"\n");
7337: free_vector(gp,1,nlstate);
7338: free_vector(gm,1,nlstate);
7339: free_matrix(mgm,1,npar,1,nlstate);
7340: free_matrix(mgp,1,npar,1,nlstate);
7341: free_matrix(gradg,1,npar,1,nlstate);
7342: free_matrix(trgradg,1,nlstate,1,npar);
7343: } /* End age */
7344:
7345: free_vector(xp,1,npar);
7346: free_matrix(doldm,1,nlstate,1,npar);
7347: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 7348:
7349: }
7350:
7351: /************ Variance of one-step probabilities ******************/
7352: 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 7353: {
7354: int i, j=0, k1, l1, tj;
7355: int k2, l2, j1, z1;
7356: int k=0, l;
7357: int first=1, first1, first2;
1.326 brouard 7358: int nres=0; /* New */
1.222 brouard 7359: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
7360: double **dnewm,**doldm;
7361: double *xp;
7362: double *gp, *gm;
7363: double **gradg, **trgradg;
7364: double **mu;
7365: double age, cov[NCOVMAX+1];
7366: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
7367: int theta;
7368: char fileresprob[FILENAMELENGTH];
7369: char fileresprobcov[FILENAMELENGTH];
7370: char fileresprobcor[FILENAMELENGTH];
7371: double ***varpij;
7372:
7373: strcpy(fileresprob,"PROB_");
7374: strcat(fileresprob,fileres);
7375: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
7376: printf("Problem with resultfile: %s\n", fileresprob);
7377: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
7378: }
7379: strcpy(fileresprobcov,"PROBCOV_");
7380: strcat(fileresprobcov,fileresu);
7381: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
7382: printf("Problem with resultfile: %s\n", fileresprobcov);
7383: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
7384: }
7385: strcpy(fileresprobcor,"PROBCOR_");
7386: strcat(fileresprobcor,fileresu);
7387: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
7388: printf("Problem with resultfile: %s\n", fileresprobcor);
7389: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
7390: }
7391: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7392: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7393: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7394: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7395: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7396: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7397: pstamp(ficresprob);
7398: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
7399: fprintf(ficresprob,"# Age");
7400: pstamp(ficresprobcov);
7401: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
7402: fprintf(ficresprobcov,"# Age");
7403: pstamp(ficresprobcor);
7404: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
7405: fprintf(ficresprobcor,"# Age");
1.126 brouard 7406:
7407:
1.222 brouard 7408: for(i=1; i<=nlstate;i++)
7409: for(j=1; j<=(nlstate+ndeath);j++){
7410: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
7411: fprintf(ficresprobcov," p%1d-%1d ",i,j);
7412: fprintf(ficresprobcor," p%1d-%1d ",i,j);
7413: }
7414: /* fprintf(ficresprob,"\n");
7415: fprintf(ficresprobcov,"\n");
7416: fprintf(ficresprobcor,"\n");
7417: */
7418: xp=vector(1,npar);
7419: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7420: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7421: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
7422: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
7423: first=1;
7424: fprintf(ficgp,"\n# Routine varprob");
7425: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
7426: fprintf(fichtm,"\n");
7427:
1.288 brouard 7428: 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 7429: 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);
7430: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 7431: and drawn. It helps understanding how is the covariance between two incidences.\
7432: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 7433: 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 7434: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
7435: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
7436: standard deviations wide on each axis. <br>\
7437: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
7438: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
7439: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
7440:
1.222 brouard 7441: cov[1]=1;
7442: /* tj=cptcoveff; */
1.225 brouard 7443: tj = (int) pow(2,cptcoveff);
1.222 brouard 7444: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
7445: j1=0;
1.332 brouard 7446:
7447: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
7448: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
1.342 brouard 7449: /* 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 7450: if(tj != 1 && TKresult[nres]!= j1)
7451: continue;
7452:
7453: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
7454: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
7455: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 7456: if (cptcovn>0) {
1.334 brouard 7457: fprintf(ficresprob, "\n#********** Variable ");
7458: fprintf(ficresprobcov, "\n#********** Variable ");
7459: fprintf(ficgp, "\n#********** Variable ");
7460: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
7461: fprintf(ficresprobcor, "\n#********** Variable ");
7462:
7463: /* Including quantitative variables of the resultline to be done */
7464: for (z1=1; z1<=cptcovs; z1++){ /* Loop on each variable of this resultline */
1.343 brouard 7465: /* printf("Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model); */
1.338 brouard 7466: fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s \n",nres, z1, modelresult[nres][z1], model);
7467: /* fprintf(ficlog,"Varprob modelresult[%d][%d]=%d model=1+age+%s resultline[%d]=%s \n",nres, z1, modelresult[nres][z1], model, nres, resultline[nres]); */
1.334 brouard 7468: if(Dummy[modelresult[nres][z1]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to z1 in resultline */
7469: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed referenced to model equation */
7470: 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 */
7471: 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 */
7472: 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 */
7473: 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 */
7474: 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 */
7475: fprintf(ficresprob,"fixed ");
7476: fprintf(ficresprobcov,"fixed ");
7477: fprintf(ficgp,"fixed ");
7478: fprintf(fichtmcov,"fixed ");
7479: fprintf(ficresprobcor,"fixed ");
7480: }else{
7481: fprintf(ficresprob,"varyi ");
7482: fprintf(ficresprobcov,"varyi ");
7483: fprintf(ficgp,"varyi ");
7484: fprintf(fichtmcov,"varyi ");
7485: fprintf(ficresprobcor,"varyi ");
7486: }
7487: }else if(Dummy[modelresult[nres][z1]]==1){ /* Quanti variable */
7488: /* For each selected (single) quantitative value */
1.337 brouard 7489: fprintf(ficresprob," V%d=%lg ",Tvqresult[nres][z1],Tqresult[nres][z1]);
1.334 brouard 7490: if(Fixed[modelresult[nres][z1]]==0){ /* Fixed */
7491: fprintf(ficresprob,"fixed ");
7492: fprintf(ficresprobcov,"fixed ");
7493: fprintf(ficgp,"fixed ");
7494: fprintf(fichtmcov,"fixed ");
7495: fprintf(ficresprobcor,"fixed ");
7496: }else{
7497: fprintf(ficresprob,"varyi ");
7498: fprintf(ficresprobcov,"varyi ");
7499: fprintf(ficgp,"varyi ");
7500: fprintf(fichtmcov,"varyi ");
7501: fprintf(ficresprobcor,"varyi ");
7502: }
7503: }else{
7504: 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 */
7505: 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 */
7506: exit(1);
7507: }
7508: } /* End loop on variable of this resultline */
7509: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]); */
1.222 brouard 7510: fprintf(ficresprob, "**********\n#\n");
7511: fprintf(ficresprobcov, "**********\n#\n");
7512: fprintf(ficgp, "**********\n#\n");
7513: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
7514: fprintf(ficresprobcor, "**********\n#");
7515: if(invalidvarcomb[j1]){
7516: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7517: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7518: continue;
7519: }
7520: }
7521: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7522: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7523: gp=vector(1,(nlstate)*(nlstate+ndeath));
7524: gm=vector(1,(nlstate)*(nlstate+ndeath));
1.334 brouard 7525: for (age=bage; age<=fage; age ++){ /* Fo each age we feed the model equation with covariates, using precov as in hpxij() ? */
1.222 brouard 7526: cov[2]=age;
7527: if(nagesqr==1)
7528: cov[3]= age*age;
1.334 brouard 7529: /* New code end of combination but for each resultline */
7530: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
7531: if(Typevar[k1]==1){ /* A product with age */
7532: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.326 brouard 7533: }else{
1.334 brouard 7534: cov[2+nagesqr+k1]=precov[nres][k1];
1.326 brouard 7535: }
1.334 brouard 7536: }/* End of loop on model equation */
7537: /* Old code */
7538: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
7539: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; *\/ */
7540: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
7541: /* /\* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates *\/ */
7542: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])]; */
7543: /* /\*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*\//\* j1 1 2 3 4 */
7544: /* * 1 1 1 1 1 */
7545: /* * 2 2 1 1 1 */
7546: /* * 3 1 2 1 1 */
7547: /* *\/ */
7548: /* /\* nbcode[1][1]=0 nbcode[1][2]=1;*\/ */
7549: /* } */
7550: /* /\* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 *\/ */
7551: /* /\* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] *\/ */
7552: /* /\*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; *\/ */
7553: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
7554: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
7555: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2]; */
7556: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7557: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
7558: /* 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]); */
7559: /* /\* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\\* Using the mean of quantitative variable Tvar[Tage[k]] /\\* Tqresult[nres][k]; *\\/ *\/ */
7560: /* /\* exit(1); *\/ */
7561: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
7562: /* } */
7563: /* /\* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
7564: /* } */
7565: /* for (k=1; k<=cptcovprod;k++){/\* For product without age *\/ */
7566: /* if(Dummy[Tvard[k][1]]==0){ */
7567: /* if(Dummy[Tvard[k][2]]==0){ */
7568: /* 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]])]; */
7569: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7570: /* }else{ /\* Should we use the mean of the quantitative variables? *\/ */
7571: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]]; */
7572: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
7573: /* } */
7574: /* }else{ */
7575: /* if(Dummy[Tvard[k][2]]==0){ */
7576: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]]; */
7577: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
7578: /* }else{ */
7579: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]]; */
7580: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
7581: /* } */
7582: /* } */
7583: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
7584: /* } */
1.326 brouard 7585: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7586: for(theta=1; theta <=npar; theta++){
7587: for(i=1; i<=npar; i++)
7588: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7589:
1.222 brouard 7590: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7591:
1.222 brouard 7592: k=0;
7593: for(i=1; i<= (nlstate); i++){
7594: for(j=1; j<=(nlstate+ndeath);j++){
7595: k=k+1;
7596: gp[k]=pmmij[i][j];
7597: }
7598: }
1.220 brouard 7599:
1.222 brouard 7600: for(i=1; i<=npar; i++)
7601: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7602:
1.222 brouard 7603: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7604: k=0;
7605: for(i=1; i<=(nlstate); i++){
7606: for(j=1; j<=(nlstate+ndeath);j++){
7607: k=k+1;
7608: gm[k]=pmmij[i][j];
7609: }
7610: }
1.220 brouard 7611:
1.222 brouard 7612: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7613: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7614: }
1.126 brouard 7615:
1.222 brouard 7616: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7617: for(theta=1; theta <=npar; theta++)
7618: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7619:
1.222 brouard 7620: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7621: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7622:
1.222 brouard 7623: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7624:
1.222 brouard 7625: k=0;
7626: for(i=1; i<=(nlstate); i++){
7627: for(j=1; j<=(nlstate+ndeath);j++){
7628: k=k+1;
7629: mu[k][(int) age]=pmmij[i][j];
7630: }
7631: }
7632: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7633: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7634: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7635:
1.222 brouard 7636: /*printf("\n%d ",(int)age);
7637: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7638: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7639: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7640: }*/
1.220 brouard 7641:
1.222 brouard 7642: fprintf(ficresprob,"\n%d ",(int)age);
7643: fprintf(ficresprobcov,"\n%d ",(int)age);
7644: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7645:
1.222 brouard 7646: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7647: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7648: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7649: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7650: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7651: }
7652: i=0;
7653: for (k=1; k<=(nlstate);k++){
7654: for (l=1; l<=(nlstate+ndeath);l++){
7655: i++;
7656: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7657: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7658: for (j=1; j<=i;j++){
7659: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7660: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7661: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7662: }
7663: }
7664: }/* end of loop for state */
7665: } /* end of loop for age */
7666: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7667: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7668: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7669: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7670:
7671: /* Confidence intervalle of pij */
7672: /*
7673: fprintf(ficgp,"\nunset parametric;unset label");
7674: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7675: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7676: 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);
7677: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7678: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7679: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7680: */
7681:
7682: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7683: first1=1;first2=2;
7684: for (k2=1; k2<=(nlstate);k2++){
7685: for (l2=1; l2<=(nlstate+ndeath);l2++){
7686: if(l2==k2) continue;
7687: j=(k2-1)*(nlstate+ndeath)+l2;
7688: for (k1=1; k1<=(nlstate);k1++){
7689: for (l1=1; l1<=(nlstate+ndeath);l1++){
7690: if(l1==k1) continue;
7691: i=(k1-1)*(nlstate+ndeath)+l1;
7692: if(i<=j) continue;
7693: for (age=bage; age<=fage; age ++){
7694: if ((int)age %5==0){
7695: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7696: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7697: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7698: mu1=mu[i][(int) age]/stepm*YEARM ;
7699: mu2=mu[j][(int) age]/stepm*YEARM;
7700: c12=cv12/sqrt(v1*v2);
7701: /* Computing eigen value of matrix of covariance */
7702: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7703: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7704: if ((lc2 <0) || (lc1 <0) ){
7705: if(first2==1){
7706: first1=0;
7707: 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);
7708: }
7709: 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);
7710: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7711: /* lc2=fabs(lc2); */
7712: }
1.220 brouard 7713:
1.222 brouard 7714: /* Eigen vectors */
1.280 brouard 7715: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7716: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7717: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7718: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7719: }else
7720: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7721: /*v21=sqrt(1.-v11*v11); *//* error */
7722: v21=(lc1-v1)/cv12*v11;
7723: v12=-v21;
7724: v22=v11;
7725: tnalp=v21/v11;
7726: if(first1==1){
7727: first1=0;
7728: 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);
7729: }
7730: 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);
7731: /*printf(fignu*/
7732: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7733: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7734: if(first==1){
7735: first=0;
7736: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7737: fprintf(ficgp,"\nset parametric;unset label");
7738: 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);
7739: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7740: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7741: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7742: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7743: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7744: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7745: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7746: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7747: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7748: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7749: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7750: 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 7751: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7752: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7753: }else{
7754: first=0;
7755: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7756: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7757: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7758: 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 7759: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7760: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7761: }/* if first */
7762: } /* age mod 5 */
7763: } /* end loop age */
7764: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7765: first=1;
7766: } /*l12 */
7767: } /* k12 */
7768: } /*l1 */
7769: }/* k1 */
1.332 brouard 7770: } /* loop on combination of covariates j1 */
1.326 brouard 7771: } /* loop on nres */
1.222 brouard 7772: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7773: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7774: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7775: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7776: free_vector(xp,1,npar);
7777: fclose(ficresprob);
7778: fclose(ficresprobcov);
7779: fclose(ficresprobcor);
7780: fflush(ficgp);
7781: fflush(fichtmcov);
7782: }
1.126 brouard 7783:
7784:
7785: /******************* Printing html file ***********/
1.201 brouard 7786: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7787: int lastpass, int stepm, int weightopt, char model[],\
7788: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7789: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7790: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7791: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7792: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7793: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7794: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7795: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7796: </ul>");
1.319 brouard 7797: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7798: /* </ul>", model); */
1.214 brouard 7799: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7800: 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",
7801: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 7802: 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 7803: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7804: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7805: fprintf(fichtm,"\
7806: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7807: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7808: fprintf(fichtm,"\
1.217 brouard 7809: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7810: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7811: fprintf(fichtm,"\
1.288 brouard 7812: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7813: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7814: fprintf(fichtm,"\
1.288 brouard 7815: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7816: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7817: fprintf(fichtm,"\
1.211 brouard 7818: - (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 7819: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7820: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7821: if(prevfcast==1){
7822: fprintf(fichtm,"\
7823: - Prevalence projections by age and states: \
1.201 brouard 7824: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7825: }
1.126 brouard 7826:
7827:
1.225 brouard 7828: m=pow(2,cptcoveff);
1.222 brouard 7829: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7830:
1.317 brouard 7831: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7832:
7833: jj1=0;
7834:
7835: fprintf(fichtm," \n<ul>");
1.337 brouard 7836: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7837: /* k1=nres; */
1.338 brouard 7838: k1=TKresult[nres];
7839: if(TKresult[nres]==0)k1=1; /* To be checked for no result */
1.337 brouard 7840: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7841: /* if(m != 1 && TKresult[nres]!= k1) */
7842: /* continue; */
1.264 brouard 7843: jj1++;
7844: if (cptcovn > 0) {
7845: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
1.337 brouard 7846: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
7847: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7848: }
1.337 brouard 7849: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7850: /* fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7851: /* } */
7852: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7853: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7854: /* } */
1.264 brouard 7855: fprintf(fichtm,"\">");
7856:
7857: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7858: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 7859: for (cpt=1; cpt<=cptcovs;cpt++){
7860: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7861: }
1.337 brouard 7862: /* fprintf(fichtm,"************ Results for covariates"); */
7863: /* for (cpt=1; cpt<=cptcoveff;cpt++){ */
7864: /* fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]); */
7865: /* } */
7866: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7867: /* fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7868: /* } */
1.264 brouard 7869: if(invalidvarcomb[k1]){
7870: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7871: continue;
7872: }
7873: fprintf(fichtm,"</a></li>");
7874: } /* cptcovn >0 */
7875: }
1.317 brouard 7876: fprintf(fichtm," \n</ul>");
1.264 brouard 7877:
1.222 brouard 7878: jj1=0;
1.237 brouard 7879:
1.337 brouard 7880: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7881: /* k1=nres; */
1.338 brouard 7882: k1=TKresult[nres];
7883: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 7884: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
7885: /* if(m != 1 && TKresult[nres]!= k1) */
7886: /* continue; */
1.220 brouard 7887:
1.222 brouard 7888: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7889: jj1++;
7890: if (cptcovn > 0) {
1.264 brouard 7891: fprintf(fichtm,"\n<p><a name=\"rescov");
1.337 brouard 7892: for (cpt=1; cpt<=cptcovs;cpt++){
7893: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.264 brouard 7894: }
1.337 brouard 7895: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
7896: /* fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]); */
7897: /* } */
1.264 brouard 7898: fprintf(fichtm,"\"</a>");
7899:
1.222 brouard 7900: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 7901: for (cpt=1; cpt<=cptcovs;cpt++){
7902: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
7903: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 7904: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7905: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7906: }
1.230 brouard 7907: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.338 brouard 7908: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7909: if(invalidvarcomb[k1]){
7910: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7911: printf("\nCombination (%d) ignored because no cases \n",k1);
7912: continue;
7913: }
7914: }
7915: /* aij, bij */
1.259 brouard 7916: 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 7917: <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 7918: /* Pij */
1.241 brouard 7919: 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> \
7920: <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 7921: /* Quasi-incidences */
7922: 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 7923: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7924: 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 7925: 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> \
7926: <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 7927: /* Survival functions (period) in state j */
7928: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7929: 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);
7930: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7931: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7932: }
7933: /* State specific survival functions (period) */
7934: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7935: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7936: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7937: <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);
7938: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7939: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7940: }
1.288 brouard 7941: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7942: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7943: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability for a person being in state (1 to %d) at different ages, to be in state %d some years after. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.338 brouard 7944: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
1.329 brouard 7945: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7946: }
1.296 brouard 7947: if(prevbcast==1){
1.288 brouard 7948: /* Backward prevalence in each health state */
1.222 brouard 7949: for(cpt=1; cpt<=nlstate;cpt++){
1.338 brouard 7950: fprintf(fichtm,"<br>\n- Convergence to mixed (stable) back prevalence in state %d. Or probability for a person to be in state %d at a younger age, knowing that she/he was in state (1 to %d) at different older ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, cpt, nlstate, subdirf2(optionfilefiname,"PB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
7951: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJB_"),subdirf2(optionfilefiname,"PIJB_"));
7952: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 7953: }
1.217 brouard 7954: }
1.222 brouard 7955: if(prevfcast==1){
1.288 brouard 7956: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7957: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7958: 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);
7959: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7960: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7961: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7962: }
7963: }
1.296 brouard 7964: if(prevbcast==1){
1.268 brouard 7965: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7966: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7967: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7968: 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 \
7969: 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 7970: 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);
7971: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7972: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7973: }
7974: }
1.220 brouard 7975:
1.222 brouard 7976: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7977: 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);
7978: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7979: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7980: }
7981: /* } /\* end i1 *\/ */
1.337 brouard 7982: }/* End k1=nres */
1.222 brouard 7983: fprintf(fichtm,"</ul>");
1.126 brouard 7984:
1.222 brouard 7985: fprintf(fichtm,"\
1.126 brouard 7986: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7987: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7988: - 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 7989: But because parameters are usually highly correlated (a higher incidence of disability \
7990: and a higher incidence of recovery can give very close observed transition) it might \
7991: be very useful to look not only at linear confidence intervals estimated from the \
7992: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7993: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7994: covariance matrix of the one-step probabilities. \
7995: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7996:
1.222 brouard 7997: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7998: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7999: fprintf(fichtm,"\
1.126 brouard 8000: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 8001: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 8002:
1.222 brouard 8003: fprintf(fichtm,"\
1.126 brouard 8004: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 8005: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
8006: fprintf(fichtm,"\
1.126 brouard 8007: - 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): \
8008: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 8009: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 8010: fprintf(fichtm,"\
1.126 brouard 8011: - (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): \
8012: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 8013: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 8014: fprintf(fichtm,"\
1.288 brouard 8015: - 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 8016: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
8017: fprintf(fichtm,"\
1.128 brouard 8018: - 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 8019: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
8020: fprintf(fichtm,"\
1.288 brouard 8021: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 8022: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 8023:
8024: /* if(popforecast==1) fprintf(fichtm,"\n */
8025: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
8026: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
8027: /* <br>",fileres,fileres,fileres,fileres); */
8028: /* else */
1.338 brouard 8029: /* fprintf(fichtm,"\n No population forecast: popforecast = %d (instead of 1) or stepm = %d (instead of 1) or model=1+age+%s (instead of .)<br><br></li>\n",popforecast, stepm, model); */
1.222 brouard 8030: fflush(fichtm);
1.126 brouard 8031:
1.225 brouard 8032: m=pow(2,cptcoveff);
1.222 brouard 8033: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 8034:
1.317 brouard 8035: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
8036:
8037: jj1=0;
8038:
8039: fprintf(fichtm," \n<ul>");
1.337 brouard 8040: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8041: /* k1=nres; */
1.338 brouard 8042: k1=TKresult[nres];
1.337 brouard 8043: /* for(k1=1; k1<=m;k1++){ /\* For each combination of covariate *\/ */
8044: /* if(m != 1 && TKresult[nres]!= k1) */
8045: /* continue; */
1.317 brouard 8046: jj1++;
8047: if (cptcovn > 0) {
8048: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
1.337 brouard 8049: for (cpt=1; cpt<=cptcovs;cpt++){
8050: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8051: }
8052: fprintf(fichtm,"\">");
8053:
8054: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
8055: fprintf(fichtm,"************ Results for covariates");
1.337 brouard 8056: for (cpt=1; cpt<=cptcovs;cpt++){
8057: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8058: }
8059: if(invalidvarcomb[k1]){
8060: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
8061: continue;
8062: }
8063: fprintf(fichtm,"</a></li>");
8064: } /* cptcovn >0 */
1.337 brouard 8065: } /* End nres */
1.317 brouard 8066: fprintf(fichtm," \n</ul>");
8067:
1.222 brouard 8068: jj1=0;
1.237 brouard 8069:
1.241 brouard 8070: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8071: /* k1=nres; */
1.338 brouard 8072: k1=TKresult[nres];
8073: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8074: /* for(k1=1; k1<=m;k1++){ */
8075: /* if(m != 1 && TKresult[nres]!= k1) */
8076: /* continue; */
1.222 brouard 8077: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
8078: jj1++;
1.126 brouard 8079: if (cptcovn > 0) {
1.317 brouard 8080: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
1.337 brouard 8081: for (cpt=1; cpt<=cptcovs;cpt++){
8082: fprintf(fichtm,"_V%d=%lg_",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.317 brouard 8083: }
8084: fprintf(fichtm,"\"</a>");
8085:
1.126 brouard 8086: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.337 brouard 8087: for (cpt=1; cpt<=cptcovs;cpt++){ /**< cptcoveff number of variables */
8088: fprintf(fichtm," V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
8089: printf(" V%d=%lg ",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);
1.237 brouard 8090: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 8091: }
1.237 brouard 8092:
1.338 brouard 8093: fprintf(fichtm," (model=1+age+%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 8094:
1.222 brouard 8095: if(invalidvarcomb[k1]){
8096: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
8097: continue;
8098: }
1.337 brouard 8099: } /* If cptcovn >0 */
1.126 brouard 8100: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 8101: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 8102: 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);
8103: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
8104: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 8105: }
8106: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 8107: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 8108: true period expectancies (those weighted with period prevalences are also\
8109: drawn in addition to the population based expectancies computed using\
1.314 brouard 8110: 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);
8111: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
8112: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 8113: /* } /\* end i1 *\/ */
1.241 brouard 8114: }/* End nres */
1.222 brouard 8115: fprintf(fichtm,"</ul>");
8116: fflush(fichtm);
1.126 brouard 8117: }
8118:
8119: /******************* Gnuplot file **************/
1.296 brouard 8120: 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 8121:
8122: char dirfileres[132],optfileres[132];
1.264 brouard 8123: char gplotcondition[132], gplotlabel[132];
1.343 brouard 8124: int cpt=0,k1=0,i=0,k=0,j=0,jk=0,k2=0,k3=0,k4=0,kf=0,kvar=0,kk=0,ipos=0,iposold=0,ij=0, ijp=0, l=0;
1.211 brouard 8125: int lv=0, vlv=0, kl=0;
1.130 brouard 8126: int ng=0;
1.201 brouard 8127: int vpopbased;
1.223 brouard 8128: int ioffset; /* variable offset for columns */
1.270 brouard 8129: int iyearc=1; /* variable column for year of projection */
8130: int iagec=1; /* variable column for age of projection */
1.235 brouard 8131: int nres=0; /* Index of resultline */
1.266 brouard 8132: int istart=1; /* For starting graphs in projections */
1.219 brouard 8133:
1.126 brouard 8134: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
8135: /* printf("Problem with file %s",optionfilegnuplot); */
8136: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
8137: /* } */
8138:
8139: /*#ifdef windows */
8140: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 8141: /*#endif */
1.225 brouard 8142: m=pow(2,cptcoveff);
1.126 brouard 8143:
1.274 brouard 8144: /* diagram of the model */
8145: fprintf(ficgp,"\n#Diagram of the model \n");
8146: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
8147: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
8148: 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);
8149:
1.343 brouard 8150: fprintf(ficgp,"\n#Centripete arrows (turning in other direction (1-i) instead of (i-1)) \nset for [i=1:%d] for [j=1:%d] arrow (%d+1)*10+i from cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.80*(cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) ), -0.80*(sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) + yoff ) ls 4\n",nlstate, nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
1.274 brouard 8151: fprintf(ficgp,"\n#show arrow\nunset label\n");
8152: 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);
8153: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
8154: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
8155: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
8156: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
8157:
1.202 brouard 8158: /* Contribution to likelihood */
8159: /* Plot the probability implied in the likelihood */
1.223 brouard 8160: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
8161: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
8162: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
8163: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 8164: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 8165: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
8166: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 8167: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
8168: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
8169: 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));
8170: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
8171: 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));
8172: for (i=1; i<= nlstate ; i ++) {
8173: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
8174: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
8175: 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);
8176: for (j=2; j<= nlstate+ndeath ; j ++) {
8177: 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);
8178: }
8179: fprintf(ficgp,";\nset out; unset ylabel;\n");
8180: }
8181: /* 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 */
8182: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
8183: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
8184: fprintf(ficgp,"\nset out;unset log\n");
8185: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 8186:
1.343 brouard 8187: /* Plot the probability implied in the likelihood by covariate value */
8188: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
8189: /* if(debugILK==1){ */
8190: for(kf=1; kf <= ncovf; kf++){ /* For each simple dummy covariate of the model */
8191: kvar=Tvar[TvarFind[kf]]; /* variable */
8192: k=18+Tvar[TvarFind[kf]];/*offset because there are 18 columns in the ILK_ file */
8193: for (i=1; i<= nlstate ; i ++) {
8194: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
8195: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
8196: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable \\\n",i,1,k,k,i,1,kvar);
8197: for (j=2; j<= nlstate+ndeath ; j ++) {
8198: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable ",i,j,k,k,i,j,kvar);
8199: }
8200: fprintf(ficgp,";\nset out; unset ylabel;\n");
8201: }
8202: } /* End of each covariate dummy */
8203: for(ncovv=1, iposold=0, kk=0; ncovv <= ncovvt ; ncovv++){
8204: /* Other example V1 + V3 + V5 + age*V1 + age*V3 + age*V5 + V1*V3 + V3*V5 + V1*V5
8205: * kmodel = 1 2 3 4 5 6 7 8 9
8206: * varying 1 2 3 4 5
8207: * ncovv 1 2 3 4 5 6 7 8
8208: * TvarVV[ncovv] V3 5 1 3 3 5 1 5
8209: * TvarVVind[ncovv]=kmodel 2 3 7 7 8 8 9 9
8210: * TvarFind[kmodel] 1 0 0 0 0 0 0 0 0
8211: * kdata ncovcol=[V1 V2] nqv=0 ntv=[V3 V4] nqtv=V5
8212: * Dummy[kmodel] 0 0 1 2 2 3 1 1 1
8213: */
8214: ipos=TvarVVind[ncovv]; /* TvarVVind={2, 5, 5] gives the position in the model of the ncovv th varying covariate */
8215: kvar=TvarVV[ncovv]; /* TvarVV={3, 1, 3} gives the name of each varying covariate */
8216: /* printf("DebugILK ficgp ncovv=%d, kvar=TvarVV[ncovv]=%d, ipos=TvarVVind[ncovv]=%d, Dummy[ipos]=%d, Typevar[ipos]=%d\n", ncovv,kvar,ipos,Dummy[ipos],Typevar[ipos]); */
8217: if(ipos!=iposold){ /* Not a product or first of a product */
8218: /* printf(" %d",ipos); */
8219: /* fprintf(ficresilk," V%d",TvarVV[ncovv]); */
8220: /* printf(" DebugILK ficgp suite ipos=%d != iposold=%d\n", ipos, iposold); */
8221: kk++; /* Position of the ncovv column in ILK_ */
8222: k=18+ncovf+kk; /*offset because there are 18 columns in the ILK_ file plus ncovf fixed covariate */
8223: if(Dummy[ipos]==0 && Typevar[ipos]==0){ /* Only if dummy time varying: Dummy(0, 1=quant singor prod without age,2 dummy*age, 3quant*age) Typevar (0 single, 1=*age,2=Vn*vm) */
8224: for (i=1; i<= nlstate ; i ++) {
8225: fprintf(ficgp,"\nset out \"%s-p%dj-%d.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i,kvar);
8226: fprintf(ficgp,"unset log;\n# For each simple dummy covariate of the model \n plot \"%s\"",subdirf(fileresilk));
8227:
8228: if(gnuplotversion >=5.2){ /* Former gnuplot versions do not have variable pointsize!! */
8229: /* printf("DebugILK gnuplotversion=%g >=5.2\n",gnuplotversion); */
8230: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable \\\n",i,1,k,k,i,1,kvar);
8231: for (j=2; j<= nlstate+ndeath ; j ++) {
8232: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? 7 : 9):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt variable ps 0.4 lc variable ",i,j,k,k,i,j,kvar);
8233: }
8234: }else{
8235: /* printf("DebugILK gnuplotversion=%g <5.2\n",gnuplotversion); */
8236: fprintf(ficgp," u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt 7 ps 0.4 lc variable \\\n",i,1,k,i,1,kvar);
8237: for (j=2; j<= nlstate+ndeath ; j ++) {
8238: fprintf(ficgp,",\\\n \"\" u 2:($5 == %d && $6==%d ? $10 : 1/0):($%d==0 ? $6 : $6+4) t \"p%d%d V%d\" with points pt 7 ps 0.4 lc variable ",i,j,k,i,j,kvar);
8239: }
8240: }
8241: fprintf(ficgp,";\nset out; unset ylabel;\n");
8242: }
8243: }/* End if dummy varying */
8244: }else{ /*Product */
8245: /* printf("*"); */
8246: /* fprintf(ficresilk,"*"); */
8247: }
8248: iposold=ipos;
8249: } /* For each time varying covariate */
8250: /* } /\* debugILK==1 *\/ */
8251: /* 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 */
8252: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
8253: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
8254: fprintf(ficgp,"\nset out;unset log\n");
8255: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
8256:
8257:
8258:
1.126 brouard 8259: strcpy(dirfileres,optionfilefiname);
8260: strcpy(optfileres,"vpl");
1.223 brouard 8261: /* 1eme*/
1.238 brouard 8262: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
1.337 brouard 8263: /* for (k1=1; k1<= m ; k1 ++){ /\* For each valid combination of covariate *\/ */
1.236 brouard 8264: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8265: k1=TKresult[nres];
1.338 brouard 8266: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.238 brouard 8267: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.337 brouard 8268: /* if(m != 1 && TKresult[nres]!= k1) */
8269: /* continue; */
1.238 brouard 8270: /* We are interested in selected combination by the resultline */
1.246 brouard 8271: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 8272: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 8273: strcpy(gplotlabel,"(");
1.337 brouard 8274: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8275: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8276: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8277:
8278: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate k get corresponding value lv for combination k1 *\/ */
8279: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the value of the covariate corresponding to k1 combination *\\/ *\/ */
8280: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8281: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8282: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8283: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8284: /* vlv= nbcode[Tvaraff[k]][lv]; /\* vlv is the value of the covariate lv, 0 or 1 *\/ */
8285: /* /\* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv *\/ */
8286: /* /\* printf(" V%d=%d ",Tvaraff[k],vlv); *\/ */
8287: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8288: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8289: /* } */
8290: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8291: /* /\* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); *\/ */
8292: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8293: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.264 brouard 8294: }
8295: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 8296: /* printf("\n#\n"); */
1.238 brouard 8297: fprintf(ficgp,"\n#\n");
8298: if(invalidvarcomb[k1]){
1.260 brouard 8299: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 8300: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8301: continue;
8302: }
1.235 brouard 8303:
1.241 brouard 8304: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
8305: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 8306: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
1.338 brouard 8307: fprintf(ficgp,"set title \"Alive state %d %s model=1+age+%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 8308: 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);
8309: /* 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); */
8310: /* k1-1 error should be nres-1*/
1.238 brouard 8311: for (i=1; i<= nlstate ; i ++) {
8312: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8313: else fprintf(ficgp," %%*lf (%%*lf)");
8314: }
1.288 brouard 8315: 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 8316: for (i=1; i<= nlstate ; i ++) {
8317: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8318: else fprintf(ficgp," %%*lf (%%*lf)");
8319: }
1.260 brouard 8320: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 8321: for (i=1; i<= nlstate ; i ++) {
8322: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8323: else fprintf(ficgp," %%*lf (%%*lf)");
8324: }
1.265 brouard 8325: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
8326:
8327: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
8328: if(cptcoveff ==0){
1.271 brouard 8329: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 8330: }else{
8331: kl=0;
8332: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8333: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8334: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 8335: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8336: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8337: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8338: vlv= nbcode[Tvaraff[k]][lv];
8339: kl++;
8340: /* 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 *\/ */
8341: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8342: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8343: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8344: if(k==cptcoveff){
8345: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
8346: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
8347: }else{
8348: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
8349: kl++;
8350: }
8351: } /* end covariate */
8352: } /* end if no covariate */
8353:
1.296 brouard 8354: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 8355: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 8356: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 8357: if(cptcoveff ==0){
1.245 brouard 8358: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 8359: }else{
8360: kl=0;
8361: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 8362: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8363: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 8364: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8365: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8366: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8367: /* vlv= nbcode[Tvaraff[k]][lv]; */
8368: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 8369: kl++;
1.238 brouard 8370: /* 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 *\/ */
8371: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8372: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8373: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
8374: if(k==cptcoveff){
1.245 brouard 8375: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 8376: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 8377: }else{
1.332 brouard 8378: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 8379: kl++;
8380: }
8381: } /* end covariate */
8382: } /* end if no covariate */
1.296 brouard 8383: if(prevbcast == 1){
1.268 brouard 8384: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
8385: /* k1-1 error should be nres-1*/
8386: for (i=1; i<= nlstate ; i ++) {
8387: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8388: else fprintf(ficgp," %%*lf (%%*lf)");
8389: }
1.271 brouard 8390: 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 8391: for (i=1; i<= nlstate ; i ++) {
8392: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8393: else fprintf(ficgp," %%*lf (%%*lf)");
8394: }
1.276 brouard 8395: 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 8396: for (i=1; i<= nlstate ; i ++) {
8397: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
8398: else fprintf(ficgp," %%*lf (%%*lf)");
8399: }
1.274 brouard 8400: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 8401: } /* end if backprojcast */
1.296 brouard 8402: } /* end if prevbcast */
1.276 brouard 8403: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
8404: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 8405: } /* nres */
1.337 brouard 8406: /* } /\* k1 *\/ */
1.201 brouard 8407: } /* cpt */
1.235 brouard 8408:
8409:
1.126 brouard 8410: /*2 eme*/
1.337 brouard 8411: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8412: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8413: k1=TKresult[nres];
1.338 brouard 8414: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8415: /* if(m != 1 && TKresult[nres]!= k1) */
8416: /* continue; */
1.238 brouard 8417: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 8418: strcpy(gplotlabel,"(");
1.337 brouard 8419: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8420: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8421: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8422: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8423: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8424: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8425: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8426: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8427: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8428: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8429: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8430: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8431: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8432: /* } */
8433: /* /\* for(k=1; k <= ncovds; k++){ *\/ */
8434: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8435: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8436: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
8437: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 8438: }
1.264 brouard 8439: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8440: fprintf(ficgp,"\n#\n");
1.223 brouard 8441: if(invalidvarcomb[k1]){
8442: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8443: continue;
8444: }
1.219 brouard 8445:
1.241 brouard 8446: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 8447: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 8448: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
8449: if(vpopbased==0){
1.238 brouard 8450: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 8451: }else
1.238 brouard 8452: fprintf(ficgp,"\nreplot ");
8453: for (i=1; i<= nlstate+1 ; i ++) {
8454: k=2*i;
1.261 brouard 8455: 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 8456: for (j=1; j<= nlstate+1 ; j ++) {
8457: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8458: else fprintf(ficgp," %%*lf (%%*lf)");
8459: }
8460: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
8461: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 8462: 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 8463: for (j=1; j<= nlstate+1 ; j ++) {
8464: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8465: else fprintf(ficgp," %%*lf (%%*lf)");
8466: }
8467: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 8468: 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 8469: for (j=1; j<= nlstate+1 ; j ++) {
8470: if (j==i) fprintf(ficgp," %%lf (%%lf)");
8471: else fprintf(ficgp," %%*lf (%%*lf)");
8472: }
8473: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
8474: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
8475: } /* state */
8476: } /* vpopbased */
1.264 brouard 8477: 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 8478: } /* end nres */
1.337 brouard 8479: /* } /\* k1 end 2 eme*\/ */
1.238 brouard 8480:
8481:
8482: /*3eme*/
1.337 brouard 8483: /* for (k1=1; k1<= m ; k1 ++){ */
1.238 brouard 8484: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8485: k1=TKresult[nres];
1.338 brouard 8486: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8487: /* if(m != 1 && TKresult[nres]!= k1) */
8488: /* continue; */
1.238 brouard 8489:
1.332 brouard 8490: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 8491: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 8492: strcpy(gplotlabel,"(");
1.337 brouard 8493: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8494: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8495: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8496: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8497: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8498: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8499: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8500: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8501: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8502: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8503: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8504: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8505: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8506: /* } */
8507: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8508: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8509: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]); */
8510: }
1.264 brouard 8511: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8512: fprintf(ficgp,"\n#\n");
8513: if(invalidvarcomb[k1]){
8514: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8515: continue;
8516: }
8517:
8518: /* k=2+nlstate*(2*cpt-2); */
8519: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 8520: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 8521: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 8522: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 8523: 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 8524: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8525: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8526: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
8527: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8528: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8529: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 8530:
1.238 brouard 8531: */
8532: for (i=1; i< nlstate ; i ++) {
1.261 brouard 8533: 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 8534: /* 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 8535:
1.238 brouard 8536: }
1.261 brouard 8537: 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 8538: }
1.264 brouard 8539: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 8540: } /* end nres */
1.337 brouard 8541: /* } /\* end kl 3eme *\/ */
1.126 brouard 8542:
1.223 brouard 8543: /* 4eme */
1.201 brouard 8544: /* Survival functions (period) from state i in state j by initial state i */
1.337 brouard 8545: /* for (k1=1; k1<=m; k1++){ /\* For each covariate and each value *\/ */
1.238 brouard 8546: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8547: k1=TKresult[nres];
1.338 brouard 8548: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8549: /* if(m != 1 && TKresult[nres]!= k1) */
8550: /* continue; */
1.238 brouard 8551: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 8552: strcpy(gplotlabel,"(");
1.337 brouard 8553: fprintf(ficgp,"\n#\n#\n# Survival functions in state %d : 'LIJ_' files, cov=%d state=%d", cpt, k1, cpt);
8554: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8555: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8556: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8557: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8558: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8559: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8560: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8561: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8562: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8563: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8564: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8565: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8566: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8567: /* } */
8568: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8569: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8570: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8571: }
1.264 brouard 8572: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8573: fprintf(ficgp,"\n#\n");
8574: if(invalidvarcomb[k1]){
8575: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8576: continue;
1.223 brouard 8577: }
1.238 brouard 8578:
1.241 brouard 8579: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 8580: 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 8581: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8582: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8583: k=3;
8584: for (i=1; i<= nlstate ; i ++){
8585: if(i==1){
8586: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8587: }else{
8588: fprintf(ficgp,", '' ");
8589: }
8590: l=(nlstate+ndeath)*(i-1)+1;
8591: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8592: for (j=2; j<= nlstate+ndeath ; j ++)
8593: fprintf(ficgp,"+$%d",k+l+j-1);
8594: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
8595: } /* nlstate */
1.264 brouard 8596: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8597: } /* end cpt state*/
8598: } /* end nres */
1.337 brouard 8599: /* } /\* end covariate k1 *\/ */
1.238 brouard 8600:
1.220 brouard 8601: /* 5eme */
1.201 brouard 8602: /* Survival functions (period) from state i in state j by final state j */
1.337 brouard 8603: /* for (k1=1; k1<= m ; k1++){ /\* For each covariate combination if any *\/ */
1.238 brouard 8604: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8605: k1=TKresult[nres];
1.338 brouard 8606: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8607: /* if(m != 1 && TKresult[nres]!= k1) */
8608: /* continue; */
1.238 brouard 8609: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 8610: strcpy(gplotlabel,"(");
1.238 brouard 8611: 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 8612: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8613: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8614: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8615: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8616: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8617: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8618: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8619: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8620: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8621: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8622: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8623: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8624: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8625: /* } */
8626: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8627: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8628: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.238 brouard 8629: }
1.264 brouard 8630: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8631: fprintf(ficgp,"\n#\n");
8632: if(invalidvarcomb[k1]){
8633: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8634: continue;
8635: }
1.227 brouard 8636:
1.241 brouard 8637: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8638: 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 8639: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8640: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8641: k=3;
8642: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8643: if(j==1)
8644: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8645: else
8646: fprintf(ficgp,", '' ");
8647: l=(nlstate+ndeath)*(cpt-1) +j;
8648: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8649: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8650: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8651: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8652: } /* nlstate */
8653: fprintf(ficgp,", '' ");
8654: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8655: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8656: l=(nlstate+ndeath)*(cpt-1) +j;
8657: if(j < nlstate)
8658: fprintf(ficgp,"$%d +",k+l);
8659: else
8660: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8661: }
1.264 brouard 8662: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8663: } /* end cpt state*/
1.337 brouard 8664: /* } /\* end covariate *\/ */
1.238 brouard 8665: } /* end nres */
1.227 brouard 8666:
1.220 brouard 8667: /* 6eme */
1.202 brouard 8668: /* CV preval stable (period) for each covariate */
1.337 brouard 8669: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8670: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8671: k1=TKresult[nres];
1.338 brouard 8672: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8673: /* if(m != 1 && TKresult[nres]!= k1) */
8674: /* continue; */
1.255 brouard 8675: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8676: strcpy(gplotlabel,"(");
1.288 brouard 8677: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8678: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8679: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8680: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8681: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8682: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8683: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8684: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8685: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8686: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8687: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8688: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8689: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8690: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8691: /* } */
8692: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8693: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8694: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8695: }
1.264 brouard 8696: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8697: fprintf(ficgp,"\n#\n");
1.223 brouard 8698: if(invalidvarcomb[k1]){
1.227 brouard 8699: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8700: continue;
1.223 brouard 8701: }
1.227 brouard 8702:
1.241 brouard 8703: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8704: 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 8705: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8706: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8707: k=3; /* Offset */
1.255 brouard 8708: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8709: if(i==1)
8710: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8711: else
8712: fprintf(ficgp,", '' ");
1.255 brouard 8713: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8714: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8715: for (j=2; j<= nlstate ; j ++)
8716: fprintf(ficgp,"+$%d",k+l+j-1);
8717: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8718: } /* nlstate */
1.264 brouard 8719: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8720: } /* end cpt state*/
8721: } /* end covariate */
1.227 brouard 8722:
8723:
1.220 brouard 8724: /* 7eme */
1.296 brouard 8725: if(prevbcast == 1){
1.288 brouard 8726: /* CV backward prevalence for each covariate */
1.337 brouard 8727: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8728: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8729: k1=TKresult[nres];
1.338 brouard 8730: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8731: /* if(m != 1 && TKresult[nres]!= k1) */
8732: /* continue; */
1.268 brouard 8733: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8734: strcpy(gplotlabel,"(");
1.288 brouard 8735: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8736: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8737: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8738: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8739: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate and each value *\/ */
8740: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate number corresponding to k1 combination *\\/ *\/ */
8741: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8742: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8743: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8744: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8745: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8746: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8747: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8748: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8749: /* } */
8750: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8751: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8752: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8753: }
1.264 brouard 8754: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8755: fprintf(ficgp,"\n#\n");
8756: if(invalidvarcomb[k1]){
8757: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8758: continue;
8759: }
8760:
1.241 brouard 8761: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8762: 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 8763: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8764: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8765: k=3; /* Offset */
1.268 brouard 8766: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8767: if(i==1)
8768: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8769: else
8770: fprintf(ficgp,", '' ");
8771: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8772: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8773: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8774: /* 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 8775: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8776: /* for (j=2; j<= nlstate ; j ++) */
8777: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8778: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8779: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8780: } /* nlstate */
1.264 brouard 8781: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8782: } /* end cpt state*/
8783: } /* end covariate */
1.296 brouard 8784: } /* End if prevbcast */
1.218 brouard 8785:
1.223 brouard 8786: /* 8eme */
1.218 brouard 8787: if(prevfcast==1){
1.288 brouard 8788: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8789:
1.337 brouard 8790: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.237 brouard 8791: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8792: k1=TKresult[nres];
1.338 brouard 8793: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8794: /* if(m != 1 && TKresult[nres]!= k1) */
8795: /* continue; */
1.211 brouard 8796: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8797: strcpy(gplotlabel,"(");
1.288 brouard 8798: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.337 brouard 8799: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8800: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8801: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8802: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8803: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8804: /* lv=codtabm(k1,TnsdVar[Tvaraff[k]]); */
8805: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8806: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8807: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8808: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8809: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8810: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8811: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8812: /* } */
8813: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8814: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8815: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.237 brouard 8816: }
1.264 brouard 8817: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8818: fprintf(ficgp,"\n#\n");
8819: if(invalidvarcomb[k1]){
8820: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8821: continue;
8822: }
8823:
8824: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8825: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8826: 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 8827: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8828: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8829:
8830: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8831: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8832: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8833: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8834: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8835: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8836: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8837: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8838: if(i==istart){
1.227 brouard 8839: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8840: }else{
8841: fprintf(ficgp,",\\\n '' ");
8842: }
8843: if(cptcoveff ==0){ /* No covariate */
8844: ioffset=2; /* Age is in 2 */
8845: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8846: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8847: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8848: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8849: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8850: if(i==nlstate+1){
1.270 brouard 8851: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8852: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8853: fprintf(ficgp,",\\\n '' ");
8854: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8855: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8856: offyear, \
1.268 brouard 8857: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8858: }else
1.227 brouard 8859: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8860: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8861: }else{ /* more than 2 covariates */
1.270 brouard 8862: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8863: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8864: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8865: iyearc=ioffset-1;
8866: iagec=ioffset;
1.227 brouard 8867: fprintf(ficgp," u %d:(",ioffset);
8868: kl=0;
8869: strcpy(gplotcondition,"(");
8870: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8871: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8872: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8873: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8874: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8875: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8876: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8877: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8878: kl++;
8879: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8880: kl++;
8881: if(k <cptcoveff && cptcoveff>1)
8882: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8883: }
8884: strcpy(gplotcondition+strlen(gplotcondition),")");
8885: /* 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 *\/ */
8886: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8887: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8888: /* '' 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*/
8889: if(i==nlstate+1){
1.270 brouard 8890: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8891: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8892: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8893: fprintf(ficgp," u %d:(",iagec);
8894: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8895: iyearc, iagec, offyear, \
8896: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8897: /* '' 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 8898: }else{
8899: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8900: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8901: }
8902: } /* end if covariate */
8903: } /* nlstate */
1.264 brouard 8904: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8905: } /* end cpt state*/
8906: } /* end covariate */
8907: } /* End if prevfcast */
1.227 brouard 8908:
1.296 brouard 8909: if(prevbcast==1){
1.268 brouard 8910: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8911:
1.337 brouard 8912: /* for (k1=1; k1<= m ; k1 ++) /\* For each covariate combination if any *\/ */
1.268 brouard 8913: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 8914: k1=TKresult[nres];
1.338 brouard 8915: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 8916: /* if(m != 1 && TKresult[nres]!= k1) */
8917: /* continue; */
1.268 brouard 8918: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8919: strcpy(gplotlabel,"(");
8920: 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 8921: for (k=1; k<=cptcovs; k++){ /* For each covariate k get corresponding value lv for combination k1 */
8922: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8923: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
8924: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
8925: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
8926: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8927: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
8928: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
8929: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
8930: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
8931: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
8932: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
8933: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
8934: /* } */
8935: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
8936: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
8937: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
1.268 brouard 8938: }
8939: strcpy(gplotlabel+strlen(gplotlabel),")");
8940: fprintf(ficgp,"\n#\n");
8941: if(invalidvarcomb[k1]){
8942: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8943: continue;
8944: }
8945:
8946: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8947: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8948: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8949: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8950: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8951:
8952: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8953: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8954: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8955: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8956: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8957: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8958: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8959: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8960: if(i==istart){
8961: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8962: }else{
8963: fprintf(ficgp,",\\\n '' ");
8964: }
8965: if(cptcoveff ==0){ /* No covariate */
8966: ioffset=2; /* Age is in 2 */
8967: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8968: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8969: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8970: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8971: fprintf(ficgp," u %d:(", ioffset);
8972: if(i==nlstate+1){
1.270 brouard 8973: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8974: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8975: fprintf(ficgp,",\\\n '' ");
8976: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8977: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8978: offbyear, \
8979: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8980: }else
8981: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8982: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8983: }else{ /* more than 2 covariates */
1.270 brouard 8984: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8985: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8986: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8987: iyearc=ioffset-1;
8988: iagec=ioffset;
1.268 brouard 8989: fprintf(ficgp," u %d:(",ioffset);
8990: kl=0;
8991: strcpy(gplotcondition,"(");
1.337 brouard 8992: for (k=1; k<=cptcovs; k++){ /* For each covariate k of the resultline, get corresponding value lv for combination k1 */
1.338 brouard 8993: if(Dummy[modelresult[nres][k]]==0){ /* To be verified */
1.337 brouard 8994: /* for (k=1; k<=cptcoveff; k++){ /\* For each covariate writing the chain of conditions *\/ */
8995: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8996: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8997: lv=Tvresult[nres][k];
8998: vlv=TinvDoQresult[nres][Tvresult[nres][k]];
8999: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
9000: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
9001: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
9002: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
9003: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
9004: kl++;
9005: /* sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]); */
9006: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%lg " ,kl,Tvresult[nres][k], kl+1,TinvDoQresult[nres][Tvresult[nres][k]]);
9007: kl++;
1.338 brouard 9008: if(k <cptcovs && cptcovs>1)
1.337 brouard 9009: sprintf(gplotcondition+strlen(gplotcondition)," && ");
9010: }
1.268 brouard 9011: }
9012: strcpy(gplotcondition+strlen(gplotcondition),")");
9013: /* 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 *\/ */
9014: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
9015: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
9016: /* '' 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*/
9017: if(i==nlstate+1){
1.270 brouard 9018: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
9019: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 9020: fprintf(ficgp,",\\\n '' ");
1.270 brouard 9021: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 9022: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 9023: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
9024: iyearc,iagec,offbyear, \
9025: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 9026: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
9027: }else{
9028: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
9029: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
9030: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
9031: }
9032: } /* end if covariate */
9033: } /* nlstate */
9034: fprintf(ficgp,"\nset out; unset label;\n");
9035: } /* end cpt state*/
9036: } /* end covariate */
1.296 brouard 9037: } /* End if prevbcast */
1.268 brouard 9038:
1.227 brouard 9039:
1.238 brouard 9040: /* 9eme writing MLE parameters */
9041: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 9042: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 9043: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 9044: for(k=1; k <=(nlstate+ndeath); k++){
9045: if (k != i) {
1.227 brouard 9046: fprintf(ficgp,"# current state %d\n",k);
9047: for(j=1; j <=ncovmodel; j++){
9048: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
9049: jk++;
9050: }
9051: fprintf(ficgp,"\n");
1.126 brouard 9052: }
9053: }
1.223 brouard 9054: }
1.187 brouard 9055: fprintf(ficgp,"##############\n#\n");
1.227 brouard 9056:
1.145 brouard 9057: /*goto avoid;*/
1.238 brouard 9058: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
9059: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 9060: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
9061: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
9062: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
9063: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
9064: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9065: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
9066: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9067: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
9068: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
9069: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
9070: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
9071: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
9072: fprintf(ficgp,"#\n");
1.223 brouard 9073: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 9074: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.338 brouard 9075: fprintf(ficgp,"#model=1+age+%s \n",model);
1.238 brouard 9076: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 9077: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
1.337 brouard 9078: /* for(k1=1; k1 <=m; k1++) /\* For each combination of covariate *\/ */
1.237 brouard 9079: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 9080: /* k1=nres; */
1.338 brouard 9081: k1=TKresult[nres];
9082: if(TKresult[nres]==0) k1=1; /* To be checked for noresult */
1.337 brouard 9083: fprintf(ficgp,"\n\n# Resultline k1=%d ",k1);
1.264 brouard 9084: strcpy(gplotlabel,"(");
1.276 brouard 9085: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.337 brouard 9086: for (k=1; k<=cptcovs; k++){ /**< cptcovs number of SIMPLE covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
9087: /* for each resultline nres, and position k, Tvresult[nres][k] gives the name of the variable and
9088: TinvDoQresult[nres][Tvresult[nres][k]] gives its value double or integer) */
9089: fprintf(ficgp," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
9090: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%lg ",Tvresult[nres][k],TinvDoQresult[nres][Tvresult[nres][k]]);
9091: }
9092: /* if(m != 1 && TKresult[nres]!= k1) */
9093: /* continue; */
9094: /* fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1); */
9095: /* strcpy(gplotlabel,"("); */
9096: /* /\*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*\/ */
9097: /* for (k=1; k<=cptcoveff; k++){ /\* For each correspondig covariate value *\/ */
9098: /* /\* lv= decodtabm(k1,k,cptcoveff); /\\* Should be the covariate value corresponding to k1 combination and kth covariate *\\/ *\/ */
9099: /* lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
9100: /* /\* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 *\/ */
9101: /* /\* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 *\/ */
9102: /* /\* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 *\/ */
9103: /* /\* vlv= nbcode[Tvaraff[k]][lv]; *\/ */
9104: /* vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]; */
9105: /* fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv); */
9106: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv); */
9107: /* } */
9108: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
9109: /* fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
9110: /* sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]); */
9111: /* } */
1.264 brouard 9112: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 9113: fprintf(ficgp,"\n#\n");
1.264 brouard 9114: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 9115: fprintf(ficgp,"\nset key outside ");
9116: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
9117: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 9118: fprintf(ficgp,"\nset ter svg size 640, 480 ");
9119: if (ng==1){
9120: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
9121: fprintf(ficgp,"\nunset log y");
9122: }else if (ng==2){
9123: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
9124: fprintf(ficgp,"\nset log y");
9125: }else if (ng==3){
9126: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
9127: fprintf(ficgp,"\nset log y");
9128: }else
9129: fprintf(ficgp,"\nunset title ");
9130: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
9131: i=1;
9132: for(k2=1; k2<=nlstate; k2++) {
9133: k3=i;
9134: for(k=1; k<=(nlstate+ndeath); k++) {
9135: if (k != k2){
9136: switch( ng) {
9137: case 1:
9138: if(nagesqr==0)
9139: fprintf(ficgp," p%d+p%d*x",i,i+1);
9140: else /* nagesqr =1 */
9141: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9142: break;
9143: case 2: /* ng=2 */
9144: if(nagesqr==0)
9145: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
9146: else /* nagesqr =1 */
9147: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
9148: break;
9149: case 3:
9150: if(nagesqr==0)
9151: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
9152: else /* nagesqr =1 */
9153: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
9154: break;
9155: }
9156: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 9157: ijp=1; /* product no age */
9158: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
9159: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 9160: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 9161: switch(Typevar[j]){
9162: case 1:
9163: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9164: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
9165: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
9166: if(DummyV[j]==0){/* Bug valgrind */
9167: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
9168: }else{ /* quantitative */
9169: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9170: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9171: }
9172: ij++;
1.268 brouard 9173: }
1.237 brouard 9174: }
1.329 brouard 9175: }
9176: break;
9177: case 2:
9178: if(cptcovprod >0){
9179: if(j==Tprod[ijp]) { /* */
9180: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9181: if(ijp <=cptcovprod) { /* Product */
9182: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9183: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9184: /* 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)]); */
9185: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9186: }else{ /* Vn is dummy and Vm is quanti */
9187: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9188: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9189: }
9190: }else{ /* Vn*Vm Vn is quanti */
9191: if(DummyV[Tvard[ijp][2]]==0){
9192: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9193: }else{ /* Both quanti */
9194: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9195: }
1.268 brouard 9196: }
1.329 brouard 9197: ijp++;
1.237 brouard 9198: }
1.329 brouard 9199: } /* end Tprod */
9200: }
9201: break;
9202: case 0:
9203: /* simple covariate */
1.264 brouard 9204: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 9205: if(Dummy[j]==0){
9206: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
9207: }else{ /* quantitative */
9208: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 9209: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 9210: }
1.329 brouard 9211: /* end simple */
9212: break;
9213: default:
9214: break;
9215: } /* end switch */
1.237 brouard 9216: } /* end j */
1.329 brouard 9217: }else{ /* k=k2 */
9218: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
9219: fprintf(ficgp," (1.");i=i-ncovmodel;
9220: }else
9221: i=i-ncovmodel;
1.223 brouard 9222: }
1.227 brouard 9223:
1.223 brouard 9224: if(ng != 1){
9225: fprintf(ficgp,")/(1");
1.227 brouard 9226:
1.264 brouard 9227: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 9228: if(nagesqr==0)
1.264 brouard 9229: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 9230: else /* nagesqr =1 */
1.264 brouard 9231: 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 9232:
1.223 brouard 9233: ij=1;
1.329 brouard 9234: ijp=1;
9235: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
9236: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
9237: switch(Typevar[j]){
9238: case 1:
9239: if(cptcovage >0){
9240: if(j==Tage[ij]) { /* Bug valgrind */
9241: if(ij <=cptcovage) { /* Bug valgrind */
9242: if(DummyV[j]==0){/* Bug valgrind */
9243: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
9244: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
9245: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
9246: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
9247: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9248: }else{ /* quantitative */
9249: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9250: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
9251: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
9252: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9253: }
9254: ij++;
9255: }
9256: }
9257: }
9258: break;
9259: case 2:
9260: if(cptcovprod >0){
9261: if(j==Tprod[ijp]) { /* */
9262: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
9263: if(ijp <=cptcovprod) { /* Product */
9264: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
9265: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
9266: /* 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)]); */
9267: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
9268: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
9269: }else{ /* Vn is dummy and Vm is quanti */
9270: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
9271: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9272: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9273: }
9274: }else{ /* Vn*Vm Vn is quanti */
9275: if(DummyV[Tvard[ijp][2]]==0){
9276: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
9277: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
9278: }else{ /* Both quanti */
9279: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
9280: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
9281: }
9282: }
9283: ijp++;
9284: }
9285: } /* end Tprod */
9286: } /* end if */
9287: break;
9288: case 0:
9289: /* simple covariate */
9290: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
9291: if(Dummy[j]==0){
9292: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9293: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
9294: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
9295: }else{ /* quantitative */
9296: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
9297: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
9298: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
9299: }
9300: /* end simple */
9301: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
9302: break;
9303: default:
9304: break;
9305: } /* end switch */
1.223 brouard 9306: }
9307: fprintf(ficgp,")");
9308: }
9309: fprintf(ficgp,")");
9310: if(ng ==2)
1.276 brouard 9311: 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 9312: else /* ng= 3 */
1.276 brouard 9313: 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 9314: }else{ /* end ng <> 1 */
1.223 brouard 9315: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 9316: 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 9317: }
9318: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
9319: fprintf(ficgp,",");
9320: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
9321: fprintf(ficgp,",");
9322: i=i+ncovmodel;
9323: } /* end k */
9324: } /* end k2 */
1.276 brouard 9325: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
9326: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.337 brouard 9327: } /* end resultline */
1.223 brouard 9328: } /* end ng */
9329: /* avoid: */
9330: fflush(ficgp);
1.126 brouard 9331: } /* end gnuplot */
9332:
9333:
9334: /*************** Moving average **************/
1.219 brouard 9335: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 9336: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 9337:
1.222 brouard 9338: int i, cpt, cptcod;
9339: int modcovmax =1;
9340: int mobilavrange, mob;
9341: int iage=0;
1.288 brouard 9342: int firstA1=0, firstA2=0;
1.222 brouard 9343:
1.266 brouard 9344: double sum=0., sumr=0.;
1.222 brouard 9345: double age;
1.266 brouard 9346: double *sumnewp, *sumnewm, *sumnewmr;
9347: double *agemingood, *agemaxgood;
9348: double *agemingoodr, *agemaxgoodr;
1.222 brouard 9349:
9350:
1.278 brouard 9351: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
9352: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 9353:
9354: sumnewp = vector(1,ncovcombmax);
9355: sumnewm = vector(1,ncovcombmax);
1.266 brouard 9356: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 9357: agemingood = vector(1,ncovcombmax);
1.266 brouard 9358: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 9359: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 9360: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 9361:
9362: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 9363: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 9364: sumnewp[cptcod]=0.;
1.266 brouard 9365: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
9366: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 9367: }
9368: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
9369:
1.266 brouard 9370: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
9371: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 9372: else mobilavrange=mobilav;
9373: for (age=bage; age<=fage; age++)
9374: for (i=1; i<=nlstate;i++)
9375: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
9376: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9377: /* We keep the original values on the extreme ages bage, fage and for
9378: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
9379: we use a 5 terms etc. until the borders are no more concerned.
9380: */
9381: for (mob=3;mob <=mobilavrange;mob=mob+2){
9382: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 9383: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
9384: sumnewm[cptcod]=0.;
9385: for (i=1; i<=nlstate;i++){
1.222 brouard 9386: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
9387: for (cpt=1;cpt<=(mob-1)/2;cpt++){
9388: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
9389: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
9390: }
9391: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 9392: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9393: } /* end i */
9394: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
9395: } /* end cptcod */
1.222 brouard 9396: }/* end age */
9397: }/* end mob */
1.266 brouard 9398: }else{
9399: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 9400: return -1;
1.266 brouard 9401: }
9402:
9403: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 9404: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
9405: if(invalidvarcomb[cptcod]){
9406: printf("\nCombination (%d) ignored because no cases \n",cptcod);
9407: continue;
9408: }
1.219 brouard 9409:
1.266 brouard 9410: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
9411: sumnewm[cptcod]=0.;
9412: sumnewmr[cptcod]=0.;
9413: for (i=1; i<=nlstate;i++){
9414: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9415: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9416: }
9417: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9418: agemingoodr[cptcod]=age;
9419: }
9420: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9421: agemingood[cptcod]=age;
9422: }
9423: } /* age */
9424: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 9425: sumnewm[cptcod]=0.;
1.266 brouard 9426: sumnewmr[cptcod]=0.;
1.222 brouard 9427: for (i=1; i<=nlstate;i++){
9428: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9429: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9430: }
9431: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9432: agemaxgoodr[cptcod]=age;
1.222 brouard 9433: }
9434: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 9435: agemaxgood[cptcod]=age;
9436: }
9437: } /* age */
9438: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
9439: /* but they will change */
1.288 brouard 9440: firstA1=0;firstA2=0;
1.266 brouard 9441: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
9442: sumnewm[cptcod]=0.;
9443: sumnewmr[cptcod]=0.;
9444: for (i=1; i<=nlstate;i++){
9445: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9446: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9447: }
9448: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9449: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
9450: agemaxgoodr[cptcod]=age; /* age min */
9451: for (i=1; i<=nlstate;i++)
9452: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9453: }else{ /* bad we change the value with the values of good ages */
9454: for (i=1; i<=nlstate;i++){
9455: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
9456: } /* i */
9457: } /* end bad */
9458: }else{
9459: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9460: agemaxgood[cptcod]=age;
9461: }else{ /* bad we change the value with the values of good ages */
9462: for (i=1; i<=nlstate;i++){
9463: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
9464: } /* i */
9465: } /* end bad */
9466: }/* end else */
9467: sum=0.;sumr=0.;
9468: for (i=1; i<=nlstate;i++){
9469: sum+=mobaverage[(int)age][i][cptcod];
9470: sumr+=probs[(int)age][i][cptcod];
9471: }
9472: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 9473: if(!firstA1){
9474: firstA1=1;
9475: 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);
9476: }
9477: 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 9478: } /* end bad */
9479: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9480: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 9481: if(!firstA2){
9482: firstA2=1;
9483: 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);
9484: }
9485: 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 9486: } /* end bad */
9487: }/* age */
1.266 brouard 9488:
9489: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 9490: sumnewm[cptcod]=0.;
1.266 brouard 9491: sumnewmr[cptcod]=0.;
1.222 brouard 9492: for (i=1; i<=nlstate;i++){
9493: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 9494: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
9495: }
9496: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
9497: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
9498: agemingoodr[cptcod]=age;
9499: for (i=1; i<=nlstate;i++)
9500: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
9501: }else{ /* bad we change the value with the values of good ages */
9502: for (i=1; i<=nlstate;i++){
9503: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
9504: } /* i */
9505: } /* end bad */
9506: }else{
9507: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
9508: agemingood[cptcod]=age;
9509: }else{ /* bad */
9510: for (i=1; i<=nlstate;i++){
9511: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
9512: } /* i */
9513: } /* end bad */
9514: }/* end else */
9515: sum=0.;sumr=0.;
9516: for (i=1; i<=nlstate;i++){
9517: sum+=mobaverage[(int)age][i][cptcod];
9518: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 9519: }
1.266 brouard 9520: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 9521: 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 9522: } /* end bad */
9523: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
9524: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 9525: 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 9526: } /* end bad */
9527: }/* age */
1.266 brouard 9528:
1.222 brouard 9529:
9530: for (age=bage; age<=fage; age++){
1.235 brouard 9531: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 9532: sumnewp[cptcod]=0.;
9533: sumnewm[cptcod]=0.;
9534: for (i=1; i<=nlstate;i++){
9535: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
9536: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
9537: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
9538: }
9539: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
9540: }
9541: /* printf("\n"); */
9542: /* } */
1.266 brouard 9543:
1.222 brouard 9544: /* brutal averaging */
1.266 brouard 9545: /* for (i=1; i<=nlstate;i++){ */
9546: /* for (age=1; age<=bage; age++){ */
9547: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
9548: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9549: /* } */
9550: /* for (age=fage; age<=AGESUP; age++){ */
9551: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
9552: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9553: /* } */
9554: /* } /\* end i status *\/ */
9555: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
9556: /* for (age=1; age<=AGESUP; age++){ */
9557: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
9558: /* mobaverage[(int)age][i][cptcod]=0.; */
9559: /* } */
9560: /* } */
1.222 brouard 9561: }/* end cptcod */
1.266 brouard 9562: free_vector(agemaxgoodr,1, ncovcombmax);
9563: free_vector(agemaxgood,1, ncovcombmax);
9564: free_vector(agemingood,1, ncovcombmax);
9565: free_vector(agemingoodr,1, ncovcombmax);
9566: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 9567: free_vector(sumnewm,1, ncovcombmax);
9568: free_vector(sumnewp,1, ncovcombmax);
9569: return 0;
9570: }/* End movingaverage */
1.218 brouard 9571:
1.126 brouard 9572:
1.296 brouard 9573:
1.126 brouard 9574: /************** Forecasting ******************/
1.296 brouard 9575: /* 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)*/
9576: 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){
9577: /* dateintemean, mean date of interviews
9578: dateprojd, year, month, day of starting projection
9579: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 9580: agemin, agemax range of age
9581: dateprev1 dateprev2 range of dates during which prevalence is computed
9582: */
1.296 brouard 9583: /* double anprojd, mprojd, jprojd; */
9584: /* double anprojf, mprojf, jprojf; */
1.267 brouard 9585: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 9586: double agec; /* generic age */
1.296 brouard 9587: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 9588: double *popeffectif,*popcount;
9589: double ***p3mat;
1.218 brouard 9590: /* double ***mobaverage; */
1.126 brouard 9591: char fileresf[FILENAMELENGTH];
9592:
9593: agelim=AGESUP;
1.211 brouard 9594: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9595: in each health status at the date of interview (if between dateprev1 and dateprev2).
9596: We still use firstpass and lastpass as another selection.
9597: */
1.214 brouard 9598: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9599: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 9600:
1.201 brouard 9601: strcpy(fileresf,"F_");
9602: strcat(fileresf,fileresu);
1.126 brouard 9603: if((ficresf=fopen(fileresf,"w"))==NULL) {
9604: printf("Problem with forecast resultfile: %s\n", fileresf);
9605: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
9606: }
1.235 brouard 9607: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
9608: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 9609:
1.225 brouard 9610: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 9611:
9612:
9613: stepsize=(int) (stepm+YEARM-1)/YEARM;
9614: if (stepm<=12) stepsize=1;
9615: if(estepm < stepm){
9616: printf ("Problem %d lower than %d\n",estepm, stepm);
9617: }
1.270 brouard 9618: else{
9619: hstepm=estepm;
9620: }
9621: if(estepm > stepm){ /* Yes every two year */
9622: stepsize=2;
9623: }
1.296 brouard 9624: hstepm=hstepm/stepm;
1.126 brouard 9625:
1.296 brouard 9626:
9627: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9628: /* fractional in yp1 *\/ */
9629: /* aintmean=yp; */
9630: /* yp2=modf((yp1*12),&yp); */
9631: /* mintmean=yp; */
9632: /* yp1=modf((yp2*30.5),&yp); */
9633: /* jintmean=yp; */
9634: /* if(jintmean==0) jintmean=1; */
9635: /* if(mintmean==0) mintmean=1; */
1.126 brouard 9636:
1.296 brouard 9637:
9638: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
9639: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
9640: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 9641: i1=pow(2,cptcoveff);
1.126 brouard 9642: if (cptcovn < 1){i1=1;}
9643:
1.296 brouard 9644: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 9645:
9646: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 9647:
1.126 brouard 9648: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 9649: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9650: 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 9651: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9652: continue;
1.227 brouard 9653: if(invalidvarcomb[k]){
9654: printf("\nCombination (%d) projection ignored because no cases \n",k);
9655: continue;
9656: }
9657: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
9658: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9659: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); */
9660: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 9661: }
1.235 brouard 9662: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 9663: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 9664: }
1.227 brouard 9665: fprintf(ficresf," yearproj age");
9666: for(j=1; j<=nlstate+ndeath;j++){
9667: for(i=1; i<=nlstate;i++)
9668: fprintf(ficresf," p%d%d",i,j);
9669: fprintf(ficresf," wp.%d",j);
9670: }
1.296 brouard 9671: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 9672: fprintf(ficresf,"\n");
1.296 brouard 9673: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 9674: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
9675: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 9676: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
9677: nhstepm = nhstepm/hstepm;
9678: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9679: oldm=oldms;savm=savms;
1.268 brouard 9680: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 9681: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 9682: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 9683: for (h=0; h<=nhstepm; h++){
9684: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 9685: break;
9686: }
9687: }
9688: fprintf(ficresf,"\n");
9689: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9690: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
9691: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* TnsdVar[Tvaraff] correct */
1.296 brouard 9692: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9693:
9694: for(j=1; j<=nlstate+ndeath;j++) {
9695: ppij=0.;
9696: for(i=1; i<=nlstate;i++) {
1.278 brouard 9697: if (mobilav>=1)
9698: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9699: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9700: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9701: }
1.268 brouard 9702: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9703: } /* end i */
9704: fprintf(ficresf," %.3f", ppij);
9705: }/* end j */
1.227 brouard 9706: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9707: } /* end agec */
1.266 brouard 9708: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9709: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9710: } /* end yearp */
9711: } /* end k */
1.219 brouard 9712:
1.126 brouard 9713: fclose(ficresf);
1.215 brouard 9714: printf("End of Computing forecasting \n");
9715: fprintf(ficlog,"End of Computing forecasting\n");
9716:
1.126 brouard 9717: }
9718:
1.269 brouard 9719: /************** Back Forecasting ******************/
1.296 brouard 9720: /* 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){ */
9721: 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){
9722: /* back1, year, month, day of starting backprojection
1.267 brouard 9723: agemin, agemax range of age
9724: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9725: anback2 year of end of backprojection (same day and month as back1).
9726: prevacurrent and prev are prevalences.
1.267 brouard 9727: */
9728: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9729: double agec; /* generic age */
1.302 brouard 9730: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9731: double *popeffectif,*popcount;
9732: double ***p3mat;
9733: /* double ***mobaverage; */
9734: char fileresfb[FILENAMELENGTH];
9735:
1.268 brouard 9736: agelim=AGEINF;
1.267 brouard 9737: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9738: in each health status at the date of interview (if between dateprev1 and dateprev2).
9739: We still use firstpass and lastpass as another selection.
9740: */
9741: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9742: /* firstpass, lastpass, stepm, weightopt, model); */
9743:
9744: /*Do we need to compute prevalence again?*/
9745:
9746: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9747:
9748: strcpy(fileresfb,"FB_");
9749: strcat(fileresfb,fileresu);
9750: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9751: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9752: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9753: }
9754: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9755: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9756:
9757: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9758:
9759:
9760: stepsize=(int) (stepm+YEARM-1)/YEARM;
9761: if (stepm<=12) stepsize=1;
9762: if(estepm < stepm){
9763: printf ("Problem %d lower than %d\n",estepm, stepm);
9764: }
1.270 brouard 9765: else{
9766: hstepm=estepm;
9767: }
9768: if(estepm >= stepm){ /* Yes every two year */
9769: stepsize=2;
9770: }
1.267 brouard 9771:
9772: hstepm=hstepm/stepm;
1.296 brouard 9773: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9774: /* fractional in yp1 *\/ */
9775: /* aintmean=yp; */
9776: /* yp2=modf((yp1*12),&yp); */
9777: /* mintmean=yp; */
9778: /* yp1=modf((yp2*30.5),&yp); */
9779: /* jintmean=yp; */
9780: /* if(jintmean==0) jintmean=1; */
9781: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9782:
9783: i1=pow(2,cptcoveff);
9784: if (cptcovn < 1){i1=1;}
9785:
1.296 brouard 9786: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9787: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9788:
9789: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9790:
9791: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9792: for(k=1; k<=i1;k++){
9793: if(i1 != 1 && TKresult[nres]!= k)
9794: continue;
9795: if(invalidvarcomb[k]){
9796: printf("\nCombination (%d) projection ignored because no cases \n",k);
9797: continue;
9798: }
1.268 brouard 9799: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9800: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9801: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.267 brouard 9802: }
9803: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9804: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9805: }
9806: fprintf(ficresfb," yearbproj age");
9807: for(j=1; j<=nlstate+ndeath;j++){
9808: for(i=1; i<=nlstate;i++)
1.268 brouard 9809: fprintf(ficresfb," b%d%d",i,j);
9810: fprintf(ficresfb," b.%d",j);
1.267 brouard 9811: }
1.296 brouard 9812: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9813: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9814: fprintf(ficresfb,"\n");
1.296 brouard 9815: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9816: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9817: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9818: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9819: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9820: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9821: nhstepm = nhstepm/hstepm;
9822: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9823: oldm=oldms;savm=savms;
1.268 brouard 9824: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9825: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9826: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9827: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9828: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9829: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9830: for (h=0; h<=nhstepm; h++){
1.268 brouard 9831: if (h*hstepm/YEARM*stepm ==-yearp) {
9832: break;
9833: }
9834: }
9835: fprintf(ficresfb,"\n");
9836: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9837: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.296 brouard 9838: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9839: for(i=1; i<=nlstate+ndeath;i++) {
9840: ppij=0.;ppi=0.;
9841: for(j=1; j<=nlstate;j++) {
9842: /* if (mobilav==1) */
1.269 brouard 9843: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9844: ppi=ppi+prevacurrent[(int)agec][j][k];
9845: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9846: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9847: /* else { */
9848: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9849: /* } */
1.268 brouard 9850: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9851: } /* end j */
9852: if(ppi <0.99){
9853: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9854: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9855: }
9856: fprintf(ficresfb," %.3f", ppij);
9857: }/* end j */
1.267 brouard 9858: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9859: } /* end agec */
9860: } /* end yearp */
9861: } /* end k */
1.217 brouard 9862:
1.267 brouard 9863: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9864:
1.267 brouard 9865: fclose(ficresfb);
9866: printf("End of Computing Back forecasting \n");
9867: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9868:
1.267 brouard 9869: }
1.217 brouard 9870:
1.269 brouard 9871: /* Variance of prevalence limit: varprlim */
9872: 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 9873: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9874:
9875: char fileresvpl[FILENAMELENGTH];
9876: FILE *ficresvpl;
9877: double **oldm, **savm;
9878: double **varpl; /* Variances of prevalence limits by age */
9879: int i1, k, nres, j ;
9880:
9881: strcpy(fileresvpl,"VPL_");
9882: strcat(fileresvpl,fileresu);
9883: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9884: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9885: exit(0);
9886: }
1.288 brouard 9887: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9888: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9889:
9890: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9891: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9892:
9893: i1=pow(2,cptcoveff);
9894: if (cptcovn < 1){i1=1;}
9895:
1.337 brouard 9896: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9897: k=TKresult[nres];
1.338 brouard 9898: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 9899: /* for(k=1; k<=i1;k++){ /\* We find the combination equivalent to result line values of dummies *\/ */
1.269 brouard 9900: if(i1 != 1 && TKresult[nres]!= k)
9901: continue;
9902: fprintf(ficresvpl,"\n#****** ");
9903: printf("\n#****** ");
9904: fprintf(ficlog,"\n#****** ");
1.337 brouard 9905: for(j=1;j<=cptcovs;j++) {
9906: fprintf(ficresvpl,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9907: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9908: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
9909: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9910: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
1.269 brouard 9911: }
1.337 brouard 9912: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
9913: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9914: /* fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9915: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9916: /* } */
1.269 brouard 9917: fprintf(ficresvpl,"******\n");
9918: printf("******\n");
9919: fprintf(ficlog,"******\n");
9920:
9921: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9922: oldm=oldms;savm=savms;
9923: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9924: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9925: /*}*/
9926: }
9927:
9928: fclose(ficresvpl);
1.288 brouard 9929: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9930: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9931:
9932: }
9933: /* Variance of back prevalence: varbprlim */
9934: 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){
9935: /*------- Variance of back (stable) prevalence------*/
9936:
9937: char fileresvbl[FILENAMELENGTH];
9938: FILE *ficresvbl;
9939:
9940: double **oldm, **savm;
9941: double **varbpl; /* Variances of back prevalence limits by age */
9942: int i1, k, nres, j ;
9943:
9944: strcpy(fileresvbl,"VBL_");
9945: strcat(fileresvbl,fileresu);
9946: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9947: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9948: exit(0);
9949: }
9950: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9951: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9952:
9953:
9954: i1=pow(2,cptcoveff);
9955: if (cptcovn < 1){i1=1;}
9956:
1.337 brouard 9957: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
9958: k=TKresult[nres];
1.338 brouard 9959: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 9960: /* for(k=1; k<=i1;k++){ */
9961: /* if(i1 != 1 && TKresult[nres]!= k) */
9962: /* continue; */
1.269 brouard 9963: fprintf(ficresvbl,"\n#****** ");
9964: printf("\n#****** ");
9965: fprintf(ficlog,"\n#****** ");
1.337 brouard 9966: for (j=1; j<= cptcovs; j++){ /* For each selected (single) quantitative value */
1.338 brouard 9967: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
9968: fprintf(ficresvbl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
9969: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][resultmodel[nres][j]]);
1.337 brouard 9970: /* for(j=1;j<=cptcoveff;j++) { */
9971: /* fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9972: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9973: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
9974: /* } */
9975: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
9976: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9977: /* fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
9978: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
1.269 brouard 9979: }
9980: fprintf(ficresvbl,"******\n");
9981: printf("******\n");
9982: fprintf(ficlog,"******\n");
9983:
9984: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9985: oldm=oldms;savm=savms;
9986:
9987: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9988: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9989: /*}*/
9990: }
9991:
9992: fclose(ficresvbl);
9993: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9994: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9995:
9996: } /* End of varbprlim */
9997:
1.126 brouard 9998: /************** Forecasting *****not tested NB*************/
1.227 brouard 9999: /* 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 10000:
1.227 brouard 10001: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
10002: /* int *popage; */
10003: /* double calagedatem, agelim, kk1, kk2; */
10004: /* double *popeffectif,*popcount; */
10005: /* double ***p3mat,***tabpop,***tabpopprev; */
10006: /* /\* double ***mobaverage; *\/ */
10007: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 10008:
1.227 brouard 10009: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10010: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10011: /* agelim=AGESUP; */
10012: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 10013:
1.227 brouard 10014: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 10015:
10016:
1.227 brouard 10017: /* strcpy(filerespop,"POP_"); */
10018: /* strcat(filerespop,fileresu); */
10019: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
10020: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
10021: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
10022: /* } */
10023: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
10024: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 10025:
1.227 brouard 10026: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 10027:
1.227 brouard 10028: /* /\* if (mobilav!=0) { *\/ */
10029: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
10030: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
10031: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
10032: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
10033: /* /\* } *\/ */
10034: /* /\* } *\/ */
1.126 brouard 10035:
1.227 brouard 10036: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
10037: /* if (stepm<=12) stepsize=1; */
1.126 brouard 10038:
1.227 brouard 10039: /* agelim=AGESUP; */
1.126 brouard 10040:
1.227 brouard 10041: /* hstepm=1; */
10042: /* hstepm=hstepm/stepm; */
1.218 brouard 10043:
1.227 brouard 10044: /* if (popforecast==1) { */
10045: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
10046: /* printf("Problem with population file : %s\n",popfile);exit(0); */
10047: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
10048: /* } */
10049: /* popage=ivector(0,AGESUP); */
10050: /* popeffectif=vector(0,AGESUP); */
10051: /* popcount=vector(0,AGESUP); */
1.126 brouard 10052:
1.227 brouard 10053: /* i=1; */
10054: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 10055:
1.227 brouard 10056: /* imx=i; */
10057: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
10058: /* } */
1.218 brouard 10059:
1.227 brouard 10060: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
10061: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
10062: /* k=k+1; */
10063: /* fprintf(ficrespop,"\n#******"); */
10064: /* for(j=1;j<=cptcoveff;j++) { */
10065: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
10066: /* } */
10067: /* fprintf(ficrespop,"******\n"); */
10068: /* fprintf(ficrespop,"# Age"); */
10069: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
10070: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 10071:
1.227 brouard 10072: /* for (cpt=0; cpt<=0;cpt++) { */
10073: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 10074:
1.227 brouard 10075: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
10076: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
10077: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 10078:
1.227 brouard 10079: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10080: /* oldm=oldms;savm=savms; */
10081: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 10082:
1.227 brouard 10083: /* for (h=0; h<=nhstepm; h++){ */
10084: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
10085: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
10086: /* } */
10087: /* for(j=1; j<=nlstate+ndeath;j++) { */
10088: /* kk1=0.;kk2=0; */
10089: /* for(i=1; i<=nlstate;i++) { */
10090: /* if (mobilav==1) */
10091: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
10092: /* else { */
10093: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
10094: /* } */
10095: /* } */
10096: /* if (h==(int)(calagedatem+12*cpt)){ */
10097: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
10098: /* /\*fprintf(ficrespop," %.3f", kk1); */
10099: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
10100: /* } */
10101: /* } */
10102: /* for(i=1; i<=nlstate;i++){ */
10103: /* kk1=0.; */
10104: /* for(j=1; j<=nlstate;j++){ */
10105: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
10106: /* } */
10107: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
10108: /* } */
1.218 brouard 10109:
1.227 brouard 10110: /* if (h==(int)(calagedatem+12*cpt)) */
10111: /* for(j=1; j<=nlstate;j++) */
10112: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
10113: /* } */
10114: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10115: /* } */
10116: /* } */
1.218 brouard 10117:
1.227 brouard 10118: /* /\******\/ */
1.218 brouard 10119:
1.227 brouard 10120: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
10121: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
10122: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
10123: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
10124: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 10125:
1.227 brouard 10126: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10127: /* oldm=oldms;savm=savms; */
10128: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
10129: /* for (h=0; h<=nhstepm; h++){ */
10130: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
10131: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
10132: /* } */
10133: /* for(j=1; j<=nlstate+ndeath;j++) { */
10134: /* kk1=0.;kk2=0; */
10135: /* for(i=1; i<=nlstate;i++) { */
10136: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
10137: /* } */
10138: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
10139: /* } */
10140: /* } */
10141: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
10142: /* } */
10143: /* } */
10144: /* } */
10145: /* } */
1.218 brouard 10146:
1.227 brouard 10147: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 10148:
1.227 brouard 10149: /* if (popforecast==1) { */
10150: /* free_ivector(popage,0,AGESUP); */
10151: /* free_vector(popeffectif,0,AGESUP); */
10152: /* free_vector(popcount,0,AGESUP); */
10153: /* } */
10154: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10155: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
10156: /* fclose(ficrespop); */
10157: /* } /\* End of popforecast *\/ */
1.218 brouard 10158:
1.126 brouard 10159: int fileappend(FILE *fichier, char *optionfich)
10160: {
10161: if((fichier=fopen(optionfich,"a"))==NULL) {
10162: printf("Problem with file: %s\n", optionfich);
10163: fprintf(ficlog,"Problem with file: %s\n", optionfich);
10164: return (0);
10165: }
10166: fflush(fichier);
10167: return (1);
10168: }
10169:
10170:
10171: /**************** function prwizard **********************/
10172: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
10173: {
10174:
10175: /* Wizard to print covariance matrix template */
10176:
1.164 brouard 10177: char ca[32], cb[32];
10178: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 10179: int numlinepar;
10180:
10181: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10182: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
10183: for(i=1; i <=nlstate; i++){
10184: jj=0;
10185: for(j=1; j <=nlstate+ndeath; j++){
10186: if(j==i) continue;
10187: jj++;
10188: /*ca[0]= k+'a'-1;ca[1]='\0';*/
10189: printf("%1d%1d",i,j);
10190: fprintf(ficparo,"%1d%1d",i,j);
10191: for(k=1; k<=ncovmodel;k++){
10192: /* printf(" %lf",param[i][j][k]); */
10193: /* fprintf(ficparo," %lf",param[i][j][k]); */
10194: printf(" 0.");
10195: fprintf(ficparo," 0.");
10196: }
10197: printf("\n");
10198: fprintf(ficparo,"\n");
10199: }
10200: }
10201: printf("# Scales (for hessian or gradient estimation)\n");
10202: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
10203: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
10204: for(i=1; i <=nlstate; i++){
10205: jj=0;
10206: for(j=1; j <=nlstate+ndeath; j++){
10207: if(j==i) continue;
10208: jj++;
10209: fprintf(ficparo,"%1d%1d",i,j);
10210: printf("%1d%1d",i,j);
10211: fflush(stdout);
10212: for(k=1; k<=ncovmodel;k++){
10213: /* printf(" %le",delti3[i][j][k]); */
10214: /* fprintf(ficparo," %le",delti3[i][j][k]); */
10215: printf(" 0.");
10216: fprintf(ficparo," 0.");
10217: }
10218: numlinepar++;
10219: printf("\n");
10220: fprintf(ficparo,"\n");
10221: }
10222: }
10223: printf("# Covariance matrix\n");
10224: /* # 121 Var(a12)\n\ */
10225: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10226: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
10227: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
10228: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
10229: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
10230: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
10231: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10232: fflush(stdout);
10233: fprintf(ficparo,"# Covariance matrix\n");
10234: /* # 121 Var(a12)\n\ */
10235: /* # 122 Cov(b12,a12) Var(b12)\n\ */
10236: /* # ...\n\ */
10237: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
10238:
10239: for(itimes=1;itimes<=2;itimes++){
10240: jj=0;
10241: for(i=1; i <=nlstate; i++){
10242: for(j=1; j <=nlstate+ndeath; j++){
10243: if(j==i) continue;
10244: for(k=1; k<=ncovmodel;k++){
10245: jj++;
10246: ca[0]= k+'a'-1;ca[1]='\0';
10247: if(itimes==1){
10248: printf("#%1d%1d%d",i,j,k);
10249: fprintf(ficparo,"#%1d%1d%d",i,j,k);
10250: }else{
10251: printf("%1d%1d%d",i,j,k);
10252: fprintf(ficparo,"%1d%1d%d",i,j,k);
10253: /* printf(" %.5le",matcov[i][j]); */
10254: }
10255: ll=0;
10256: for(li=1;li <=nlstate; li++){
10257: for(lj=1;lj <=nlstate+ndeath; lj++){
10258: if(lj==li) continue;
10259: for(lk=1;lk<=ncovmodel;lk++){
10260: ll++;
10261: if(ll<=jj){
10262: cb[0]= lk +'a'-1;cb[1]='\0';
10263: if(ll<jj){
10264: if(itimes==1){
10265: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10266: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
10267: }else{
10268: printf(" 0.");
10269: fprintf(ficparo," 0.");
10270: }
10271: }else{
10272: if(itimes==1){
10273: printf(" Var(%s%1d%1d)",ca,i,j);
10274: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
10275: }else{
10276: printf(" 0.");
10277: fprintf(ficparo," 0.");
10278: }
10279: }
10280: }
10281: } /* end lk */
10282: } /* end lj */
10283: } /* end li */
10284: printf("\n");
10285: fprintf(ficparo,"\n");
10286: numlinepar++;
10287: } /* end k*/
10288: } /*end j */
10289: } /* end i */
10290: } /* end itimes */
10291:
10292: } /* end of prwizard */
10293: /******************* Gompertz Likelihood ******************************/
10294: double gompertz(double x[])
10295: {
1.302 brouard 10296: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 10297: int i,n=0; /* n is the size of the sample */
10298:
1.220 brouard 10299: for (i=1;i<=imx ; i++) {
1.126 brouard 10300: sump=sump+weight[i];
10301: /* sump=sump+1;*/
10302: num=num+1;
10303: }
1.302 brouard 10304: L=0.0;
10305: /* agegomp=AGEGOMP; */
1.126 brouard 10306: /* for (i=0; i<=imx; i++)
10307: 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]);*/
10308:
1.302 brouard 10309: for (i=1;i<=imx ; i++) {
10310: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
10311: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
10312: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
10313: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
10314: * +
10315: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
10316: */
10317: if (wav[i] > 1 || agedc[i] < AGESUP) {
10318: if (cens[i] == 1){
10319: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
10320: } else if (cens[i] == 0){
1.126 brouard 10321: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 10322: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
10323: } else
10324: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 10325: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 10326: L=L+A*weight[i];
1.126 brouard 10327: /* 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 10328: }
10329: }
1.126 brouard 10330:
1.302 brouard 10331: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 10332:
10333: return -2*L*num/sump;
10334: }
10335:
1.136 brouard 10336: #ifdef GSL
10337: /******************* Gompertz_f Likelihood ******************************/
10338: double gompertz_f(const gsl_vector *v, void *params)
10339: {
1.302 brouard 10340: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 10341: double *x= (double *) v->data;
10342: int i,n=0; /* n is the size of the sample */
10343:
10344: for (i=0;i<=imx-1 ; i++) {
10345: sump=sump+weight[i];
10346: /* sump=sump+1;*/
10347: num=num+1;
10348: }
10349:
10350:
10351: /* for (i=0; i<=imx; i++)
10352: 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]);*/
10353: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
10354: for (i=1;i<=imx ; i++)
10355: {
10356: if (cens[i] == 1 && wav[i]>1)
10357: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
10358:
10359: if (cens[i] == 0 && wav[i]>1)
10360: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
10361: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
10362:
10363: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
10364: if (wav[i] > 1 ) { /* ??? */
10365: LL=LL+A*weight[i];
10366: /* 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]);*/
10367: }
10368: }
10369:
10370: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
10371: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
10372:
10373: return -2*LL*num/sump;
10374: }
10375: #endif
10376:
1.126 brouard 10377: /******************* Printing html file ***********/
1.201 brouard 10378: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 10379: int lastpass, int stepm, int weightopt, char model[],\
10380: int imx, double p[],double **matcov,double agemortsup){
10381: int i,k;
10382:
10383: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
10384: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
10385: for (i=1;i<=2;i++)
10386: 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 10387: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 10388: fprintf(fichtm,"</ul>");
10389:
10390: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
10391:
10392: 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>");
10393:
10394: for (k=agegomp;k<(agemortsup-2);k++)
10395: 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]);
10396:
10397:
10398: fflush(fichtm);
10399: }
10400:
10401: /******************* Gnuplot file **************/
1.201 brouard 10402: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 10403:
10404: char dirfileres[132],optfileres[132];
1.164 brouard 10405:
1.126 brouard 10406: int ng;
10407:
10408:
10409: /*#ifdef windows */
10410: fprintf(ficgp,"cd \"%s\" \n",pathc);
10411: /*#endif */
10412:
10413:
10414: strcpy(dirfileres,optionfilefiname);
10415: strcpy(optfileres,"vpl");
1.199 brouard 10416: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 10417: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 10418: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 10419: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 10420: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
10421:
10422: }
10423:
1.136 brouard 10424: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
10425: {
1.126 brouard 10426:
1.136 brouard 10427: /*-------- data file ----------*/
10428: FILE *fic;
10429: char dummy[]=" ";
1.240 brouard 10430: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 10431: int lstra;
1.136 brouard 10432: int linei, month, year,iout;
1.302 brouard 10433: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 10434: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 10435: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 10436: char *stratrunc;
1.223 brouard 10437:
1.240 brouard 10438: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
10439: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 10440: for(v=1;v<NCOVMAX;v++){
10441: DummyV[v]=0;
10442: FixedV[v]=0;
10443: }
1.126 brouard 10444:
1.240 brouard 10445: for(v=1; v <=ncovcol;v++){
10446: DummyV[v]=0;
10447: FixedV[v]=0;
10448: }
10449: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
10450: DummyV[v]=1;
10451: FixedV[v]=0;
10452: }
10453: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
10454: DummyV[v]=0;
10455: FixedV[v]=1;
10456: }
10457: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
10458: DummyV[v]=1;
10459: FixedV[v]=1;
10460: }
10461: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
10462: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
10463: 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]);
10464: }
1.339 brouard 10465:
10466: ncovcolt=ncovcol+nqv+ntv+nqtv; /* total of covariates in the data, not in the model equation */
10467:
1.136 brouard 10468: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 10469: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10470: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 10471: }
1.126 brouard 10472:
1.302 brouard 10473: /* Is it a BOM UTF-8 Windows file? */
10474: /* First data line */
10475: linei=0;
10476: while(fgets(line, MAXLINE, fic)) {
10477: noffset=0;
10478: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
10479: {
10480: noffset=noffset+3;
10481: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
10482: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
10483: fflush(ficlog); return 1;
10484: }
10485: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
10486: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
10487: {
10488: noffset=noffset+2;
1.304 brouard 10489: 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);
10490: 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 10491: fflush(ficlog); return 1;
10492: }
10493: else if( line[0] == 0 && line[1] == 0)
10494: {
10495: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
10496: noffset=noffset+4;
1.304 brouard 10497: 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);
10498: 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 10499: fflush(ficlog); return 1;
10500: }
10501: } else{
10502: ;/*printf(" Not a BOM file\n");*/
10503: }
10504: /* If line starts with a # it is a comment */
10505: if (line[noffset] == '#') {
10506: linei=linei+1;
10507: break;
10508: }else{
10509: break;
10510: }
10511: }
10512: fclose(fic);
10513: if((fic=fopen(datafile,"r"))==NULL) {
10514: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
10515: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
10516: }
10517: /* Not a Bom file */
10518:
1.136 brouard 10519: i=1;
10520: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
10521: linei=linei+1;
10522: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
10523: if(line[j] == '\t')
10524: line[j] = ' ';
10525: }
10526: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
10527: ;
10528: };
10529: line[j+1]=0; /* Trims blanks at end of line */
10530: if(line[0]=='#'){
10531: fprintf(ficlog,"Comment line\n%s\n",line);
10532: printf("Comment line\n%s\n",line);
10533: continue;
10534: }
10535: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 10536: strcpy(line, linetmp);
1.223 brouard 10537:
10538: /* Loops on waves */
10539: for (j=maxwav;j>=1;j--){
10540: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 10541: cutv(stra, strb, line, ' ');
10542: if(strb[0]=='.') { /* Missing value */
10543: lval=-1;
10544: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
1.341 brouard 10545: cotvar[j][ncovcol+nqv+ntv+iv][i]=-1; /* For performance reasons */
1.238 brouard 10546: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
10547: 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);
10548: 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);
10549: return 1;
10550: }
10551: }else{
10552: errno=0;
10553: /* what_kind_of_number(strb); */
10554: dval=strtod(strb,&endptr);
10555: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
10556: /* if(strb != endptr && *endptr == '\0') */
10557: /* dval=dlval; */
10558: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10559: if( strb[0]=='\0' || (*endptr != '\0')){
10560: 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);
10561: 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);
10562: return 1;
10563: }
10564: cotqvar[j][iv][i]=dval;
1.341 brouard 10565: cotvar[j][ncovcol+nqv+ntv+iv][i]=dval; /* because cotvar starts now at first ntv */
1.238 brouard 10566: }
10567: strcpy(line,stra);
1.223 brouard 10568: }/* end loop ntqv */
1.225 brouard 10569:
1.223 brouard 10570: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 10571: cutv(stra, strb, line, ' ');
10572: if(strb[0]=='.') { /* Missing value */
10573: lval=-1;
10574: }else{
10575: errno=0;
10576: lval=strtol(strb,&endptr,10);
10577: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10578: if( strb[0]=='\0' || (*endptr != '\0')){
10579: 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);
10580: 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);
10581: return 1;
10582: }
10583: }
10584: if(lval <-1 || lval >1){
10585: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10586: 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 10587: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10588: For example, for multinomial values like 1, 2 and 3,\n \
10589: build V1=0 V2=0 for the reference value (1),\n \
10590: V1=1 V2=0 for (2) \n \
1.223 brouard 10591: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10592: output of IMaCh is often meaningless.\n \
1.319 brouard 10593: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 10594: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10595: 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 10596: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10597: For example, for multinomial values like 1, 2 and 3,\n \
10598: build V1=0 V2=0 for the reference value (1),\n \
10599: V1=1 V2=0 for (2) \n \
1.223 brouard 10600: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10601: output of IMaCh is often meaningless.\n \
1.319 brouard 10602: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 10603: return 1;
10604: }
1.341 brouard 10605: cotvar[j][ncovcol+nqv+iv][i]=(double)(lval);
1.238 brouard 10606: strcpy(line,stra);
1.223 brouard 10607: }/* end loop ntv */
1.225 brouard 10608:
1.223 brouard 10609: /* Statuses at wave */
1.137 brouard 10610: cutv(stra, strb, line, ' ');
1.223 brouard 10611: if(strb[0]=='.') { /* Missing value */
1.238 brouard 10612: lval=-1;
1.136 brouard 10613: }else{
1.238 brouard 10614: errno=0;
10615: lval=strtol(strb,&endptr,10);
10616: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10617: if( strb[0]=='\0' || (*endptr != '\0')){
10618: 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);
10619: 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);
10620: return 1;
10621: }
1.136 brouard 10622: }
1.225 brouard 10623:
1.136 brouard 10624: s[j][i]=lval;
1.225 brouard 10625:
1.223 brouard 10626: /* Date of Interview */
1.136 brouard 10627: strcpy(line,stra);
10628: cutv(stra, strb,line,' ');
1.169 brouard 10629: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10630: }
1.169 brouard 10631: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 10632: month=99;
10633: year=9999;
1.136 brouard 10634: }else{
1.225 brouard 10635: 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);
10636: 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);
10637: return 1;
1.136 brouard 10638: }
10639: anint[j][i]= (double) year;
1.302 brouard 10640: mint[j][i]= (double)month;
10641: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
10642: /* 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]); */
10643: /* 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]); */
10644: /* } */
1.136 brouard 10645: strcpy(line,stra);
1.223 brouard 10646: } /* End loop on waves */
1.225 brouard 10647:
1.223 brouard 10648: /* Date of death */
1.136 brouard 10649: cutv(stra, strb,line,' ');
1.169 brouard 10650: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10651: }
1.169 brouard 10652: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 10653: month=99;
10654: year=9999;
10655: }else{
1.141 brouard 10656: 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 10657: 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);
10658: return 1;
1.136 brouard 10659: }
10660: andc[i]=(double) year;
10661: moisdc[i]=(double) month;
10662: strcpy(line,stra);
10663:
1.223 brouard 10664: /* Date of birth */
1.136 brouard 10665: cutv(stra, strb,line,' ');
1.169 brouard 10666: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10667: }
1.169 brouard 10668: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 10669: month=99;
10670: year=9999;
10671: }else{
1.141 brouard 10672: 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);
10673: 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 10674: return 1;
1.136 brouard 10675: }
10676: if (year==9999) {
1.141 brouard 10677: 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);
10678: 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 10679: return 1;
10680:
1.136 brouard 10681: }
10682: annais[i]=(double)(year);
1.302 brouard 10683: moisnais[i]=(double)(month);
10684: for (j=1;j<=maxwav;j++){
10685: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
10686: 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]);
10687: 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]);
10688: }
10689: }
10690:
1.136 brouard 10691: strcpy(line,stra);
1.225 brouard 10692:
1.223 brouard 10693: /* Sample weight */
1.136 brouard 10694: cutv(stra, strb,line,' ');
10695: errno=0;
10696: dval=strtod(strb,&endptr);
10697: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 10698: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
10699: 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 10700: fflush(ficlog);
10701: return 1;
10702: }
10703: weight[i]=dval;
10704: strcpy(line,stra);
1.225 brouard 10705:
1.223 brouard 10706: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10707: cutv(stra, strb, line, ' ');
10708: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10709: lval=-1;
1.311 brouard 10710: coqvar[iv][i]=NAN;
10711: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10712: }else{
1.225 brouard 10713: errno=0;
10714: /* what_kind_of_number(strb); */
10715: dval=strtod(strb,&endptr);
10716: /* if(strb != endptr && *endptr == '\0') */
10717: /* dval=dlval; */
10718: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10719: if( strb[0]=='\0' || (*endptr != '\0')){
10720: 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);
10721: 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);
10722: return 1;
10723: }
10724: coqvar[iv][i]=dval;
1.226 brouard 10725: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10726: }
10727: strcpy(line,stra);
10728: }/* end loop nqv */
1.136 brouard 10729:
1.223 brouard 10730: /* Covariate values */
1.136 brouard 10731: for (j=ncovcol;j>=1;j--){
10732: cutv(stra, strb,line,' ');
1.223 brouard 10733: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10734: lval=-1;
1.136 brouard 10735: }else{
1.225 brouard 10736: errno=0;
10737: lval=strtol(strb,&endptr,10);
10738: if( strb[0]=='\0' || (*endptr != '\0')){
10739: 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);
10740: 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);
10741: return 1;
10742: }
1.136 brouard 10743: }
10744: if(lval <-1 || lval >1){
1.225 brouard 10745: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10746: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10747: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10748: For example, for multinomial values like 1, 2 and 3,\n \
10749: build V1=0 V2=0 for the reference value (1),\n \
10750: V1=1 V2=0 for (2) \n \
1.136 brouard 10751: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10752: output of IMaCh is often meaningless.\n \
1.136 brouard 10753: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10754: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10755: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10756: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10757: For example, for multinomial values like 1, 2 and 3,\n \
10758: build V1=0 V2=0 for the reference value (1),\n \
10759: V1=1 V2=0 for (2) \n \
1.136 brouard 10760: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10761: output of IMaCh is often meaningless.\n \
1.136 brouard 10762: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10763: return 1;
1.136 brouard 10764: }
10765: covar[j][i]=(double)(lval);
10766: strcpy(line,stra);
10767: }
10768: lstra=strlen(stra);
1.225 brouard 10769:
1.136 brouard 10770: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10771: stratrunc = &(stra[lstra-9]);
10772: num[i]=atol(stratrunc);
10773: }
10774: else
10775: num[i]=atol(stra);
10776: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10777: 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;}*/
10778:
10779: i=i+1;
10780: } /* End loop reading data */
1.225 brouard 10781:
1.136 brouard 10782: *imax=i-1; /* Number of individuals */
10783: fclose(fic);
1.225 brouard 10784:
1.136 brouard 10785: return (0);
1.164 brouard 10786: /* endread: */
1.225 brouard 10787: printf("Exiting readdata: ");
10788: fclose(fic);
10789: return (1);
1.223 brouard 10790: }
1.126 brouard 10791:
1.234 brouard 10792: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10793: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10794: while (*p2 == ' ')
1.234 brouard 10795: p2++;
10796: /* while ((*p1++ = *p2++) !=0) */
10797: /* ; */
10798: /* do */
10799: /* while (*p2 == ' ') */
10800: /* p2++; */
10801: /* while (*p1++ == *p2++); */
10802: *stri=p2;
1.145 brouard 10803: }
10804:
1.330 brouard 10805: int decoderesult( char resultline[], int nres)
1.230 brouard 10806: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10807: {
1.235 brouard 10808: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10809: char resultsav[MAXLINE];
1.330 brouard 10810: /* int resultmodel[MAXLINE]; */
1.334 brouard 10811: /* int modelresult[MAXLINE]; */
1.230 brouard 10812: char stra[80], strb[80], strc[80], strd[80],stre[80];
10813:
1.234 brouard 10814: removefirstspace(&resultline);
1.332 brouard 10815: printf("decoderesult:%s\n",resultline);
1.230 brouard 10816:
1.332 brouard 10817: strcpy(resultsav,resultline);
1.342 brouard 10818: /* printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline); */
1.230 brouard 10819: if (strlen(resultsav) >1){
1.334 brouard 10820: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' in this resultline */
1.230 brouard 10821: }
1.253 brouard 10822: if(j == 0){ /* Resultline but no = */
10823: TKresult[nres]=0; /* Combination for the nresult and the model */
10824: return (0);
10825: }
1.234 brouard 10826: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.334 brouard 10827: 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);
10828: 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 10829: /* return 1;*/
1.234 brouard 10830: }
1.334 brouard 10831: for(k=1; k<=j;k++){ /* Loop on any covariate of the RESULT LINE */
1.234 brouard 10832: if(nbocc(resultsav,'=') >1){
1.318 brouard 10833: 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 10834: /* 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 10835: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 10836: /* If a blank, then strc="V4=" and strd='\0' */
10837: if(strc[0]=='\0'){
10838: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
10839: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
10840: return 1;
10841: }
1.234 brouard 10842: }else
10843: cutl(strc,strd,resultsav,'=');
1.318 brouard 10844: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10845:
1.230 brouard 10846: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10847: 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 10848: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10849: /* cptcovsel++; */
10850: if (nbocc(stra,'=') >0)
10851: strcpy(resultsav,stra); /* and analyzes it */
10852: }
1.235 brouard 10853: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10854: /* 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 10855: 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 10856: if(Typevar[k1]==0){ /* Single covariate in model */
10857: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 10858: match=0;
1.318 brouard 10859: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10860: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10861: modelresult[nres][k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10862: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10863: break;
10864: }
10865: }
10866: if(match == 0){
1.338 brouard 10867: printf("Error in result line (Dummy single): V%d is missing in result: %s according to model=1+age+%s. Tvar[k1=%d]=%d is different from Tvarsel[k2=%d]=%d.\n",Tvar[k1], resultline, model,k1, Tvar[k1], k2, Tvarsel[k2]);
10868: fprintf(ficlog,"Error in result line (Dummy single): V%d is missing in result: %s according to model=1+age+%s\n",Tvar[k1], resultline, model);
1.310 brouard 10869: return 1;
1.234 brouard 10870: }
1.332 brouard 10871: }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*/
10872: /* We feed resultmodel[k1]=k2; */
10873: match=0;
10874: 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 */
10875: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.334 brouard 10876: 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 10877: resultmodel[nres][k1]=k2; /* Added here */
1.342 brouard 10878: /* printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]); */
1.332 brouard 10879: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10880: break;
10881: }
10882: }
10883: if(match == 0){
1.338 brouard 10884: printf("Error in result line (Product with age): V%d is missing in result: %s according to model=1+age+%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
10885: fprintf(ficlog,"Error in result line (Product with age): V%d is missing in result: %s according to model=1+age+%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
1.332 brouard 10886: return 1;
10887: }
10888: }else if(Typevar[k1]==2){ /* Product No age We want to get the position in the resultline of the product in the model line*/
10889: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
10890: match=0;
1.342 brouard 10891: /* printf("Decoderesult very first Product Tvardk[k1=%d][1]=%d Tvardk[k1=%d][2]=%d V%d * V%d\n",k1,Tvardk[k1][1],k1,Tvardk[k1][2],Tvardk[k1][1],Tvardk[k1][2]); */
1.332 brouard 10892: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10893: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10894: /* modelresult[k2]=k1; */
1.342 brouard 10895: /* printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]); */
1.332 brouard 10896: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10897: }
10898: }
10899: if(match == 0){
1.338 brouard 10900: printf("Error in result line (Product without age first variable): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][1], resultline, model);
10901: fprintf(ficlog,"Error in result line (Product without age first variable): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][1], resultline, model);
1.332 brouard 10902: return 1;
10903: }
10904: match=0;
10905: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10906: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10907: /* modelresult[k2]=k1;*/
1.342 brouard 10908: /* printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]); */
1.332 brouard 10909: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10910: break;
10911: }
10912: }
10913: if(match == 0){
1.338 brouard 10914: printf("Error in result line (Product without age second variable): V%d is missing in result: %s according to model=1+age+%s\n",Tvardk[k1][2], resultline, model);
10915: fprintf(ficlog,"Error in result line (Product without age second variable): V%d is missing in result : %s according to model=1+age+%s\n",Tvardk[k1][2], resultline, model);
1.332 brouard 10916: return 1;
10917: }
10918: }/* End of testing */
1.333 brouard 10919: }/* End loop cptcovt */
1.235 brouard 10920: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10921: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.334 brouard 10922: 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)
10923: * Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 10924: match=0;
1.318 brouard 10925: 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 10926: if(Typevar[k1]==0){ /* Single only */
1.237 brouard 10927: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 brouard 10928: 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 10929: 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 10930: ++match;
10931: }
10932: }
10933: }
10934: if(match == 0){
1.338 brouard 10935: printf("Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
10936: fprintf(ficlog,"Error in result line: variable V%d is missing in model; result: %s, model=1+age+%s\n",Tvarsel[k2], resultline, model);
1.310 brouard 10937: return 1;
1.234 brouard 10938: }else if(match > 1){
1.338 brouard 10939: printf("Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
10940: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=1+age+%s\n",k2, resultline, model);
1.310 brouard 10941: return 1;
1.234 brouard 10942: }
10943: }
1.334 brouard 10944: /* cptcovres=j /\* Number of variables in the resultline is equal to cptcovs and thus useless *\/ */
1.234 brouard 10945: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10946: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 10947: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10948: /* 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*/
10949: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 10950: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10951: /* 1 0 0 0 */
10952: /* 2 1 0 0 */
10953: /* 3 0 1 0 */
1.330 brouard 10954: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 10955: /* 5 0 0 1 */
1.330 brouard 10956: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 10957: /* 7 0 1 1 */
10958: /* 8 1 1 1 */
1.237 brouard 10959: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10960: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10961: /* V5*age V5 known which value for nres? */
10962: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.334 brouard 10963: 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.
10964: * loop on position k1 in the MODEL LINE */
1.331 brouard 10965: /* k counting number of combination of single dummies in the equation model */
10966: /* k4 counting single dummies in the equation model */
10967: /* k4q counting single quantitatives in the equation model */
1.344 ! brouard 10968: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Dummy and Single, fixed or timevarying, k1 is sorting according to MODEL, but k3 to resultline */
1.334 brouard 10969: /* 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 10970: /* 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 10971: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 10972: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
10973: /* k3 is the position in the nres result line of the k1th variable of the model equation */
10974: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
10975: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
10976: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
1.334 brouard 10977: /* Tvresult[nres][result_position]= name of the dummy variable at the result_position in the nres resultline */
1.332 brouard 10978: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 10979: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 10980: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
10981: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
10982: 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 10983: 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 10984: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][Name]=Value; stores the value into the name of the variable. */
1.332 brouard 10985: /* Tinvresult[nres][4]=1 */
1.334 brouard 10986: /* Tresult[nres][k4+1]=Tvalsel[k3];/\* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) *\/ */
10987: Tresult[nres][k3]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
10988: /* Tvresult[nres][k4+1]=(int)Tvarsel[k3];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
10989: Tvresult[nres][k3]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
1.237 brouard 10990: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.334 brouard 10991: precov[nres][k1]=Tvalsel[k3]; /* Value from resultline of the variable at the k1 position in the model */
1.342 brouard 10992: /* 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 10993: k4++;;
1.331 brouard 10994: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 10995: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 10996: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 10997: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 10998: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
10999: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
11000: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 11001: /* Tqresult[nres][k4q+1]=Tvalsel[k3q]; /\* Tqresult[nres][1]=25.1 *\/ */
11002: /* Tvresult[nres][k4q+1]=(int)Tvarsel[k3q];/\* Tvresult[nres][1]=4 Tvresult[nres][3]=1 *\/ */
11003: /* Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /\* Tvqresult[nres][1]=5 *\/ */
11004: Tqresult[nres][k3q]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
11005: Tvresult[nres][k3q]=(int)Tvarsel[k3q];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
11006: Tvqresult[nres][k3q]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
1.237 brouard 11007: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 11008: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 11009: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 11010: /* 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 11011: k4q++;;
1.331 brouard 11012: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
11013: /* Tvar[k1]; */ /* Age variable */
1.332 brouard 11014: /* Wrong we want the value of variable name Tvar[k1] */
11015:
11016: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
1.331 brouard 11017: 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 11018: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* TinvDoQresult[nres][4]=1 */
1.332 brouard 11019: precov[nres][k1]=Tvalsel[k3];
1.342 brouard 11020: /* 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 11021: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.332 brouard 11022: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.331 brouard 11023: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.334 brouard 11024: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* TinvDoQresult[nres][5]=25.1 */
1.332 brouard 11025: precov[nres][k1]=Tvalsel[k3q];
1.342 brouard 11026: /* 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 11027: }else if(Typevar[k1]==2 ){ /* For product quant or dummy (not with age) */
1.332 brouard 11028: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
1.342 brouard 11029: /* 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 11030: }else{
1.332 brouard 11031: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
11032: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 11033: }
11034: }
1.234 brouard 11035:
1.334 brouard 11036: TKresult[nres]=++k; /* Number of combinations of dummies for the nresult and the model =Tvalsel[k3]*pow(2,k4) + 1*/
1.230 brouard 11037: return (0);
11038: }
1.235 brouard 11039:
1.230 brouard 11040: int decodemodel( char model[], int lastobs)
11041: /**< This routine decodes the model and returns:
1.224 brouard 11042: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
11043: * - nagesqr = 1 if age*age in the model, otherwise 0.
11044: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
11045: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
11046: * - cptcovage number of covariates with age*products =2
11047: * - cptcovs number of simple covariates
1.339 brouard 11048: * ncovcolt=ncovcol+nqv+ntv+nqtv total of covariates in the data, not in the model equation
1.224 brouard 11049: * - Tvar[k] is the id of the kth covariate Tvar[1]@12 {1, 2, 3, 8, 10, 11, 8, 3, 7, 8, 5, 6}, thus Tvar[5=V7*V8]=10
1.339 brouard 11050: * which is a new column after the 9 (ncovcol+nqv+ntv+nqtv) variables.
1.319 brouard 11051: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 11052: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
11053: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
11054: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
11055: */
1.319 brouard 11056: /* 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 11057: {
1.238 brouard 11058: int i, j, k, ks, v;
1.227 brouard 11059: int j1, k1, k2, k3, k4;
1.136 brouard 11060: char modelsav[80];
1.145 brouard 11061: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 11062: char *strpt;
1.136 brouard 11063:
1.145 brouard 11064: /*removespace(model);*/
1.136 brouard 11065: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 11066: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 11067: if (strstr(model,"AGE") !=0){
1.192 brouard 11068: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
11069: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 11070: return 1;
11071: }
1.141 brouard 11072: if (strstr(model,"v") !=0){
1.338 brouard 11073: printf("Error. 'v' must be in upper case 'V' model=1+age+%s ",model);
11074: fprintf(ficlog,"Error. 'v' must be in upper case model=1+age+%s ",model);fflush(ficlog);
1.141 brouard 11075: return 1;
11076: }
1.187 brouard 11077: strcpy(modelsav,model);
11078: if ((strpt=strstr(model,"age*age")) !=0){
1.338 brouard 11079: printf(" strpt=%s, model=1+age+%s\n",strpt, model);
1.187 brouard 11080: if(strpt != model){
1.338 brouard 11081: printf("Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 11082: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 11083: corresponding column of parameters.\n",model);
1.338 brouard 11084: fprintf(ficlog,"Error in model: 'model=1+age+%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 11085: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 11086: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 11087: return 1;
1.225 brouard 11088: }
1.187 brouard 11089: nagesqr=1;
11090: if (strstr(model,"+age*age") !=0)
1.234 brouard 11091: substrchaine(modelsav, model, "+age*age");
1.187 brouard 11092: else if (strstr(model,"age*age+") !=0)
1.234 brouard 11093: substrchaine(modelsav, model, "age*age+");
1.187 brouard 11094: else
1.234 brouard 11095: substrchaine(modelsav, model, "age*age");
1.187 brouard 11096: }else
11097: nagesqr=0;
11098: if (strlen(modelsav) >1){
11099: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
11100: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 11101: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 11102: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 11103: * cst, age and age*age
11104: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
11105: /* including age products which are counted in cptcovage.
11106: * but the covariates which are products must be treated
11107: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 11108: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
11109: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 11110:
11111:
1.187 brouard 11112: /* Design
11113: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
11114: * < ncovcol=8 >
11115: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
11116: * k= 1 2 3 4 5 6 7 8
11117: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
11118: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 11119: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
11120: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 11121: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
11122: * Tage[++cptcovage]=k
11123: * if products, new covar are created after ncovcol with k1
11124: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
11125: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
11126: * 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
11127: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
11128: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
11129: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
11130: * < ncovcol=8 >
11131: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
11132: * k= 1 2 3 4 5 6 7 8 9 10 11 12
11133: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 11134: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 11135: * p Tprod[1]@2={ 6, 5}
11136: *p Tvard[1][1]@4= {7, 8, 5, 6}
11137: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
11138: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 11139: *How to reorganize? Tvars(orted)
1.187 brouard 11140: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
11141: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
11142: * {2, 1, 4, 8, 5, 6, 3, 7}
11143: * Struct []
11144: */
1.225 brouard 11145:
1.187 brouard 11146: /* This loop fills the array Tvar from the string 'model'.*/
11147: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
11148: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
11149: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
11150: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
11151: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
11152: /* k=1 Tvar[1]=2 (from V2) */
11153: /* k=5 Tvar[5] */
11154: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 11155: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 11156: /* } */
1.198 brouard 11157: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 11158: /*
11159: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 11160: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
11161: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
11162: }
1.187 brouard 11163: cptcovage=0;
1.319 brouard 11164: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
11165: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
11166: 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" */
11167: if (nbocc(modelsav,'+')==0)
11168: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 11169: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
11170: /*scanf("%d",i);*/
1.319 brouard 11171: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
11172: 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 11173: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
11174: /* covar is not filled and then is empty */
11175: cptcovprod--;
11176: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 11177: 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 11178: Typevar[k]=1; /* 1 for age product */
1.319 brouard 11179: cptcovage++; /* Counts the number of covariates which include age as a product */
11180: 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 11181: /*printf("stre=%s ", stre);*/
11182: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
11183: cptcovprod--;
11184: cutl(stre,strb,strc,'V');
11185: Tvar[k]=atoi(stre);
11186: Typevar[k]=1; /* 1 for age product */
11187: cptcovage++;
11188: Tage[cptcovage]=k;
11189: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
11190: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
11191: cptcovn++;
11192: cptcovprodnoage++;k1++;
11193: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
1.339 brouard 11194: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* ncovcolt+k1; For model-covariate k tells which data-covariate to use but
1.234 brouard 11195: because this model-covariate is a construction we invent a new column
11196: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.335 brouard 11197: If already ncovcol=4 and model= V2 + V1 + V1*V4 + age*V3 + V3*V2
1.319 brouard 11198: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
1.339 brouard 11199: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=3 etc */
1.335 brouard 11200: /* Please remark that the new variables are model dependent */
11201: /* If we have 4 variable but the model uses only 3, like in
11202: * model= V1 + age*V1 + V2 + V3 + age*V2 + age*V3 + V1*V2 + V1*V3
11203: * k= 1 2 3 4 5 6 7 8
11204: * Tvar[k]=1 1 2 3 2 3 (5 6) (and not 4 5 because of V4 missing)
11205: * Tage[kk] [1]= 2 [2]=5 [3]=6 kk=1 to cptcovage=3
11206: * Tvar[Tage[kk]][1]=2 [2]=2 [3]=3
11207: */
1.339 brouard 11208: Typevar[k]=2; /* 2 for product */
1.234 brouard 11209: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
11210: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 11211: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 11212: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 brouard 11213: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 11214: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 brouard 11215: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 11216: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
11217: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
11218: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 11219: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 11220: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
1.339 brouard 11221: if( FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){ /* If the product is a fixed covariate then we feed the new column with Vn*Vm */
11222: for (i=1; i<=lastobs;i++){/* For fixed product */
1.234 brouard 11223: /* Computes the new covariate which is a product of
11224: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
1.339 brouard 11225: covar[ncovcolt+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
11226: }
11227: } /*End of FixedV */
1.234 brouard 11228: } /* End age is not in the model */
11229: } /* End if model includes a product */
1.319 brouard 11230: else { /* not a product */
1.234 brouard 11231: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
11232: /* scanf("%d",i);*/
11233: cutl(strd,strc,strb,'V');
11234: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
11235: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
11236: Tvar[k]=atoi(strd);
11237: Typevar[k]=0; /* 0 for simple covariates */
11238: }
11239: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 11240: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 11241: scanf("%d",i);*/
1.187 brouard 11242: } /* end of loop + on total covariates */
11243: } /* end if strlen(modelsave == 0) age*age might exist */
11244: } /* end if strlen(model == 0) */
1.136 brouard 11245:
11246: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
11247: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 11248:
1.136 brouard 11249: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 11250: printf("cptcovprod=%d ", cptcovprod);
11251: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
11252: scanf("%d ",i);*/
11253:
11254:
1.230 brouard 11255: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
11256: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 11257: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
11258: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
11259: k = 1 2 3 4 5 6 7 8 9
11260: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 11261: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 11262: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
11263: Dummy[k] 1 0 0 0 3 1 1 2 3
11264: Tmodelind[combination of covar]=k;
1.225 brouard 11265: */
11266: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 11267: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 11268: /* 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 11269: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 11270: printf("Model=1+age+%s\n\
1.227 brouard 11271: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11272: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11273: 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 11274: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 11275: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
11276: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
11277: Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product\n",model);
1.342 brouard 11278: for(k=-1;k<=NCOVMAX; k++){ Fixed[k]=0; Dummy[k]=0;}
11279: for(k=1;k<=NCOVMAX; k++){TvarFind[k]=0; TvarVind[k]=0;}
1.343 brouard 11280: for(k=1, ncovf=0, nsd=0, nsq=0, ncovv=0, ncova=0, ncoveff=0, nqfveff=0, ntveff=0, nqtveff=0, ncovvt=0;k<=cptcovt; k++){ /* or cptocvt loop on k from model */
1.234 brouard 11281: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 11282: Fixed[k]= 0;
11283: Dummy[k]= 0;
1.225 brouard 11284: ncoveff++;
1.232 brouard 11285: ncovf++;
1.234 brouard 11286: nsd++;
11287: modell[k].maintype= FTYPE;
11288: TvarsD[nsd]=Tvar[k];
11289: TvarsDind[nsd]=k;
1.330 brouard 11290: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 11291: TvarF[ncovf]=Tvar[k];
11292: TvarFind[ncovf]=k;
11293: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
11294: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.339 brouard 11295: /* }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /\* Product of fixed dummy (<=ncovcol) covariates For a fixed product k is higher than ncovcol *\/ */
11296: }else if( Tposprod[k]>0 && Typevar[k]==2 && FixedV[Tvardk[k][1]] == 0 && FixedV[Tvardk[k][2]] == 0){ /* Needs a fixed product Product of fixed dummy (<=ncovcol) covariates For a fixed product k is higher than ncovcol */
1.234 brouard 11297: Fixed[k]= 0;
11298: Dummy[k]= 0;
11299: ncoveff++;
11300: ncovf++;
11301: modell[k].maintype= FTYPE;
11302: TvarF[ncovf]=Tvar[k];
1.330 brouard 11303: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 11304: TvarFind[ncovf]=k;
1.230 brouard 11305: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 11306: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 11307: }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 11308: Fixed[k]= 0;
11309: Dummy[k]= 1;
1.230 brouard 11310: nqfveff++;
1.234 brouard 11311: modell[k].maintype= FTYPE;
11312: modell[k].subtype= FQ;
11313: nsq++;
1.334 brouard 11314: TvarsQ[nsq]=Tvar[k]; /* Gives the variable name (extended to products) of first nsq simple quantitative covariates (fixed or time vary see below */
11315: TvarsQind[nsq]=k; /* Gives the position in the model equation of the first nsq simple quantitative covariates (fixed or time vary) */
1.232 brouard 11316: ncovf++;
1.234 brouard 11317: TvarF[ncovf]=Tvar[k];
11318: TvarFind[ncovf]=k;
1.231 brouard 11319: 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 11320: 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 11321: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.339 brouard 11322: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11323: /* model V1+V3+age*V1+age*V3+V1*V3 */
11324: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11325: ncovvt++;
11326: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11327: TvarVVind[ncovvt]=k; /* TvarVVind[1]=2 (second position in the model equation */
11328:
1.227 brouard 11329: Fixed[k]= 1;
11330: Dummy[k]= 0;
1.225 brouard 11331: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 11332: modell[k].maintype= VTYPE;
11333: modell[k].subtype= VD;
11334: nsd++;
11335: TvarsD[nsd]=Tvar[k];
11336: TvarsDind[nsd]=k;
1.330 brouard 11337: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 11338: ncovv++; /* Only simple time varying variables */
11339: TvarV[ncovv]=Tvar[k];
1.242 brouard 11340: 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 11341: 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 */
11342: 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 11343: 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);
11344: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 11345: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.339 brouard 11346: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11347: /* model V1+V3+age*V1+age*V3+V1*V3 */
11348: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11349: ncovvt++;
11350: TvarVV[ncovvt]=Tvar[k]; /* TvarVV[1]=V3 (first time varying in the model equation */
11351: TvarVVind[ncovvt]=k; /* TvarVV[1]=V3 (first time varying in the model equation */
11352:
1.234 brouard 11353: Fixed[k]= 1;
11354: Dummy[k]= 1;
11355: nqtveff++;
11356: modell[k].maintype= VTYPE;
11357: modell[k].subtype= VQ;
11358: ncovv++; /* Only simple time varying variables */
11359: nsq++;
1.334 brouard 11360: 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) */
11361: 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 11362: TvarV[ncovv]=Tvar[k];
1.242 brouard 11363: 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 11364: 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 */
11365: 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 11366: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
11367: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
1.342 brouard 11368: /* 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); */
11369: /* printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv); */
1.227 brouard 11370: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 11371: ncova++;
11372: TvarA[ncova]=Tvar[k];
11373: TvarAind[ncova]=k;
1.231 brouard 11374: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 11375: Fixed[k]= 2;
11376: Dummy[k]= 2;
11377: modell[k].maintype= ATYPE;
11378: modell[k].subtype= APFD;
11379: /* ncoveff++; */
1.227 brouard 11380: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 11381: Fixed[k]= 2;
11382: Dummy[k]= 3;
11383: modell[k].maintype= ATYPE;
11384: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
11385: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 11386: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 11387: Fixed[k]= 3;
11388: Dummy[k]= 2;
11389: modell[k].maintype= ATYPE;
11390: modell[k].subtype= APVD; /* Product age * varying dummy */
11391: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 11392: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 11393: Fixed[k]= 3;
11394: Dummy[k]= 3;
11395: modell[k].maintype= ATYPE;
11396: modell[k].subtype= APVQ; /* Product age * varying quantitative */
11397: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 11398: }
1.339 brouard 11399: }else if (Typevar[k] == 2) { /* product Vn * Vm without age, V1+V3+age*V1+age*V3+V1*V3 looking at V1*V3, Typevar={0, 0, 1, 1, 2}, k=5, V1 is fixed, V3 is timevary, V5 is a product */
11400: /*# ID V1 V2 weight birth death 1st s1 V3 V4 V5 2nd s2 */
11401: /* model V1+V3+age*V1+age*V3+V1*V3 */
11402: /* Tvar={1, 3, 1, 3, 6}, the 6 comes from the fact that there are already V1, V2, V3, V4, V5 native covariates */
11403: k1=Tposprod[k]; /* Position in the products of product k, Tposprod={0, 0, 0, 0, 1} k1=1 first product but second time varying because of V3 */
11404: ncovvt++;
11405: TvarVV[ncovvt]=Tvard[k1][1]; /* TvarVV[2]=V1 (because TvarVV[1] was V3, first time varying covariates */
11406: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11407: ncovvt++;
11408: TvarVV[ncovvt]=Tvard[k1][2]; /* TvarVV[3]=V3 */
11409: TvarVVind[ncovvt]=k; /* TvarVVind[2]=5 (because TvarVVind[2] was V1*V3 at position 5 */
11410:
11411:
11412: if(Tvard[k1][1] <=ncovcol){ /* Vn is dummy fixed, (Tvard[1][1]=V1), (Tvard[1][1]=V3 time varying) */
11413: if(Tvard[k1][2] <=ncovcol){ /* Vm is dummy fixed */
1.240 brouard 11414: Fixed[k]= 1;
11415: Dummy[k]= 0;
11416: modell[k].maintype= FTYPE;
11417: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
11418: ncovf++; /* Fixed variables without age */
11419: TvarF[ncovf]=Tvar[k];
11420: TvarFind[ncovf]=k;
1.339 brouard 11421: }else if(Tvard[k1][2] <=ncovcol+nqv){ /* Vm is quanti fixed */
11422: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11423: Dummy[k]= 1;
11424: modell[k].maintype= FTYPE;
11425: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
11426: ncovf++; /* Varying variables without age */
11427: TvarF[ncovf]=Tvar[k];
11428: TvarFind[ncovf]=k;
1.339 brouard 11429: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is a time varying dummy covariate */
1.240 brouard 11430: Fixed[k]= 1;
11431: Dummy[k]= 0;
11432: modell[k].maintype= VTYPE;
11433: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
11434: ncovv++; /* Varying variables without age */
1.339 brouard 11435: TvarV[ncovv]=Tvar[k]; /* TvarV[1]=Tvar[5]=5 because there is a V4 */
11436: TvarVind[ncovv]=k;/* TvarVind[1]=5 */
11437: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is a time varying quantitative covariate */
1.240 brouard 11438: Fixed[k]= 1;
11439: Dummy[k]= 1;
11440: modell[k].maintype= VTYPE;
11441: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
11442: ncovv++; /* Varying variables without age */
11443: TvarV[ncovv]=Tvar[k];
11444: TvarVind[ncovv]=k;
11445: }
1.339 brouard 11446: }else if(Tvard[k1][1] <=ncovcol+nqv){ /* Vn is fixed quanti */
11447: if(Tvard[k1][2] <=ncovcol){ /* Vm is fixed dummy */
11448: Fixed[k]= 0; /* Fixed product */
1.240 brouard 11449: Dummy[k]= 1;
11450: modell[k].maintype= FTYPE;
11451: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
11452: ncovf++; /* Fixed variables without age */
11453: TvarF[ncovf]=Tvar[k];
11454: TvarFind[ncovf]=k;
1.339 brouard 11455: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){ /* Vm is time varying */
1.240 brouard 11456: Fixed[k]= 1;
11457: Dummy[k]= 1;
11458: modell[k].maintype= VTYPE;
11459: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
11460: ncovv++; /* Varying variables without age */
11461: TvarV[ncovv]=Tvar[k];
11462: TvarVind[ncovv]=k;
1.339 brouard 11463: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){ /* Vm is time varying quanti */
1.240 brouard 11464: Fixed[k]= 1;
11465: Dummy[k]= 1;
11466: modell[k].maintype= VTYPE;
11467: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
11468: ncovv++; /* Varying variables without age */
11469: TvarV[ncovv]=Tvar[k];
11470: TvarVind[ncovv]=k;
11471: ncovv++; /* Varying variables without age */
11472: TvarV[ncovv]=Tvar[k];
11473: TvarVind[ncovv]=k;
11474: }
1.339 brouard 11475: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){ /* Vn is time varying dummy */
1.240 brouard 11476: if(Tvard[k1][2] <=ncovcol){
11477: Fixed[k]= 1;
11478: Dummy[k]= 1;
11479: modell[k].maintype= VTYPE;
11480: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
11481: ncovv++; /* Varying variables without age */
11482: TvarV[ncovv]=Tvar[k];
11483: TvarVind[ncovv]=k;
11484: }else if(Tvard[k1][2] <=ncovcol+nqv){
11485: Fixed[k]= 1;
11486: Dummy[k]= 1;
11487: modell[k].maintype= VTYPE;
11488: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
11489: ncovv++; /* Varying variables without age */
11490: TvarV[ncovv]=Tvar[k];
11491: TvarVind[ncovv]=k;
11492: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11493: Fixed[k]= 1;
11494: Dummy[k]= 0;
11495: modell[k].maintype= VTYPE;
11496: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
11497: ncovv++; /* Varying variables without age */
11498: TvarV[ncovv]=Tvar[k];
11499: TvarVind[ncovv]=k;
11500: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11501: Fixed[k]= 1;
11502: Dummy[k]= 1;
11503: modell[k].maintype= VTYPE;
11504: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
11505: ncovv++; /* Varying variables without age */
11506: TvarV[ncovv]=Tvar[k];
11507: TvarVind[ncovv]=k;
11508: }
1.339 brouard 11509: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){ /* Vn is time varying quanti */
1.240 brouard 11510: if(Tvard[k1][2] <=ncovcol){
11511: Fixed[k]= 1;
11512: Dummy[k]= 1;
11513: modell[k].maintype= VTYPE;
11514: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
11515: ncovv++; /* Varying variables without age */
11516: TvarV[ncovv]=Tvar[k];
11517: TvarVind[ncovv]=k;
11518: }else if(Tvard[k1][2] <=ncovcol+nqv){
11519: Fixed[k]= 1;
11520: Dummy[k]= 1;
11521: modell[k].maintype= VTYPE;
11522: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
11523: ncovv++; /* Varying variables without age */
11524: TvarV[ncovv]=Tvar[k];
11525: TvarVind[ncovv]=k;
11526: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
11527: Fixed[k]= 1;
11528: Dummy[k]= 1;
11529: modell[k].maintype= VTYPE;
11530: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
11531: ncovv++; /* Varying variables without age */
11532: TvarV[ncovv]=Tvar[k];
11533: TvarVind[ncovv]=k;
11534: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
11535: Fixed[k]= 1;
11536: Dummy[k]= 1;
11537: modell[k].maintype= VTYPE;
11538: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
11539: ncovv++; /* Varying variables without age */
11540: TvarV[ncovv]=Tvar[k];
11541: TvarVind[ncovv]=k;
11542: }
1.227 brouard 11543: }else{
1.240 brouard 11544: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11545: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
11546: } /*end k1*/
1.225 brouard 11547: }else{
1.226 brouard 11548: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
11549: 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 11550: }
1.342 brouard 11551: /* 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]); */
11552: /* printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype); */
1.227 brouard 11553: 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]);
11554: }
11555: /* Searching for doublons in the model */
11556: for(k1=1; k1<= cptcovt;k1++){
11557: for(k2=1; k2 <k1;k2++){
1.285 brouard 11558: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
11559: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 11560: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
11561: if(Tvar[k1]==Tvar[k2]){
1.338 brouard 11562: printf("Error duplication in the model=1+age+%s at positions (+) %d and %d, Tvar[%d]=V%d, Tvar[%d]=V%d, Typevar=%d, Fixed=%d, Dummy=%d\n", model, k1,k2, k1, Tvar[k1], k2, Tvar[k2],Typevar[k1],Fixed[k1],Dummy[k1]);
11563: fprintf(ficlog,"Error duplication in the model=1+age+%s at positions (+) %d and %d, Tvar[%d]=V%d, Tvar[%d]=V%d, Typevar=%d, Fixed=%d, Dummy=%d\n", model, k1,k2, k1, Tvar[k1], k2, Tvar[k2],Typevar[k1],Fixed[k1],Dummy[k1]); fflush(ficlog);
1.234 brouard 11564: return(1);
11565: }
11566: }else if (Typevar[k1] ==2){
11567: k3=Tposprod[k1];
11568: k4=Tposprod[k2];
11569: if( ((Tvard[k3][1]== Tvard[k4][1])&&(Tvard[k3][2]== Tvard[k4][2])) || ((Tvard[k3][1]== Tvard[k4][2])&&(Tvard[k3][2]== Tvard[k4][1])) ){
1.338 brouard 11570: printf("Error duplication in the model=1+age+%s at positions (+) %d and %d, V%d*V%d, Typevar=%d, Fixed=%d, Dummy=%d\n",model, k1,k2, Tvard[k3][1], Tvard[k3][2],Typevar[k1],Fixed[Tvar[k1]],Dummy[Tvar[k1]]);
11571: fprintf(ficlog,"Error duplication in the model=1+age+%s at positions (+) %d and %d, V%d*V%d, Typevar=%d, Fixed=%d, Dummy=%d\n",model, k1,k2, Tvard[k3][1], Tvard[k3][2],Typevar[k1],Fixed[Tvar[k1]],Dummy[Tvar[k1]]); fflush(ficlog);
1.234 brouard 11572: return(1);
11573: }
11574: }
1.227 brouard 11575: }
11576: }
1.225 brouard 11577: }
11578: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
11579: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 11580: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
11581: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 11582: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 11583: /*endread:*/
1.225 brouard 11584: printf("Exiting decodemodel: ");
11585: return (1);
1.136 brouard 11586: }
11587:
1.169 brouard 11588: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 11589: {/* Check ages at death */
1.136 brouard 11590: int i, m;
1.218 brouard 11591: int firstone=0;
11592:
1.136 brouard 11593: for (i=1; i<=imx; i++) {
11594: for(m=2; (m<= maxwav); m++) {
11595: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
11596: anint[m][i]=9999;
1.216 brouard 11597: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
11598: s[m][i]=-1;
1.136 brouard 11599: }
11600: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 11601: *nberr = *nberr + 1;
1.218 brouard 11602: if(firstone == 0){
11603: firstone=1;
1.260 brouard 11604: 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 11605: }
1.262 brouard 11606: 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 11607: s[m][i]=-1; /* Droping the death status */
1.136 brouard 11608: }
11609: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 11610: (*nberr)++;
1.259 brouard 11611: 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 11612: 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 11613: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 11614: }
11615: }
11616: }
11617:
11618: for (i=1; i<=imx; i++) {
11619: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
11620: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 11621: 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 11622: if (s[m][i] >= nlstate+1) {
1.169 brouard 11623: if(agedc[i]>0){
11624: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 11625: agev[m][i]=agedc[i];
1.214 brouard 11626: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 11627: }else {
1.136 brouard 11628: if ((int)andc[i]!=9999){
11629: nbwarn++;
11630: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
11631: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
11632: agev[m][i]=-1;
11633: }
11634: }
1.169 brouard 11635: } /* agedc > 0 */
1.214 brouard 11636: } /* end if */
1.136 brouard 11637: else if(s[m][i] !=9){ /* Standard case, age in fractional
11638: years but with the precision of a month */
11639: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
11640: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
11641: agev[m][i]=1;
11642: else if(agev[m][i] < *agemin){
11643: *agemin=agev[m][i];
11644: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
11645: }
11646: else if(agev[m][i] >*agemax){
11647: *agemax=agev[m][i];
1.156 brouard 11648: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 11649: }
11650: /*agev[m][i]=anint[m][i]-annais[i];*/
11651: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 11652: } /* en if 9*/
1.136 brouard 11653: else { /* =9 */
1.214 brouard 11654: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 11655: agev[m][i]=1;
11656: s[m][i]=-1;
11657: }
11658: }
1.214 brouard 11659: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 11660: agev[m][i]=1;
1.214 brouard 11661: else{
11662: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11663: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11664: agev[m][i]=0;
11665: }
11666: } /* End for lastpass */
11667: }
1.136 brouard 11668:
11669: for (i=1; i<=imx; i++) {
11670: for(m=firstpass; (m<=lastpass); m++){
11671: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 11672: (*nberr)++;
1.136 brouard 11673: 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);
11674: 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);
11675: return 1;
11676: }
11677: }
11678: }
11679:
11680: /*for (i=1; i<=imx; i++){
11681: for (m=firstpass; (m<lastpass); m++){
11682: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
11683: }
11684:
11685: }*/
11686:
11687:
1.139 brouard 11688: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
11689: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 11690:
11691: return (0);
1.164 brouard 11692: /* endread:*/
1.136 brouard 11693: printf("Exiting calandcheckages: ");
11694: return (1);
11695: }
11696:
1.172 brouard 11697: #if defined(_MSC_VER)
11698: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11699: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11700: //#include "stdafx.h"
11701: //#include <stdio.h>
11702: //#include <tchar.h>
11703: //#include <windows.h>
11704: //#include <iostream>
11705: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
11706:
11707: LPFN_ISWOW64PROCESS fnIsWow64Process;
11708:
11709: BOOL IsWow64()
11710: {
11711: BOOL bIsWow64 = FALSE;
11712:
11713: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
11714: // (HANDLE, PBOOL);
11715:
11716: //LPFN_ISWOW64PROCESS fnIsWow64Process;
11717:
11718: HMODULE module = GetModuleHandle(_T("kernel32"));
11719: const char funcName[] = "IsWow64Process";
11720: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
11721: GetProcAddress(module, funcName);
11722:
11723: if (NULL != fnIsWow64Process)
11724: {
11725: if (!fnIsWow64Process(GetCurrentProcess(),
11726: &bIsWow64))
11727: //throw std::exception("Unknown error");
11728: printf("Unknown error\n");
11729: }
11730: return bIsWow64 != FALSE;
11731: }
11732: #endif
1.177 brouard 11733:
1.191 brouard 11734: void syscompilerinfo(int logged)
1.292 brouard 11735: {
11736: #include <stdint.h>
11737:
11738: /* #include "syscompilerinfo.h"*/
1.185 brouard 11739: /* command line Intel compiler 32bit windows, XP compatible:*/
11740: /* /GS /W3 /Gy
11741: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
11742: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
11743: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 11744: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
11745: */
11746: /* 64 bits */
1.185 brouard 11747: /*
11748: /GS /W3 /Gy
11749: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
11750: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
11751: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
11752: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
11753: /* Optimization are useless and O3 is slower than O2 */
11754: /*
11755: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
11756: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
11757: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
11758: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
11759: */
1.186 brouard 11760: /* Link is */ /* /OUT:"visual studio
1.185 brouard 11761: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
11762: /PDB:"visual studio
11763: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
11764: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
11765: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
11766: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
11767: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
11768: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
11769: uiAccess='false'"
11770: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
11771: /NOLOGO /TLBID:1
11772: */
1.292 brouard 11773:
11774:
1.177 brouard 11775: #if defined __INTEL_COMPILER
1.178 brouard 11776: #if defined(__GNUC__)
11777: struct utsname sysInfo; /* For Intel on Linux and OS/X */
11778: #endif
1.177 brouard 11779: #elif defined(__GNUC__)
1.179 brouard 11780: #ifndef __APPLE__
1.174 brouard 11781: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 11782: #endif
1.177 brouard 11783: struct utsname sysInfo;
1.178 brouard 11784: int cross = CROSS;
11785: if (cross){
11786: printf("Cross-");
1.191 brouard 11787: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 11788: }
1.174 brouard 11789: #endif
11790:
1.191 brouard 11791: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 11792: #if defined(__clang__)
1.191 brouard 11793: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 11794: #endif
11795: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 11796: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 11797: #endif
11798: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 11799: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 11800: #endif
11801: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 11802: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 11803: #endif
11804: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 11805: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 11806: #endif
11807: #if defined(_MSC_VER)
1.191 brouard 11808: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 11809: #endif
11810: #if defined(__PGI)
1.191 brouard 11811: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 11812: #endif
11813: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 11814: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 11815: #endif
1.191 brouard 11816: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 11817:
1.167 brouard 11818: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
11819: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
11820: // Windows (x64 and x86)
1.191 brouard 11821: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 11822: #elif __unix__ // all unices, not all compilers
11823: // Unix
1.191 brouard 11824: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 11825: #elif __linux__
11826: // linux
1.191 brouard 11827: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 11828: #elif __APPLE__
1.174 brouard 11829: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 11830: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11831: #endif
11832:
11833: /* __MINGW32__ */
11834: /* __CYGWIN__ */
11835: /* __MINGW64__ */
11836: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11837: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11838: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11839: /* _WIN64 // Defined for applications for Win64. */
11840: /* _M_X64 // Defined for compilations that target x64 processors. */
11841: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11842:
1.167 brouard 11843: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11844: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11845: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11846: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11847: #else
1.191 brouard 11848: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11849: #endif
11850:
1.169 brouard 11851: #if defined(__GNUC__)
11852: # if defined(__GNUC_PATCHLEVEL__)
11853: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11854: + __GNUC_MINOR__ * 100 \
11855: + __GNUC_PATCHLEVEL__)
11856: # else
11857: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11858: + __GNUC_MINOR__ * 100)
11859: # endif
1.174 brouard 11860: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11861: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11862:
11863: if (uname(&sysInfo) != -1) {
11864: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11865: 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 11866: }
11867: else
11868: perror("uname() error");
1.179 brouard 11869: //#ifndef __INTEL_COMPILER
11870: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11871: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11872: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11873: #endif
1.169 brouard 11874: #endif
1.172 brouard 11875:
1.286 brouard 11876: // void main ()
1.172 brouard 11877: // {
1.169 brouard 11878: #if defined(_MSC_VER)
1.174 brouard 11879: if (IsWow64()){
1.191 brouard 11880: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11881: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11882: }
11883: else{
1.191 brouard 11884: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11885: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11886: }
1.172 brouard 11887: // printf("\nPress Enter to continue...");
11888: // getchar();
11889: // }
11890:
1.169 brouard 11891: #endif
11892:
1.167 brouard 11893:
1.219 brouard 11894: }
1.136 brouard 11895:
1.219 brouard 11896: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11897: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 11898: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 11899: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11900: /* double ftolpl = 1.e-10; */
1.180 brouard 11901: double age, agebase, agelim;
1.203 brouard 11902: double tot;
1.180 brouard 11903:
1.202 brouard 11904: strcpy(filerespl,"PL_");
11905: strcat(filerespl,fileresu);
11906: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11907: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11908: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11909: }
1.288 brouard 11910: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11911: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11912: pstamp(ficrespl);
1.288 brouard 11913: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11914: fprintf(ficrespl,"#Age ");
11915: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11916: fprintf(ficrespl,"\n");
1.180 brouard 11917:
1.219 brouard 11918: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11919:
1.219 brouard 11920: agebase=ageminpar;
11921: agelim=agemaxpar;
1.180 brouard 11922:
1.227 brouard 11923: /* i1=pow(2,ncoveff); */
1.234 brouard 11924: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11925: if (cptcovn < 1){i1=1;}
1.180 brouard 11926:
1.337 brouard 11927: /* for(k=1; k<=i1;k++){ /\* For each combination k of dummy covariates in the model *\/ */
1.238 brouard 11928: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 11929: k=TKresult[nres];
1.338 brouard 11930: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 11931: /* if(i1 != 1 && TKresult[nres]!= k) /\* We found the combination k corresponding to the resultline value of dummies *\/ */
11932: /* continue; */
1.235 brouard 11933:
1.238 brouard 11934: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11935: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11936: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11937: /* k=k+1; */
11938: /* to clean */
1.332 brouard 11939: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11940: fprintf(ficrespl,"#******");
11941: printf("#******");
11942: fprintf(ficlog,"#******");
1.337 brouard 11943: 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 11944: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
1.337 brouard 11945: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11946: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11947: fprintf(ficrespl," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11948: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11949: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
11950: }
11951: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
11952: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11953: /* fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11954: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
11955: /* } */
1.238 brouard 11956: fprintf(ficrespl,"******\n");
11957: printf("******\n");
11958: fprintf(ficlog,"******\n");
11959: if(invalidvarcomb[k]){
11960: printf("\nCombination (%d) ignored because no case \n",k);
11961: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
11962: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
11963: continue;
11964: }
1.219 brouard 11965:
1.238 brouard 11966: fprintf(ficrespl,"#Age ");
1.337 brouard 11967: /* for(j=1;j<=cptcoveff;j++) { */
11968: /* fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11969: /* } */
11970: for(j=1;j<=cptcovs;j++) { /* New the quanti variable is added */
11971: fprintf(ficrespl,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11972: }
11973: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
11974: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 11975:
1.238 brouard 11976: for (age=agebase; age<=agelim; age++){
11977: /* for (age=agebase; age<=agebase; age++){ */
1.337 brouard 11978: /**< Computes the prevalence limit in each live state at age x and for covariate combination (k and) nres */
11979: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres); /* Nicely done */
1.238 brouard 11980: fprintf(ficrespl,"%.0f ",age );
1.337 brouard 11981: /* for(j=1;j<=cptcoveff;j++) */
11982: /* fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
11983: for(j=1;j<=cptcovs;j++)
11984: fprintf(ficrespl,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 11985: tot=0.;
11986: for(i=1; i<=nlstate;i++){
11987: tot += prlim[i][i];
11988: fprintf(ficrespl," %.5f", prlim[i][i]);
11989: }
11990: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
11991: } /* Age */
11992: /* was end of cptcod */
1.337 brouard 11993: } /* nres */
11994: /* } /\* for each combination *\/ */
1.219 brouard 11995: return 0;
1.180 brouard 11996: }
11997:
1.218 brouard 11998: 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 11999: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 12000:
12001: /* Computes the back prevalence limit for any combination of covariate values
12002: * at any age between ageminpar and agemaxpar
12003: */
1.235 brouard 12004: int i, j, k, i1, nres=0 ;
1.217 brouard 12005: /* double ftolpl = 1.e-10; */
12006: double age, agebase, agelim;
12007: double tot;
1.218 brouard 12008: /* double ***mobaverage; */
12009: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 12010:
12011: strcpy(fileresplb,"PLB_");
12012: strcat(fileresplb,fileresu);
12013: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 12014: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
12015: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 12016: }
1.288 brouard 12017: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
12018: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 12019: pstamp(ficresplb);
1.288 brouard 12020: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 12021: fprintf(ficresplb,"#Age ");
12022: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
12023: fprintf(ficresplb,"\n");
12024:
1.218 brouard 12025:
12026: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
12027:
12028: agebase=ageminpar;
12029: agelim=agemaxpar;
12030:
12031:
1.227 brouard 12032: i1=pow(2,cptcoveff);
1.218 brouard 12033: if (cptcovn < 1){i1=1;}
1.227 brouard 12034:
1.238 brouard 12035: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.338 brouard 12036: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
12037: k=TKresult[nres];
12038: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
12039: /* if(i1 != 1 && TKresult[nres]!= k) */
12040: /* continue; */
12041: /* /\*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*\/ */
1.238 brouard 12042: fprintf(ficresplb,"#******");
12043: printf("#******");
12044: fprintf(ficlog,"#******");
1.338 brouard 12045: 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) */
12046: printf(" V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12047: fprintf(ficresplb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12048: fprintf(ficlog," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12049: }
1.338 brouard 12050: /* for(j=1;j<=cptcoveff ;j++) {/\* all covariates *\/ */
12051: /* fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12052: /* printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12053: /* fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12054: /* } */
12055: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12056: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12057: /* fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12058: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12059: /* } */
1.238 brouard 12060: fprintf(ficresplb,"******\n");
12061: printf("******\n");
12062: fprintf(ficlog,"******\n");
12063: if(invalidvarcomb[k]){
12064: printf("\nCombination (%d) ignored because no cases \n",k);
12065: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
12066: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
12067: continue;
12068: }
1.218 brouard 12069:
1.238 brouard 12070: fprintf(ficresplb,"#Age ");
1.338 brouard 12071: for(j=1;j<=cptcovs;j++) {
12072: fprintf(ficresplb,"V%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12073: }
12074: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
12075: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 12076:
12077:
1.238 brouard 12078: for (age=agebase; age<=agelim; age++){
12079: /* for (age=agebase; age<=agebase; age++){ */
12080: if(mobilavproj > 0){
12081: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
12082: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 12083: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 12084: }else if (mobilavproj == 0){
12085: 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);
12086: 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);
12087: exit(1);
12088: }else{
12089: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 12090: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 12091: /* printf("TOTOT\n"); */
12092: /* exit(1); */
1.238 brouard 12093: }
12094: fprintf(ficresplb,"%.0f ",age );
1.338 brouard 12095: for(j=1;j<=cptcovs;j++)
12096: fprintf(ficresplb,"%d %lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.238 brouard 12097: tot=0.;
12098: for(i=1; i<=nlstate;i++){
12099: tot += bprlim[i][i];
12100: fprintf(ficresplb," %.5f", bprlim[i][i]);
12101: }
12102: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
12103: } /* Age */
12104: /* was end of cptcod */
1.255 brouard 12105: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.338 brouard 12106: /* } /\* end of any combination *\/ */
1.238 brouard 12107: } /* end of nres */
1.218 brouard 12108: /* hBijx(p, bage, fage); */
12109: /* fclose(ficrespijb); */
12110:
12111: return 0;
1.217 brouard 12112: }
1.218 brouard 12113:
1.180 brouard 12114: int hPijx(double *p, int bage, int fage){
12115: /*------------- h Pij x at various ages ------------*/
1.336 brouard 12116: /* to be optimized with precov */
1.180 brouard 12117: int stepsize;
12118: int agelim;
12119: int hstepm;
12120: int nhstepm;
1.235 brouard 12121: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 12122:
12123: double agedeb;
12124: double ***p3mat;
12125:
1.337 brouard 12126: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
12127: if((ficrespij=fopen(filerespij,"w"))==NULL) {
12128: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
12129: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
12130: }
12131: printf("Computing pij: result on file '%s' \n", filerespij);
12132: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
12133:
12134: stepsize=(int) (stepm+YEARM-1)/YEARM;
12135: /*if (stepm<=24) stepsize=2;*/
12136:
12137: agelim=AGESUP;
12138: hstepm=stepsize*YEARM; /* Every year of age */
12139: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
12140:
12141: /* hstepm=1; aff par mois*/
12142: pstamp(ficrespij);
12143: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
12144: i1= pow(2,cptcoveff);
12145: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12146: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12147: /* k=k+1; */
12148: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
12149: k=TKresult[nres];
1.338 brouard 12150: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12151: /* for(k=1; k<=i1;k++){ */
12152: /* if(i1 != 1 && TKresult[nres]!= k) */
12153: /* continue; */
12154: fprintf(ficrespij,"\n#****** ");
12155: for(j=1;j<=cptcovs;j++){
12156: fprintf(ficrespij," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
12157: /* fprintf(ficrespij,"@wV%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12158: /* for (k4=1; k4<= nsq; k4++){ /\* For each selected (single) quantitative value *\/ */
12159: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12160: /* fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
12161: }
12162: fprintf(ficrespij,"******\n");
12163:
12164: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
12165: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
12166: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
12167:
12168: /* nhstepm=nhstepm*YEARM; aff par mois*/
12169:
12170: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12171: oldm=oldms;savm=savms;
12172: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
12173: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
12174: for(i=1; i<=nlstate;i++)
12175: for(j=1; j<=nlstate+ndeath;j++)
12176: fprintf(ficrespij," %1d-%1d",i,j);
12177: fprintf(ficrespij,"\n");
12178: for (h=0; h<=nhstepm; h++){
12179: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12180: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.183 brouard 12181: for(i=1; i<=nlstate;i++)
12182: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12183: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.183 brouard 12184: fprintf(ficrespij,"\n");
12185: }
1.337 brouard 12186: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12187: fprintf(ficrespij,"\n");
1.180 brouard 12188: }
1.337 brouard 12189: }
12190: /*}*/
12191: return 0;
1.180 brouard 12192: }
1.218 brouard 12193:
12194: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 12195: /*------------- h Bij x at various ages ------------*/
1.336 brouard 12196: /* To be optimized with precov */
1.217 brouard 12197: int stepsize;
1.218 brouard 12198: /* int agelim; */
12199: int ageminl;
1.217 brouard 12200: int hstepm;
12201: int nhstepm;
1.238 brouard 12202: int h, i, i1, j, k, nres;
1.218 brouard 12203:
1.217 brouard 12204: double agedeb;
12205: double ***p3mat;
1.218 brouard 12206:
12207: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
12208: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
12209: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12210: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
12211: }
12212: printf("Computing pij back: result on file '%s' \n", filerespijb);
12213: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
12214:
12215: stepsize=(int) (stepm+YEARM-1)/YEARM;
12216: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 12217:
1.218 brouard 12218: /* agelim=AGESUP; */
1.289 brouard 12219: ageminl=AGEINF; /* was 30 */
1.218 brouard 12220: hstepm=stepsize*YEARM; /* Every year of age */
12221: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
12222:
12223: /* hstepm=1; aff par mois*/
12224: pstamp(ficrespijb);
1.255 brouard 12225: 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 12226: i1= pow(2,cptcoveff);
1.218 brouard 12227: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
12228: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
12229: /* k=k+1; */
1.238 brouard 12230: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.337 brouard 12231: k=TKresult[nres];
1.338 brouard 12232: if(TKresult[nres]==0) k=1; /* To be checked for noresult */
1.337 brouard 12233: /* for(k=1; k<=i1;k++){ /\* For any combination of dummy covariates, fixed and varying *\/ */
12234: /* if(i1 != 1 && TKresult[nres]!= k) */
12235: /* continue; */
12236: fprintf(ficrespijb,"\n#****** ");
12237: for(j=1;j<=cptcovs;j++){
1.338 brouard 12238: fprintf(ficrespijb," V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]);
1.337 brouard 12239: /* for(j=1;j<=cptcoveff;j++) */
12240: /* fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
12241: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
12242: /* fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
12243: }
12244: fprintf(ficrespijb,"******\n");
12245: if(invalidvarcomb[k]){ /* Is it necessary here? */
12246: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
12247: continue;
12248: }
12249:
12250: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
12251: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
12252: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
12253: 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 */
12254: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
12255:
12256: /* nhstepm=nhstepm*YEARM; aff par mois*/
12257:
12258: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
12259: /* and memory limitations if stepm is small */
12260:
12261: /* oldm=oldms;savm=savms; */
12262: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
12263: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
12264: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
12265: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
12266: for(i=1; i<=nlstate;i++)
12267: for(j=1; j<=nlstate+ndeath;j++)
12268: fprintf(ficrespijb," %1d-%1d",i,j);
12269: fprintf(ficrespijb,"\n");
12270: for (h=0; h<=nhstepm; h++){
12271: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
12272: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
12273: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
1.217 brouard 12274: for(i=1; i<=nlstate;i++)
12275: for(j=1; j<=nlstate+ndeath;j++)
1.337 brouard 12276: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.217 brouard 12277: fprintf(ficrespijb,"\n");
1.337 brouard 12278: }
12279: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
12280: fprintf(ficrespijb,"\n");
12281: } /* end age deb */
12282: /* } /\* end combination *\/ */
1.238 brouard 12283: } /* end nres */
1.218 brouard 12284: return 0;
12285: } /* hBijx */
1.217 brouard 12286:
1.180 brouard 12287:
1.136 brouard 12288: /***********************************************/
12289: /**************** Main Program *****************/
12290: /***********************************************/
12291:
12292: int main(int argc, char *argv[])
12293: {
12294: #ifdef GSL
12295: const gsl_multimin_fminimizer_type *T;
12296: size_t iteri = 0, it;
12297: int rval = GSL_CONTINUE;
12298: int status = GSL_SUCCESS;
12299: double ssval;
12300: #endif
12301: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 12302: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
12303: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 12304: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 12305: int jj, ll, li, lj, lk;
1.136 brouard 12306: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 12307: int num_filled;
1.136 brouard 12308: int itimes;
12309: int NDIM=2;
12310: int vpopbased=0;
1.235 brouard 12311: int nres=0;
1.258 brouard 12312: int endishere=0;
1.277 brouard 12313: int noffset=0;
1.274 brouard 12314: int ncurrv=0; /* Temporary variable */
12315:
1.164 brouard 12316: char ca[32], cb[32];
1.136 brouard 12317: /* FILE *fichtm; *//* Html File */
12318: /* FILE *ficgp;*/ /*Gnuplot File */
12319: struct stat info;
1.191 brouard 12320: double agedeb=0.;
1.194 brouard 12321:
12322: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 12323: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 12324:
1.165 brouard 12325: double fret;
1.191 brouard 12326: double dum=0.; /* Dummy variable */
1.136 brouard 12327: double ***p3mat;
1.218 brouard 12328: /* double ***mobaverage; */
1.319 brouard 12329: double wald;
1.164 brouard 12330:
12331: char line[MAXLINE];
1.197 brouard 12332: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
12333:
1.234 brouard 12334: char modeltemp[MAXLINE];
1.332 brouard 12335: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 12336:
1.136 brouard 12337: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 12338: char *tok, *val; /* pathtot */
1.334 brouard 12339: /* int firstobs=1, lastobs=10; /\* nobs = lastobs-firstobs declared globally ;*\/ */
1.195 brouard 12340: int c, h , cpt, c2;
1.191 brouard 12341: int jl=0;
12342: int i1, j1, jk, stepsize=0;
1.194 brouard 12343: int count=0;
12344:
1.164 brouard 12345: int *tab;
1.136 brouard 12346: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 12347: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
12348: /* double anprojf, mprojf, jprojf; */
12349: /* double jintmean,mintmean,aintmean; */
12350: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12351: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
12352: double yrfproj= 10.0; /* Number of years of forward projections */
12353: double yrbproj= 10.0; /* Number of years of backward projections */
12354: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 12355: int mobilav=0,popforecast=0;
1.191 brouard 12356: int hstepm=0, nhstepm=0;
1.136 brouard 12357: int agemortsup;
12358: float sumlpop=0.;
12359: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
12360: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
12361:
1.191 brouard 12362: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 12363: double ftolpl=FTOL;
12364: double **prlim;
1.217 brouard 12365: double **bprlim;
1.317 brouard 12366: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
12367: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 12368: double ***paramstart; /* Matrix of starting parameter values */
12369: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 12370: double **matcov; /* Matrix of covariance */
1.203 brouard 12371: double **hess; /* Hessian matrix */
1.136 brouard 12372: double ***delti3; /* Scale */
12373: double *delti; /* Scale */
12374: double ***eij, ***vareij;
12375: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 12376:
1.136 brouard 12377: double *epj, vepp;
1.164 brouard 12378:
1.273 brouard 12379: double dateprev1, dateprev2;
1.296 brouard 12380: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
12381: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
12382:
1.217 brouard 12383:
1.136 brouard 12384: double **ximort;
1.145 brouard 12385: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 12386: int *dcwave;
12387:
1.164 brouard 12388: char z[1]="c";
1.136 brouard 12389:
12390: /*char *strt;*/
12391: char strtend[80];
1.126 brouard 12392:
1.164 brouard 12393:
1.126 brouard 12394: /* setlocale (LC_ALL, ""); */
12395: /* bindtextdomain (PACKAGE, LOCALEDIR); */
12396: /* textdomain (PACKAGE); */
12397: /* setlocale (LC_CTYPE, ""); */
12398: /* setlocale (LC_MESSAGES, ""); */
12399:
12400: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 12401: rstart_time = time(NULL);
12402: /* (void) gettimeofday(&start_time,&tzp);*/
12403: start_time = *localtime(&rstart_time);
1.126 brouard 12404: curr_time=start_time;
1.157 brouard 12405: /*tml = *localtime(&start_time.tm_sec);*/
12406: /* strcpy(strstart,asctime(&tml)); */
12407: strcpy(strstart,asctime(&start_time));
1.126 brouard 12408:
12409: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 12410: /* tp.tm_sec = tp.tm_sec +86400; */
12411: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 12412: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
12413: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
12414: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 12415: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 12416: /* strt=asctime(&tmg); */
12417: /* printf("Time(after) =%s",strstart); */
12418: /* (void) time (&time_value);
12419: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
12420: * tm = *localtime(&time_value);
12421: * strstart=asctime(&tm);
12422: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
12423: */
12424:
12425: nberr=0; /* Number of errors and warnings */
12426: nbwarn=0;
1.184 brouard 12427: #ifdef WIN32
12428: _getcwd(pathcd, size);
12429: #else
1.126 brouard 12430: getcwd(pathcd, size);
1.184 brouard 12431: #endif
1.191 brouard 12432: syscompilerinfo(0);
1.196 brouard 12433: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 12434: if(argc <=1){
12435: printf("\nEnter the parameter file name: ");
1.205 brouard 12436: if(!fgets(pathr,FILENAMELENGTH,stdin)){
12437: printf("ERROR Empty parameter file name\n");
12438: goto end;
12439: }
1.126 brouard 12440: i=strlen(pathr);
12441: if(pathr[i-1]=='\n')
12442: pathr[i-1]='\0';
1.156 brouard 12443: i=strlen(pathr);
1.205 brouard 12444: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 12445: pathr[i-1]='\0';
1.205 brouard 12446: }
12447: i=strlen(pathr);
12448: if( i==0 ){
12449: printf("ERROR Empty parameter file name\n");
12450: goto end;
12451: }
12452: for (tok = pathr; tok != NULL; ){
1.126 brouard 12453: printf("Pathr |%s|\n",pathr);
12454: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
12455: printf("val= |%s| pathr=%s\n",val,pathr);
12456: strcpy (pathtot, val);
12457: if(pathr[0] == '\0') break; /* Dirty */
12458: }
12459: }
1.281 brouard 12460: else if (argc<=2){
12461: strcpy(pathtot,argv[1]);
12462: }
1.126 brouard 12463: else{
12464: strcpy(pathtot,argv[1]);
1.281 brouard 12465: strcpy(z,argv[2]);
12466: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 12467: }
12468: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
12469: /*cygwin_split_path(pathtot,path,optionfile);
12470: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
12471: /* cutv(path,optionfile,pathtot,'\\');*/
12472:
12473: /* Split argv[0], imach program to get pathimach */
12474: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
12475: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12476: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
12477: /* strcpy(pathimach,argv[0]); */
12478: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
12479: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
12480: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 12481: #ifdef WIN32
12482: _chdir(path); /* Can be a relative path */
12483: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
12484: #else
1.126 brouard 12485: chdir(path); /* Can be a relative path */
1.184 brouard 12486: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
12487: #endif
12488: printf("Current directory %s!\n",pathcd);
1.126 brouard 12489: strcpy(command,"mkdir ");
12490: strcat(command,optionfilefiname);
12491: if((outcmd=system(command)) != 0){
1.169 brouard 12492: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 12493: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
12494: /* fclose(ficlog); */
12495: /* exit(1); */
12496: }
12497: /* if((imk=mkdir(optionfilefiname))<0){ */
12498: /* perror("mkdir"); */
12499: /* } */
12500:
12501: /*-------- arguments in the command line --------*/
12502:
1.186 brouard 12503: /* Main Log file */
1.126 brouard 12504: strcat(filelog, optionfilefiname);
12505: strcat(filelog,".log"); /* */
12506: if((ficlog=fopen(filelog,"w"))==NULL) {
12507: printf("Problem with logfile %s\n",filelog);
12508: goto end;
12509: }
12510: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 12511: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 12512: fprintf(ficlog,"\nEnter the parameter file name: \n");
12513: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
12514: path=%s \n\
12515: optionfile=%s\n\
12516: optionfilext=%s\n\
1.156 brouard 12517: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 12518:
1.197 brouard 12519: syscompilerinfo(1);
1.167 brouard 12520:
1.126 brouard 12521: printf("Local time (at start):%s",strstart);
12522: fprintf(ficlog,"Local time (at start): %s",strstart);
12523: fflush(ficlog);
12524: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 12525: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 12526:
12527: /* */
12528: strcpy(fileres,"r");
12529: strcat(fileres, optionfilefiname);
1.201 brouard 12530: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 12531: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 12532: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 12533:
1.186 brouard 12534: /* Main ---------arguments file --------*/
1.126 brouard 12535:
12536: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 12537: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
12538: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 12539: fflush(ficlog);
1.149 brouard 12540: /* goto end; */
12541: exit(70);
1.126 brouard 12542: }
12543:
12544: strcpy(filereso,"o");
1.201 brouard 12545: strcat(filereso,fileresu);
1.126 brouard 12546: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
12547: printf("Problem with Output resultfile: %s\n", filereso);
12548: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
12549: fflush(ficlog);
12550: goto end;
12551: }
1.278 brouard 12552: /*-------- Rewriting parameter file ----------*/
12553: strcpy(rfileres,"r"); /* "Rparameterfile */
12554: strcat(rfileres,optionfilefiname); /* Parameter file first name */
12555: strcat(rfileres,"."); /* */
12556: strcat(rfileres,optionfilext); /* Other files have txt extension */
12557: if((ficres =fopen(rfileres,"w"))==NULL) {
12558: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
12559: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
12560: fflush(ficlog);
12561: goto end;
12562: }
12563: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 12564:
1.278 brouard 12565:
1.126 brouard 12566: /* Reads comments: lines beginning with '#' */
12567: numlinepar=0;
1.277 brouard 12568: /* Is it a BOM UTF-8 Windows file? */
12569: /* First parameter line */
1.197 brouard 12570: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 12571: noffset=0;
12572: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
12573: {
12574: noffset=noffset+3;
12575: printf("# File is an UTF8 Bom.\n"); // 0xBF
12576: }
1.302 brouard 12577: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
12578: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 12579: {
12580: noffset=noffset+2;
12581: printf("# File is an UTF16BE BOM file\n");
12582: }
12583: else if( line[0] == 0 && line[1] == 0)
12584: {
12585: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
12586: noffset=noffset+4;
12587: printf("# File is an UTF16BE BOM file\n");
12588: }
12589: } else{
12590: ;/*printf(" Not a BOM file\n");*/
12591: }
12592:
1.197 brouard 12593: /* If line starts with a # it is a comment */
1.277 brouard 12594: if (line[noffset] == '#') {
1.197 brouard 12595: numlinepar++;
12596: fputs(line,stdout);
12597: fputs(line,ficparo);
1.278 brouard 12598: fputs(line,ficres);
1.197 brouard 12599: fputs(line,ficlog);
12600: continue;
12601: }else
12602: break;
12603: }
12604: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
12605: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
12606: if (num_filled != 5) {
12607: printf("Should be 5 parameters\n");
1.283 brouard 12608: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 12609: }
1.126 brouard 12610: numlinepar++;
1.197 brouard 12611: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 12612: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12613: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
12614: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 12615: }
12616: /* Second parameter line */
12617: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 12618: /* while(fscanf(ficpar,"%[^\n]", line)) { */
12619: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 12620: if (line[0] == '#') {
12621: numlinepar++;
1.283 brouard 12622: printf("%s",line);
12623: fprintf(ficres,"%s",line);
12624: fprintf(ficparo,"%s",line);
12625: fprintf(ficlog,"%s",line);
1.197 brouard 12626: continue;
12627: }else
12628: break;
12629: }
1.223 brouard 12630: 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", \
12631: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
12632: if (num_filled != 11) {
12633: 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 12634: printf("but line=%s\n",line);
1.283 brouard 12635: 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");
12636: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 12637: }
1.286 brouard 12638: if( lastpass > maxwav){
12639: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12640: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12641: fflush(ficlog);
12642: goto end;
12643: }
12644: 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 12645: 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 12646: 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 12647: 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 12648: }
1.203 brouard 12649: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 12650: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 12651: /* Third parameter line */
12652: while(fgets(line, MAXLINE, ficpar)) {
12653: /* If line starts with a # it is a comment */
12654: if (line[0] == '#') {
12655: numlinepar++;
1.283 brouard 12656: printf("%s",line);
12657: fprintf(ficres,"%s",line);
12658: fprintf(ficparo,"%s",line);
12659: fprintf(ficlog,"%s",line);
1.197 brouard 12660: continue;
12661: }else
12662: break;
12663: }
1.201 brouard 12664: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 12665: if (num_filled != 1){
1.302 brouard 12666: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
12667: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 12668: model[0]='\0';
12669: goto end;
12670: }
12671: else{
12672: if (model[0]=='+'){
12673: for(i=1; i<=strlen(model);i++)
12674: modeltemp[i-1]=model[i];
1.201 brouard 12675: strcpy(model,modeltemp);
1.197 brouard 12676: }
12677: }
1.338 brouard 12678: /* printf(" model=1+age%s modeltemp= %s, model=1+age+%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 12679: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 12680: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
12681: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
12682: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 12683: }
12684: /* 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); */
12685: /* numlinepar=numlinepar+3; /\* In general *\/ */
12686: /* 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 12687: /* 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); */
12688: /* 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 12689: fflush(ficlog);
1.190 brouard 12690: /* if(model[0]=='#'|| model[0]== '\0'){ */
12691: if(model[0]=='#'){
1.279 brouard 12692: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
12693: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
12694: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 12695: if(mle != -1){
1.279 brouard 12696: 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 12697: exit(1);
12698: }
12699: }
1.126 brouard 12700: while((c=getc(ficpar))=='#' && c!= EOF){
12701: ungetc(c,ficpar);
12702: fgets(line, MAXLINE, ficpar);
12703: numlinepar++;
1.195 brouard 12704: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
12705: z[0]=line[1];
1.342 brouard 12706: }else if(line[1]=='d'){ /* For debugging individual values of covariates in ficresilk */
1.343 brouard 12707: debugILK=1;printf("DebugILK\n");
1.195 brouard 12708: }
12709: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 12710: fputs(line, stdout);
12711: //puts(line);
1.126 brouard 12712: fputs(line,ficparo);
12713: fputs(line,ficlog);
12714: }
12715: ungetc(c,ficpar);
12716:
12717:
1.290 brouard 12718: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
12719: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
12720: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
1.341 brouard 12721: /* if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /\**< Time varying covariate (dummy and quantitative)*\/ */
12722: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs); /**< Might be better */
1.136 brouard 12723: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
12724: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
12725: v1+v2*age+v2*v3 makes cptcovn = 3
12726: */
12727: if (strlen(model)>1)
1.187 brouard 12728: 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 12729: else
1.187 brouard 12730: ncovmodel=2; /* Constant and age */
1.133 brouard 12731: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
12732: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 12733: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
12734: 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);
12735: 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);
12736: fflush(stdout);
12737: fclose (ficlog);
12738: goto end;
12739: }
1.126 brouard 12740: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12741: delti=delti3[1][1];
12742: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
12743: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 12744: /* We could also provide initial parameters values giving by simple logistic regression
12745: * only one way, that is without matrix product. We will have nlstate maximizations */
12746: /* for(i=1;i<nlstate;i++){ */
12747: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12748: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12749: /* } */
1.126 brouard 12750: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 12751: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
12752: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12753: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12754: fclose (ficparo);
12755: fclose (ficlog);
12756: goto end;
12757: exit(0);
1.220 brouard 12758: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 12759: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 12760: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
12761: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12762: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12763: matcov=matrix(1,npar,1,npar);
1.203 brouard 12764: hess=matrix(1,npar,1,npar);
1.220 brouard 12765: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 12766: /* Read guessed parameters */
1.126 brouard 12767: /* Reads comments: lines beginning with '#' */
12768: while((c=getc(ficpar))=='#' && c!= EOF){
12769: ungetc(c,ficpar);
12770: fgets(line, MAXLINE, ficpar);
12771: numlinepar++;
1.141 brouard 12772: fputs(line,stdout);
1.126 brouard 12773: fputs(line,ficparo);
12774: fputs(line,ficlog);
12775: }
12776: ungetc(c,ficpar);
12777:
12778: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 12779: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 12780: for(i=1; i <=nlstate; i++){
1.234 brouard 12781: j=0;
1.126 brouard 12782: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 12783: if(jj==i) continue;
12784: j++;
1.292 brouard 12785: while((c=getc(ficpar))=='#' && c!= EOF){
12786: ungetc(c,ficpar);
12787: fgets(line, MAXLINE, ficpar);
12788: numlinepar++;
12789: fputs(line,stdout);
12790: fputs(line,ficparo);
12791: fputs(line,ficlog);
12792: }
12793: ungetc(c,ficpar);
1.234 brouard 12794: fscanf(ficpar,"%1d%1d",&i1,&j1);
12795: if ((i1 != i) || (j1 != jj)){
12796: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 12797: It might be a problem of design; if ncovcol and the model are correct\n \
12798: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 12799: exit(1);
12800: }
12801: fprintf(ficparo,"%1d%1d",i1,j1);
12802: if(mle==1)
12803: printf("%1d%1d",i,jj);
12804: fprintf(ficlog,"%1d%1d",i,jj);
12805: for(k=1; k<=ncovmodel;k++){
12806: fscanf(ficpar," %lf",¶m[i][j][k]);
12807: if(mle==1){
12808: printf(" %lf",param[i][j][k]);
12809: fprintf(ficlog," %lf",param[i][j][k]);
12810: }
12811: else
12812: fprintf(ficlog," %lf",param[i][j][k]);
12813: fprintf(ficparo," %lf",param[i][j][k]);
12814: }
12815: fscanf(ficpar,"\n");
12816: numlinepar++;
12817: if(mle==1)
12818: printf("\n");
12819: fprintf(ficlog,"\n");
12820: fprintf(ficparo,"\n");
1.126 brouard 12821: }
12822: }
12823: fflush(ficlog);
1.234 brouard 12824:
1.251 brouard 12825: /* Reads parameters values */
1.126 brouard 12826: p=param[1][1];
1.251 brouard 12827: pstart=paramstart[1][1];
1.126 brouard 12828:
12829: /* Reads comments: lines beginning with '#' */
12830: while((c=getc(ficpar))=='#' && c!= EOF){
12831: ungetc(c,ficpar);
12832: fgets(line, MAXLINE, ficpar);
12833: numlinepar++;
1.141 brouard 12834: fputs(line,stdout);
1.126 brouard 12835: fputs(line,ficparo);
12836: fputs(line,ficlog);
12837: }
12838: ungetc(c,ficpar);
12839:
12840: for(i=1; i <=nlstate; i++){
12841: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 12842: fscanf(ficpar,"%1d%1d",&i1,&j1);
12843: if ( (i1-i) * (j1-j) != 0){
12844: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
12845: exit(1);
12846: }
12847: printf("%1d%1d",i,j);
12848: fprintf(ficparo,"%1d%1d",i1,j1);
12849: fprintf(ficlog,"%1d%1d",i1,j1);
12850: for(k=1; k<=ncovmodel;k++){
12851: fscanf(ficpar,"%le",&delti3[i][j][k]);
12852: printf(" %le",delti3[i][j][k]);
12853: fprintf(ficparo," %le",delti3[i][j][k]);
12854: fprintf(ficlog," %le",delti3[i][j][k]);
12855: }
12856: fscanf(ficpar,"\n");
12857: numlinepar++;
12858: printf("\n");
12859: fprintf(ficparo,"\n");
12860: fprintf(ficlog,"\n");
1.126 brouard 12861: }
12862: }
12863: fflush(ficlog);
1.234 brouard 12864:
1.145 brouard 12865: /* Reads covariance matrix */
1.126 brouard 12866: delti=delti3[1][1];
1.220 brouard 12867:
12868:
1.126 brouard 12869: /* 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 12870:
1.126 brouard 12871: /* Reads comments: lines beginning with '#' */
12872: while((c=getc(ficpar))=='#' && c!= EOF){
12873: ungetc(c,ficpar);
12874: fgets(line, MAXLINE, ficpar);
12875: numlinepar++;
1.141 brouard 12876: fputs(line,stdout);
1.126 brouard 12877: fputs(line,ficparo);
12878: fputs(line,ficlog);
12879: }
12880: ungetc(c,ficpar);
1.220 brouard 12881:
1.126 brouard 12882: matcov=matrix(1,npar,1,npar);
1.203 brouard 12883: hess=matrix(1,npar,1,npar);
1.131 brouard 12884: for(i=1; i <=npar; i++)
12885: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12886:
1.194 brouard 12887: /* Scans npar lines */
1.126 brouard 12888: for(i=1; i <=npar; i++){
1.226 brouard 12889: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12890: if(count != 3){
1.226 brouard 12891: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12892: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12893: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12894: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12895: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12896: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12897: exit(1);
1.220 brouard 12898: }else{
1.226 brouard 12899: if(mle==1)
12900: printf("%1d%1d%d",i1,j1,jk);
12901: }
12902: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12903: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12904: for(j=1; j <=i; j++){
1.226 brouard 12905: fscanf(ficpar," %le",&matcov[i][j]);
12906: if(mle==1){
12907: printf(" %.5le",matcov[i][j]);
12908: }
12909: fprintf(ficlog," %.5le",matcov[i][j]);
12910: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12911: }
12912: fscanf(ficpar,"\n");
12913: numlinepar++;
12914: if(mle==1)
1.220 brouard 12915: printf("\n");
1.126 brouard 12916: fprintf(ficlog,"\n");
12917: fprintf(ficparo,"\n");
12918: }
1.194 brouard 12919: /* End of read covariance matrix npar lines */
1.126 brouard 12920: for(i=1; i <=npar; i++)
12921: for(j=i+1;j<=npar;j++)
1.226 brouard 12922: matcov[i][j]=matcov[j][i];
1.126 brouard 12923:
12924: if(mle==1)
12925: printf("\n");
12926: fprintf(ficlog,"\n");
12927:
12928: fflush(ficlog);
12929:
12930: } /* End of mle != -3 */
1.218 brouard 12931:
1.186 brouard 12932: /* Main data
12933: */
1.290 brouard 12934: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12935: /* num=lvector(1,n); */
12936: /* moisnais=vector(1,n); */
12937: /* annais=vector(1,n); */
12938: /* moisdc=vector(1,n); */
12939: /* andc=vector(1,n); */
12940: /* weight=vector(1,n); */
12941: /* agedc=vector(1,n); */
12942: /* cod=ivector(1,n); */
12943: /* for(i=1;i<=n;i++){ */
12944: num=lvector(firstobs,lastobs);
12945: moisnais=vector(firstobs,lastobs);
12946: annais=vector(firstobs,lastobs);
12947: moisdc=vector(firstobs,lastobs);
12948: andc=vector(firstobs,lastobs);
12949: weight=vector(firstobs,lastobs);
12950: agedc=vector(firstobs,lastobs);
12951: cod=ivector(firstobs,lastobs);
12952: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 12953: num[i]=0;
12954: moisnais[i]=0;
12955: annais[i]=0;
12956: moisdc[i]=0;
12957: andc[i]=0;
12958: agedc[i]=0;
12959: cod[i]=0;
12960: weight[i]=1.0; /* Equal weights, 1 by default */
12961: }
1.290 brouard 12962: mint=matrix(1,maxwav,firstobs,lastobs);
12963: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 12964: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.336 brouard 12965: /* printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel)); */
1.126 brouard 12966: tab=ivector(1,NCOVMAX);
1.144 brouard 12967: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 12968: 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 12969:
1.136 brouard 12970: /* Reads data from file datafile */
12971: if (readdata(datafile, firstobs, lastobs, &imx)==1)
12972: goto end;
12973:
12974: /* Calculation of the number of parameters from char model */
1.234 brouard 12975: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 12976: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
12977: k=3 V4 Tvar[k=3]= 4 (from V4)
12978: k=2 V1 Tvar[k=2]= 1 (from V1)
12979: k=1 Tvar[1]=2 (from V2)
1.234 brouard 12980: */
12981:
12982: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
12983: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 12984: TnsdVar=ivector(1,NCOVMAX); /* */
1.335 brouard 12985: /* for(i=1; i<=NCOVMAX;i++) TnsdVar[i]=3; */
1.234 brouard 12986: TvarsD=ivector(1,NCOVMAX); /* */
12987: TvarsQind=ivector(1,NCOVMAX); /* */
12988: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 12989: TvarF=ivector(1,NCOVMAX); /* */
12990: TvarFind=ivector(1,NCOVMAX); /* */
12991: TvarV=ivector(1,NCOVMAX); /* */
12992: TvarVind=ivector(1,NCOVMAX); /* */
12993: TvarA=ivector(1,NCOVMAX); /* */
12994: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 12995: TvarFD=ivector(1,NCOVMAX); /* */
12996: TvarFDind=ivector(1,NCOVMAX); /* */
12997: TvarFQ=ivector(1,NCOVMAX); /* */
12998: TvarFQind=ivector(1,NCOVMAX); /* */
12999: TvarVD=ivector(1,NCOVMAX); /* */
13000: TvarVDind=ivector(1,NCOVMAX); /* */
13001: TvarVQ=ivector(1,NCOVMAX); /* */
13002: TvarVQind=ivector(1,NCOVMAX); /* */
1.339 brouard 13003: TvarVV=ivector(1,NCOVMAX); /* */
13004: TvarVVind=ivector(1,NCOVMAX); /* */
1.231 brouard 13005:
1.230 brouard 13006: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 13007: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 13008: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
13009: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
13010: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 13011: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
13012: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
13013: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
13014: */
13015: /* For model-covariate k tells which data-covariate to use but
13016: because this model-covariate is a construction we invent a new column
13017: ncovcol + k1
13018: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
13019: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 13020: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
13021: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 13022: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
13023: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 13024: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 13025: */
1.145 brouard 13026: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
13027: 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 13028: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
13029: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 brouard 13030: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 13031: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 13032: 4 covariates (3 plus signs)
13033: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 13034: */
13035: for(i=1;i<NCOVMAX;i++)
13036: Tage[i]=0;
1.230 brouard 13037: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 13038: * individual dummy, fixed or varying:
13039: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
13040: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 13041: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
13042: * V1 df, V2 qf, V3 & V4 dv, V5 qv
13043: * Tmodelind[1]@9={9,0,3,2,}*/
13044: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
13045: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 13046: * individual quantitative, fixed or varying:
13047: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
13048: * 3, 1, 0, 0, 0, 0, 0, 0},
13049: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 13050: /* Main decodemodel */
13051:
1.187 brouard 13052:
1.223 brouard 13053: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 13054: goto end;
13055:
1.137 brouard 13056: if((double)(lastobs-imx)/(double)imx > 1.10){
13057: nbwarn++;
13058: 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);
13059: 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);
13060: }
1.136 brouard 13061: /* if(mle==1){*/
1.137 brouard 13062: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
13063: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 13064: }
13065:
13066: /*-calculation of age at interview from date of interview and age at death -*/
13067: agev=matrix(1,maxwav,1,imx);
13068:
13069: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
13070: goto end;
13071:
1.126 brouard 13072:
1.136 brouard 13073: agegomp=(int)agemin;
1.290 brouard 13074: free_vector(moisnais,firstobs,lastobs);
13075: free_vector(annais,firstobs,lastobs);
1.126 brouard 13076: /* free_matrix(mint,1,maxwav,1,n);
13077: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 13078: /* free_vector(moisdc,1,n); */
13079: /* free_vector(andc,1,n); */
1.145 brouard 13080: /* */
13081:
1.126 brouard 13082: wav=ivector(1,imx);
1.214 brouard 13083: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
13084: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
13085: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
13086: 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.*/
13087: bh=imatrix(1,lastpass-firstpass+2,1,imx);
13088: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 13089:
13090: /* Concatenates waves */
1.214 brouard 13091: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
13092: Death is a valid wave (if date is known).
13093: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
13094: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
13095: and mw[mi+1][i]. dh depends on stepm.
13096: */
13097:
1.126 brouard 13098: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 13099: /* Concatenates waves */
1.145 brouard 13100:
1.290 brouard 13101: free_vector(moisdc,firstobs,lastobs);
13102: free_vector(andc,firstobs,lastobs);
1.215 brouard 13103:
1.126 brouard 13104: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
13105: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
13106: ncodemax[1]=1;
1.145 brouard 13107: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 13108: cptcoveff=0;
1.220 brouard 13109: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
1.335 brouard 13110: 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 13111: }
13112:
13113: ncovcombmax=pow(2,cptcoveff);
1.338 brouard 13114: invalidvarcomb=ivector(0, ncovcombmax);
13115: for(i=0;i<ncovcombmax;i++)
1.227 brouard 13116: invalidvarcomb[i]=0;
13117:
1.211 brouard 13118: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 13119: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 13120: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 13121:
1.200 brouard 13122: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 13123: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 13124: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 13125: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
13126: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
13127: * (currently 0 or 1) in the data.
13128: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
13129: * corresponding modality (h,j).
13130: */
13131:
1.145 brouard 13132: h=0;
13133: /*if (cptcovn > 0) */
1.126 brouard 13134: m=pow(2,cptcoveff);
13135:
1.144 brouard 13136: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 13137: * For k=4 covariates, h goes from 1 to m=2**k
13138: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
13139: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 13140: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
13141: *______________________________ *______________________
13142: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
13143: * 2 2 1 1 1 * 1 0 0 0 1
13144: * 3 i=2 1 2 1 1 * 2 0 0 1 0
13145: * 4 2 2 1 1 * 3 0 0 1 1
13146: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
13147: * 6 2 1 2 1 * 5 0 1 0 1
13148: * 7 i=4 1 2 2 1 * 6 0 1 1 0
13149: * 8 2 2 2 1 * 7 0 1 1 1
13150: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
13151: * 10 2 1 1 2 * 9 1 0 0 1
13152: * 11 i=6 1 2 1 2 * 10 1 0 1 0
13153: * 12 2 2 1 2 * 11 1 0 1 1
13154: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
13155: * 14 2 1 2 2 * 13 1 1 0 1
13156: * 15 i=8 1 2 2 2 * 14 1 1 1 0
13157: * 16 2 2 2 2 * 15 1 1 1 1
13158: */
1.212 brouard 13159: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 13160: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
13161: * and the value of each covariate?
13162: * V1=1, V2=1, V3=2, V4=1 ?
13163: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
13164: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
13165: * In order to get the real value in the data, we use nbcode
13166: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
13167: * We are keeping this crazy system in order to be able (in the future?)
13168: * to have more than 2 values (0 or 1) for a covariate.
13169: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
13170: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
13171: * bbbbbbbb
13172: * 76543210
13173: * h-1 00000101 (6-1=5)
1.219 brouard 13174: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 13175: * &
13176: * 1 00000001 (1)
1.219 brouard 13177: * 00000000 = 1 & ((h-1) >> (k-1))
13178: * +1= 00000001 =1
1.211 brouard 13179: *
13180: * h=14, k=3 => h'=h-1=13, k'=k-1=2
13181: * h' 1101 =2^3+2^2+0x2^1+2^0
13182: * >>k' 11
13183: * & 00000001
13184: * = 00000001
13185: * +1 = 00000010=2 = codtabm(14,3)
13186: * Reverse h=6 and m=16?
13187: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
13188: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
13189: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
13190: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
13191: * V3=decodtabm(14,3,2**4)=2
13192: * h'=13 1101 =2^3+2^2+0x2^1+2^0
13193: *(h-1) >> (j-1) 0011 =13 >> 2
13194: * &1 000000001
13195: * = 000000001
13196: * +1= 000000010 =2
13197: * 2211
13198: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
13199: * V3=2
1.220 brouard 13200: * codtabm and decodtabm are identical
1.211 brouard 13201: */
13202:
1.145 brouard 13203:
13204: free_ivector(Ndum,-1,NCOVMAX);
13205:
13206:
1.126 brouard 13207:
1.186 brouard 13208: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 13209: strcpy(optionfilegnuplot,optionfilefiname);
13210: if(mle==-3)
1.201 brouard 13211: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 13212: strcat(optionfilegnuplot,".gp");
13213:
13214: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
13215: printf("Problem with file %s",optionfilegnuplot);
13216: }
13217: else{
1.204 brouard 13218: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 13219: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 13220: //fprintf(ficgp,"set missing 'NaNq'\n");
13221: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 13222: }
13223: /* fclose(ficgp);*/
1.186 brouard 13224:
13225:
13226: /* Initialisation of --------- index.htm --------*/
1.126 brouard 13227:
13228: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
13229: if(mle==-3)
1.201 brouard 13230: strcat(optionfilehtm,"-MORT_");
1.126 brouard 13231: strcat(optionfilehtm,".htm");
13232: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 13233: printf("Problem with %s \n",optionfilehtm);
13234: exit(0);
1.126 brouard 13235: }
13236:
13237: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
13238: strcat(optionfilehtmcov,"-cov.htm");
13239: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
13240: printf("Problem with %s \n",optionfilehtmcov), exit(0);
13241: }
13242: else{
13243: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
13244: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13245: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 13246: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
13247: }
13248:
1.335 brouard 13249: fprintf(fichtm,"<html><head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n\
13250: <title>IMaCh %s</title></head>\n\
13251: <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n\
13252: <font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>\
13253: -EUROREVES-Institut de longévité-2013-2022-Japan Society for the Promotion of Sciences 日本学術振興会 \
13254: (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - \
13255: <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \n", optionfilehtm);
13256:
13257: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 13258: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 13259: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.337 brouard 13260: 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 13261: \n\
13262: <hr size=\"2\" color=\"#EC5E5E\">\
13263: <ul><li><h4>Parameter files</h4>\n\
13264: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
13265: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
13266: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
13267: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
13268: - Date and time at start: %s</ul>\n",\
1.335 brouard 13269: version,fullversion,optionfilehtm,optionfilehtm,title,datafile,datafile,firstpass,lastpass,stepm, weightopt, model, \
1.126 brouard 13270: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
13271: fileres,fileres,\
13272: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
13273: fflush(fichtm);
13274:
13275: strcpy(pathr,path);
13276: strcat(pathr,optionfilefiname);
1.184 brouard 13277: #ifdef WIN32
13278: _chdir(optionfilefiname); /* Move to directory named optionfile */
13279: #else
1.126 brouard 13280: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 13281: #endif
13282:
1.126 brouard 13283:
1.220 brouard 13284: /* Calculates basic frequencies. Computes observed prevalence at single age
13285: and for any valid combination of covariates
1.126 brouard 13286: and prints on file fileres'p'. */
1.251 brouard 13287: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 13288: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 13289:
13290: fprintf(fichtm,"\n");
1.286 brouard 13291: 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 13292: ftol, stepm);
13293: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
13294: ncurrv=1;
13295: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
13296: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
13297: ncurrv=i;
13298: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13299: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 13300: ncurrv=i;
13301: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 13302: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 13303: ncurrv=i;
13304: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
13305: 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", \
13306: nlstate, ndeath, maxwav, mle, weightopt);
13307:
13308: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
13309: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
13310:
13311:
1.317 brouard 13312: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 13313: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
13314: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 13315: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 13316: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 13317: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13318: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13319: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
13320: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 13321:
1.126 brouard 13322: /* For Powell, parameters are in a vector p[] starting at p[1]
13323: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
13324: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
13325:
13326: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 13327: /* For mortality only */
1.126 brouard 13328: if (mle==-3){
1.136 brouard 13329: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 13330: for(i=1;i<=NDIM;i++)
13331: for(j=1;j<=NDIM;j++)
13332: ximort[i][j]=0.;
1.186 brouard 13333: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 13334: cens=ivector(firstobs,lastobs);
13335: ageexmed=vector(firstobs,lastobs);
13336: agecens=vector(firstobs,lastobs);
13337: dcwave=ivector(firstobs,lastobs);
1.223 brouard 13338:
1.126 brouard 13339: for (i=1; i<=imx; i++){
13340: dcwave[i]=-1;
13341: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 13342: if (s[m][i]>nlstate) {
13343: dcwave[i]=m;
13344: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
13345: break;
13346: }
1.126 brouard 13347: }
1.226 brouard 13348:
1.126 brouard 13349: for (i=1; i<=imx; i++) {
13350: if (wav[i]>0){
1.226 brouard 13351: ageexmed[i]=agev[mw[1][i]][i];
13352: j=wav[i];
13353: agecens[i]=1.;
13354:
13355: if (ageexmed[i]> 1 && wav[i] > 0){
13356: agecens[i]=agev[mw[j][i]][i];
13357: cens[i]= 1;
13358: }else if (ageexmed[i]< 1)
13359: cens[i]= -1;
13360: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
13361: cens[i]=0 ;
1.126 brouard 13362: }
13363: else cens[i]=-1;
13364: }
13365:
13366: for (i=1;i<=NDIM;i++) {
13367: for (j=1;j<=NDIM;j++)
1.226 brouard 13368: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 13369: }
13370:
1.302 brouard 13371: p[1]=0.0268; p[NDIM]=0.083;
13372: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 13373:
13374:
1.136 brouard 13375: #ifdef GSL
13376: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 13377: #else
1.126 brouard 13378: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 13379: #endif
1.201 brouard 13380: strcpy(filerespow,"POW-MORT_");
13381: strcat(filerespow,fileresu);
1.126 brouard 13382: if((ficrespow=fopen(filerespow,"w"))==NULL) {
13383: printf("Problem with resultfile: %s\n", filerespow);
13384: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
13385: }
1.136 brouard 13386: #ifdef GSL
13387: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 13388: #else
1.126 brouard 13389: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 13390: #endif
1.126 brouard 13391: /* for (i=1;i<=nlstate;i++)
13392: for(j=1;j<=nlstate+ndeath;j++)
13393: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
13394: */
13395: fprintf(ficrespow,"\n");
1.136 brouard 13396: #ifdef GSL
13397: /* gsl starts here */
13398: T = gsl_multimin_fminimizer_nmsimplex;
13399: gsl_multimin_fminimizer *sfm = NULL;
13400: gsl_vector *ss, *x;
13401: gsl_multimin_function minex_func;
13402:
13403: /* Initial vertex size vector */
13404: ss = gsl_vector_alloc (NDIM);
13405:
13406: if (ss == NULL){
13407: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
13408: }
13409: /* Set all step sizes to 1 */
13410: gsl_vector_set_all (ss, 0.001);
13411:
13412: /* Starting point */
1.126 brouard 13413:
1.136 brouard 13414: x = gsl_vector_alloc (NDIM);
13415:
13416: if (x == NULL){
13417: gsl_vector_free(ss);
13418: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
13419: }
13420:
13421: /* Initialize method and iterate */
13422: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 13423: /* gsl_vector_set(x, 0, 0.0268); */
13424: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 13425: gsl_vector_set(x, 0, p[1]);
13426: gsl_vector_set(x, 1, p[2]);
13427:
13428: minex_func.f = &gompertz_f;
13429: minex_func.n = NDIM;
13430: minex_func.params = (void *)&p; /* ??? */
13431:
13432: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
13433: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
13434:
13435: printf("Iterations beginning .....\n\n");
13436: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
13437:
13438: iteri=0;
13439: while (rval == GSL_CONTINUE){
13440: iteri++;
13441: status = gsl_multimin_fminimizer_iterate(sfm);
13442:
13443: if (status) printf("error: %s\n", gsl_strerror (status));
13444: fflush(0);
13445:
13446: if (status)
13447: break;
13448:
13449: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
13450: ssval = gsl_multimin_fminimizer_size (sfm);
13451:
13452: if (rval == GSL_SUCCESS)
13453: printf ("converged to a local maximum at\n");
13454:
13455: printf("%5d ", iteri);
13456: for (it = 0; it < NDIM; it++){
13457: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
13458: }
13459: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
13460: }
13461:
13462: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
13463:
13464: gsl_vector_free(x); /* initial values */
13465: gsl_vector_free(ss); /* inital step size */
13466: for (it=0; it<NDIM; it++){
13467: p[it+1]=gsl_vector_get(sfm->x,it);
13468: fprintf(ficrespow," %.12lf", p[it]);
13469: }
13470: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
13471: #endif
13472: #ifdef POWELL
13473: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
13474: #endif
1.126 brouard 13475: fclose(ficrespow);
13476:
1.203 brouard 13477: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 13478:
13479: for(i=1; i <=NDIM; i++)
13480: for(j=i+1;j<=NDIM;j++)
1.220 brouard 13481: matcov[i][j]=matcov[j][i];
1.126 brouard 13482:
13483: printf("\nCovariance matrix\n ");
1.203 brouard 13484: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 13485: for(i=1; i <=NDIM; i++) {
13486: for(j=1;j<=NDIM;j++){
1.220 brouard 13487: printf("%f ",matcov[i][j]);
13488: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 13489: }
1.203 brouard 13490: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 13491: }
13492:
13493: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 13494: for (i=1;i<=NDIM;i++) {
1.126 brouard 13495: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 13496: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
13497: }
1.302 brouard 13498: lsurv=vector(agegomp,AGESUP);
13499: lpop=vector(agegomp,AGESUP);
13500: tpop=vector(agegomp,AGESUP);
1.126 brouard 13501: lsurv[agegomp]=100000;
13502:
13503: for (k=agegomp;k<=AGESUP;k++) {
13504: agemortsup=k;
13505: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
13506: }
13507:
13508: for (k=agegomp;k<agemortsup;k++)
13509: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
13510:
13511: for (k=agegomp;k<agemortsup;k++){
13512: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
13513: sumlpop=sumlpop+lpop[k];
13514: }
13515:
13516: tpop[agegomp]=sumlpop;
13517: for (k=agegomp;k<(agemortsup-3);k++){
13518: /* tpop[k+1]=2;*/
13519: tpop[k+1]=tpop[k]-lpop[k];
13520: }
13521:
13522:
13523: printf("\nAge lx qx dx Lx Tx e(x)\n");
13524: for (k=agegomp;k<(agemortsup-2);k++)
13525: 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]);
13526:
13527:
13528: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 13529: ageminpar=50;
13530: agemaxpar=100;
1.194 brouard 13531: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
13532: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13533: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13534: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
13535: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
13536: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13537: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13538: }else{
13539: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
13540: 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 13541: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 13542: }
1.201 brouard 13543: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 13544: stepm, weightopt,\
13545: model,imx,p,matcov,agemortsup);
13546:
1.302 brouard 13547: free_vector(lsurv,agegomp,AGESUP);
13548: free_vector(lpop,agegomp,AGESUP);
13549: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 13550: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 13551: free_ivector(dcwave,firstobs,lastobs);
13552: free_vector(agecens,firstobs,lastobs);
13553: free_vector(ageexmed,firstobs,lastobs);
13554: free_ivector(cens,firstobs,lastobs);
1.220 brouard 13555: #ifdef GSL
1.136 brouard 13556: #endif
1.186 brouard 13557: } /* Endof if mle==-3 mortality only */
1.205 brouard 13558: /* Standard */
13559: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
13560: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13561: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 13562: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 13563: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
13564: for (k=1; k<=npar;k++)
13565: printf(" %d %8.5f",k,p[k]);
13566: printf("\n");
1.205 brouard 13567: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
13568: /* mlikeli uses func not funcone */
1.247 brouard 13569: /* for(i=1;i<nlstate;i++){ */
13570: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
13571: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
13572: /* } */
1.205 brouard 13573: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
13574: }
13575: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
13576: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
13577: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
13578: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13579: }
13580: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 13581: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
13582: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
1.335 brouard 13583: /* exit(0); */
1.126 brouard 13584: for (k=1; k<=npar;k++)
13585: printf(" %d %8.5f",k,p[k]);
13586: printf("\n");
13587:
13588: /*--------- results files --------------*/
1.283 brouard 13589: /* 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 13590:
13591:
13592: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13593: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 13594: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 13595:
13596: printf("#model= 1 + age ");
13597: fprintf(ficres,"#model= 1 + age ");
13598: fprintf(ficlog,"#model= 1 + age ");
13599: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
13600: </ul>", model);
13601:
13602: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
13603: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
13604: if(nagesqr==1){
13605: printf(" + age*age ");
13606: fprintf(ficres," + age*age ");
13607: fprintf(ficlog," + age*age ");
13608: fprintf(fichtm, "<th>+ age*age</th>");
13609: }
13610: for(j=1;j <=ncovmodel-2;j++){
13611: if(Typevar[j]==0) {
13612: printf(" + V%d ",Tvar[j]);
13613: fprintf(ficres," + V%d ",Tvar[j]);
13614: fprintf(ficlog," + V%d ",Tvar[j]);
13615: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13616: }else if(Typevar[j]==1) {
13617: printf(" + V%d*age ",Tvar[j]);
13618: fprintf(ficres," + V%d*age ",Tvar[j]);
13619: fprintf(ficlog," + V%d*age ",Tvar[j]);
13620: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13621: }else if(Typevar[j]==2) {
13622: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13623: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13624: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13625: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13626: }
13627: }
13628: printf("\n");
13629: fprintf(ficres,"\n");
13630: fprintf(ficlog,"\n");
13631: fprintf(fichtm, "</tr>");
13632: fprintf(fichtm, "\n");
13633:
13634:
1.126 brouard 13635: for(i=1,jk=1; i <=nlstate; i++){
13636: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 13637: if (k != i) {
1.319 brouard 13638: fprintf(fichtm, "<tr>");
1.225 brouard 13639: printf("%d%d ",i,k);
13640: fprintf(ficlog,"%d%d ",i,k);
13641: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 13642: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13643: for(j=1; j <=ncovmodel; j++){
13644: printf("%12.7f ",p[jk]);
13645: fprintf(ficlog,"%12.7f ",p[jk]);
13646: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 13647: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 13648: jk++;
13649: }
13650: printf("\n");
13651: fprintf(ficlog,"\n");
13652: fprintf(ficres,"\n");
1.319 brouard 13653: fprintf(fichtm, "</tr>\n");
1.225 brouard 13654: }
1.126 brouard 13655: }
13656: }
1.319 brouard 13657: /* fprintf(fichtm,"</tr>\n"); */
13658: fprintf(fichtm,"</table>\n");
13659: fprintf(fichtm, "\n");
13660:
1.203 brouard 13661: if(mle != 0){
13662: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 13663: ftolhess=ftol; /* Usually correct */
1.203 brouard 13664: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
13665: 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");
13666: 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 13667: 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 13668: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
13669: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
13670: if(nagesqr==1){
13671: printf(" + age*age ");
13672: fprintf(ficres," + age*age ");
13673: fprintf(ficlog," + age*age ");
13674: fprintf(fichtm, "<th>+ age*age</th>");
13675: }
13676: for(j=1;j <=ncovmodel-2;j++){
13677: if(Typevar[j]==0) {
13678: printf(" + V%d ",Tvar[j]);
13679: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13680: }else if(Typevar[j]==1) {
13681: printf(" + V%d*age ",Tvar[j]);
13682: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13683: }else if(Typevar[j]==2) {
13684: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13685: }
13686: }
13687: fprintf(fichtm, "</tr>\n");
13688:
1.203 brouard 13689: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 13690: for(k=1; k <=(nlstate+ndeath); k++){
13691: if (k != i) {
1.319 brouard 13692: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 13693: printf("%d%d ",i,k);
13694: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 13695: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13696: for(j=1; j <=ncovmodel; j++){
1.319 brouard 13697: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 13698: 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]));
13699: 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 13700: if(fabs(wald) > 1.96){
1.321 brouard 13701: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 13702: }else{
13703: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
13704: }
1.324 brouard 13705: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 13706: 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 13707: jk++;
13708: }
13709: printf("\n");
13710: fprintf(ficlog,"\n");
1.319 brouard 13711: fprintf(fichtm, "</tr>\n");
1.225 brouard 13712: }
13713: }
1.193 brouard 13714: }
1.203 brouard 13715: } /* end of hesscov and Wald tests */
1.319 brouard 13716: fprintf(fichtm,"</table>\n");
1.225 brouard 13717:
1.203 brouard 13718: /* */
1.126 brouard 13719: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
13720: printf("# Scales (for hessian or gradient estimation)\n");
13721: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
13722: for(i=1,jk=1; i <=nlstate; i++){
13723: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 13724: if (j!=i) {
13725: fprintf(ficres,"%1d%1d",i,j);
13726: printf("%1d%1d",i,j);
13727: fprintf(ficlog,"%1d%1d",i,j);
13728: for(k=1; k<=ncovmodel;k++){
13729: printf(" %.5e",delti[jk]);
13730: fprintf(ficlog," %.5e",delti[jk]);
13731: fprintf(ficres," %.5e",delti[jk]);
13732: jk++;
13733: }
13734: printf("\n");
13735: fprintf(ficlog,"\n");
13736: fprintf(ficres,"\n");
13737: }
1.126 brouard 13738: }
13739: }
13740:
13741: 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 13742: if(mle >= 1) /* To big for the screen */
1.126 brouard 13743: 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");
13744: 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");
13745: /* # 121 Var(a12)\n\ */
13746: /* # 122 Cov(b12,a12) Var(b12)\n\ */
13747: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
13748: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
13749: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
13750: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
13751: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
13752: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
13753:
13754:
13755: /* Just to have a covariance matrix which will be more understandable
13756: even is we still don't want to manage dictionary of variables
13757: */
13758: for(itimes=1;itimes<=2;itimes++){
13759: jj=0;
13760: for(i=1; i <=nlstate; i++){
1.225 brouard 13761: for(j=1; j <=nlstate+ndeath; j++){
13762: if(j==i) continue;
13763: for(k=1; k<=ncovmodel;k++){
13764: jj++;
13765: ca[0]= k+'a'-1;ca[1]='\0';
13766: if(itimes==1){
13767: if(mle>=1)
13768: printf("#%1d%1d%d",i,j,k);
13769: fprintf(ficlog,"#%1d%1d%d",i,j,k);
13770: fprintf(ficres,"#%1d%1d%d",i,j,k);
13771: }else{
13772: if(mle>=1)
13773: printf("%1d%1d%d",i,j,k);
13774: fprintf(ficlog,"%1d%1d%d",i,j,k);
13775: fprintf(ficres,"%1d%1d%d",i,j,k);
13776: }
13777: ll=0;
13778: for(li=1;li <=nlstate; li++){
13779: for(lj=1;lj <=nlstate+ndeath; lj++){
13780: if(lj==li) continue;
13781: for(lk=1;lk<=ncovmodel;lk++){
13782: ll++;
13783: if(ll<=jj){
13784: cb[0]= lk +'a'-1;cb[1]='\0';
13785: if(ll<jj){
13786: if(itimes==1){
13787: if(mle>=1)
13788: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13789: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13790: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13791: }else{
13792: if(mle>=1)
13793: printf(" %.5e",matcov[jj][ll]);
13794: fprintf(ficlog," %.5e",matcov[jj][ll]);
13795: fprintf(ficres," %.5e",matcov[jj][ll]);
13796: }
13797: }else{
13798: if(itimes==1){
13799: if(mle>=1)
13800: printf(" Var(%s%1d%1d)",ca,i,j);
13801: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
13802: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
13803: }else{
13804: if(mle>=1)
13805: printf(" %.7e",matcov[jj][ll]);
13806: fprintf(ficlog," %.7e",matcov[jj][ll]);
13807: fprintf(ficres," %.7e",matcov[jj][ll]);
13808: }
13809: }
13810: }
13811: } /* end lk */
13812: } /* end lj */
13813: } /* end li */
13814: if(mle>=1)
13815: printf("\n");
13816: fprintf(ficlog,"\n");
13817: fprintf(ficres,"\n");
13818: numlinepar++;
13819: } /* end k*/
13820: } /*end j */
1.126 brouard 13821: } /* end i */
13822: } /* end itimes */
13823:
13824: fflush(ficlog);
13825: fflush(ficres);
1.225 brouard 13826: while(fgets(line, MAXLINE, ficpar)) {
13827: /* If line starts with a # it is a comment */
13828: if (line[0] == '#') {
13829: numlinepar++;
13830: fputs(line,stdout);
13831: fputs(line,ficparo);
13832: fputs(line,ficlog);
1.299 brouard 13833: fputs(line,ficres);
1.225 brouard 13834: continue;
13835: }else
13836: break;
13837: }
13838:
1.209 brouard 13839: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
13840: /* ungetc(c,ficpar); */
13841: /* fgets(line, MAXLINE, ficpar); */
13842: /* fputs(line,stdout); */
13843: /* fputs(line,ficparo); */
13844: /* } */
13845: /* ungetc(c,ficpar); */
1.126 brouard 13846:
13847: estepm=0;
1.209 brouard 13848: 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 13849:
13850: if (num_filled != 6) {
13851: 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);
13852: 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);
13853: goto end;
13854: }
13855: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
13856: }
13857: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
13858: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
13859:
1.209 brouard 13860: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 13861: if (estepm==0 || estepm < stepm) estepm=stepm;
13862: if (fage <= 2) {
13863: bage = ageminpar;
13864: fage = agemaxpar;
13865: }
13866:
13867: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 13868: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
13869: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 13870:
1.186 brouard 13871: /* Other stuffs, more or less useful */
1.254 brouard 13872: while(fgets(line, MAXLINE, ficpar)) {
13873: /* If line starts with a # it is a comment */
13874: if (line[0] == '#') {
13875: numlinepar++;
13876: fputs(line,stdout);
13877: fputs(line,ficparo);
13878: fputs(line,ficlog);
1.299 brouard 13879: fputs(line,ficres);
1.254 brouard 13880: continue;
13881: }else
13882: break;
13883: }
13884:
13885: 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){
13886:
13887: if (num_filled != 7) {
13888: 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);
13889: 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);
13890: goto end;
13891: }
13892: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13893: 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);
13894: 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);
13895: 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 13896: }
1.254 brouard 13897:
13898: while(fgets(line, MAXLINE, ficpar)) {
13899: /* If line starts with a # it is a comment */
13900: if (line[0] == '#') {
13901: numlinepar++;
13902: fputs(line,stdout);
13903: fputs(line,ficparo);
13904: fputs(line,ficlog);
1.299 brouard 13905: fputs(line,ficres);
1.254 brouard 13906: continue;
13907: }else
13908: break;
1.126 brouard 13909: }
13910:
13911:
13912: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13913: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13914:
1.254 brouard 13915: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13916: if (num_filled != 1) {
13917: 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);
13918: 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);
13919: goto end;
13920: }
13921: printf("pop_based=%d\n",popbased);
13922: fprintf(ficlog,"pop_based=%d\n",popbased);
13923: fprintf(ficparo,"pop_based=%d\n",popbased);
13924: fprintf(ficres,"pop_based=%d\n",popbased);
13925: }
13926:
1.258 brouard 13927: /* Results */
1.332 brouard 13928: /* Value of covariate in each resultine will be compututed (if product) and sorted according to model rank */
13929: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
13930: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 13931: endishere=0;
1.258 brouard 13932: nresult=0;
1.308 brouard 13933: parameterline=0;
1.258 brouard 13934: do{
13935: if(!fgets(line, MAXLINE, ficpar)){
13936: endishere=1;
1.308 brouard 13937: parameterline=15;
1.258 brouard 13938: }else if (line[0] == '#') {
13939: /* If line starts with a # it is a comment */
1.254 brouard 13940: numlinepar++;
13941: fputs(line,stdout);
13942: fputs(line,ficparo);
13943: fputs(line,ficlog);
1.299 brouard 13944: fputs(line,ficres);
1.254 brouard 13945: continue;
1.258 brouard 13946: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13947: parameterline=11;
1.296 brouard 13948: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 13949: parameterline=12;
1.307 brouard 13950: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 13951: parameterline=13;
1.307 brouard 13952: }
1.258 brouard 13953: else{
13954: parameterline=14;
1.254 brouard 13955: }
1.308 brouard 13956: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 13957: case 11:
1.296 brouard 13958: 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)){
13959: 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 13960: 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);
13961: 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);
13962: 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);
13963: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 13964: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
13965: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 13966: prvforecast = 1;
13967: }
13968: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 13969: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13970: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13971: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 13972: prvforecast = 2;
13973: }
13974: else {
13975: 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);
13976: 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);
13977: goto end;
1.258 brouard 13978: }
1.254 brouard 13979: break;
1.258 brouard 13980: case 12:
1.296 brouard 13981: 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)){
13982: 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);
13983: 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);
13984: 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);
13985: 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);
13986: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 13987: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
13988: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 13989: prvbackcast = 1;
13990: }
13991: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 13992: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13993: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13994: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 13995: prvbackcast = 2;
13996: }
13997: else {
13998: 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);
13999: 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);
14000: goto end;
1.258 brouard 14001: }
1.230 brouard 14002: break;
1.258 brouard 14003: case 13:
1.332 brouard 14004: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 14005: nresult++; /* Sum of resultlines */
1.342 brouard 14006: /* printf("Result %d: result:%s\n",nresult, resultlineori); */
1.332 brouard 14007: /* removefirstspace(&resultlineori); */
14008:
14009: if(strstr(resultlineori,"v") !=0){
14010: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
14011: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
14012: return 1;
14013: }
14014: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
1.342 brouard 14015: /* printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori); */
1.318 brouard 14016: if(nresult > MAXRESULTLINESPONE-1){
14017: 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);
14018: 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 14019: goto end;
14020: }
1.332 brouard 14021:
1.310 brouard 14022: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 14023: fprintf(ficparo,"result: %s\n",resultline);
14024: fprintf(ficres,"result: %s\n",resultline);
14025: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 14026: } else
14027: goto end;
1.307 brouard 14028: break;
14029: case 14:
14030: printf("Error: Unknown command '%s'\n",line);
14031: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 14032: if(line[0] == ' ' || line[0] == '\n'){
14033: printf("It should not be an empty line '%s'\n",line);
14034: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
14035: }
1.307 brouard 14036: if(ncovmodel >=2 && nresult==0 ){
14037: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
14038: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 14039: }
1.307 brouard 14040: /* goto end; */
14041: break;
1.308 brouard 14042: case 15:
14043: printf("End of resultlines.\n");
14044: fprintf(ficlog,"End of resultlines.\n");
14045: break;
14046: default: /* parameterline =0 */
1.307 brouard 14047: nresult=1;
14048: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 14049: } /* End switch parameterline */
14050: }while(endishere==0); /* End do */
1.126 brouard 14051:
1.230 brouard 14052: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 14053: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 14054:
14055: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 14056: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 14057: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 14058: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
14059: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 14060: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 14061: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
14062: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 14063: }else{
1.270 brouard 14064: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 14065: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
14066: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
14067: if(prvforecast==1){
14068: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
14069: jprojd=jproj1;
14070: mprojd=mproj1;
14071: anprojd=anproj1;
14072: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
14073: jprojf=jproj2;
14074: mprojf=mproj2;
14075: anprojf=anproj2;
14076: } else if(prvforecast == 2){
14077: dateprojd=dateintmean;
14078: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
14079: dateprojf=dateintmean+yrfproj;
14080: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
14081: }
14082: if(prvbackcast==1){
14083: datebackd=(jback1+12*mback1+365*anback1)/365;
14084: jbackd=jback1;
14085: mbackd=mback1;
14086: anbackd=anback1;
14087: datebackf=(jback2+12*mback2+365*anback2)/365;
14088: jbackf=jback2;
14089: mbackf=mback2;
14090: anbackf=anback2;
14091: } else if(prvbackcast == 2){
14092: datebackd=dateintmean;
14093: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
14094: datebackf=dateintmean-yrbproj;
14095: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
14096: }
14097:
14098: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 14099: }
14100: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 14101: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
14102: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 14103:
1.225 brouard 14104: /*------------ free_vector -------------*/
14105: /* chdir(path); */
1.220 brouard 14106:
1.215 brouard 14107: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
14108: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
14109: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
14110: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 14111: free_lvector(num,firstobs,lastobs);
14112: free_vector(agedc,firstobs,lastobs);
1.126 brouard 14113: /*free_matrix(covar,0,NCOVMAX,1,n);*/
14114: /*free_matrix(covar,1,NCOVMAX,1,n);*/
14115: fclose(ficparo);
14116: fclose(ficres);
1.220 brouard 14117:
14118:
1.186 brouard 14119: /* Other results (useful)*/
1.220 brouard 14120:
14121:
1.126 brouard 14122: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 14123: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
14124: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 14125: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 14126: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 14127: fclose(ficrespl);
14128:
14129: /*------------- h Pij x at various ages ------------*/
1.180 brouard 14130: /*#include "hpijx.h"*/
1.332 brouard 14131: /** 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?*/
14132: /* calls hpxij with combination k */
1.180 brouard 14133: hPijx(p, bage, fage);
1.145 brouard 14134: fclose(ficrespij);
1.227 brouard 14135:
1.220 brouard 14136: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 14137: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 14138: k=1;
1.126 brouard 14139: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 14140:
1.269 brouard 14141: /* Prevalence for each covariate combination in probs[age][status][cov] */
14142: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14143: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 14144: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 14145: for(k=1;k<=ncovcombmax;k++)
14146: probs[i][j][k]=0.;
1.269 brouard 14147: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
14148: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 14149: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 14150: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
14151: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 14152: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 14153: for(k=1;k<=ncovcombmax;k++)
14154: mobaverages[i][j][k]=0.;
1.219 brouard 14155: mobaverage=mobaverages;
14156: if (mobilav!=0) {
1.235 brouard 14157: printf("Movingaveraging observed prevalence\n");
1.258 brouard 14158: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 14159: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
14160: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
14161: printf(" Error in movingaverage mobilav=%d\n",mobilav);
14162: }
1.269 brouard 14163: } else if (mobilavproj !=0) {
1.235 brouard 14164: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 14165: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 14166: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
14167: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
14168: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
14169: }
1.269 brouard 14170: }else{
14171: printf("Internal error moving average\n");
14172: fflush(stdout);
14173: exit(1);
1.219 brouard 14174: }
14175: }/* end if moving average */
1.227 brouard 14176:
1.126 brouard 14177: /*---------- Forecasting ------------------*/
1.296 brouard 14178: if(prevfcast==1){
14179: /* /\* if(stepm ==1){*\/ */
14180: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14181: /*This done previously after freqsummary.*/
14182: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
14183: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
14184:
14185: /* } else if (prvforecast==2){ */
14186: /* /\* if(stepm ==1){*\/ */
14187: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
14188: /* } */
14189: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
14190: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 14191: }
1.269 brouard 14192:
1.296 brouard 14193: /* Prevbcasting */
14194: if(prevbcast==1){
1.219 brouard 14195: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14196: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14197: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
14198:
14199: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
14200:
14201: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 14202:
1.219 brouard 14203: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
14204: fclose(ficresplb);
14205:
1.222 brouard 14206: hBijx(p, bage, fage, mobaverage);
14207: fclose(ficrespijb);
1.219 brouard 14208:
1.296 brouard 14209: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
14210: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
14211: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
14212: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
14213: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
14214: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
14215:
14216:
1.269 brouard 14217: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14218:
14219:
1.269 brouard 14220: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 14221: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14222: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
14223: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 14224: } /* end Prevbcasting */
1.268 brouard 14225:
1.186 brouard 14226:
14227: /* ------ Other prevalence ratios------------ */
1.126 brouard 14228:
1.215 brouard 14229: free_ivector(wav,1,imx);
14230: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
14231: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
14232: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 14233:
14234:
1.127 brouard 14235: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 14236:
1.201 brouard 14237: strcpy(filerese,"E_");
14238: strcat(filerese,fileresu);
1.126 brouard 14239: if((ficreseij=fopen(filerese,"w"))==NULL) {
14240: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14241: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
14242: }
1.208 brouard 14243: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
14244: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 14245:
14246: pstamp(ficreseij);
1.219 brouard 14247:
1.235 brouard 14248: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14249: if (cptcovn < 1){i1=1;}
14250:
14251: for(nres=1; nres <= nresult; nres++) /* For each resultline */
14252: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 14253: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 14254: continue;
1.219 brouard 14255: fprintf(ficreseij,"\n#****** ");
1.235 brouard 14256: printf("\n#****** ");
1.225 brouard 14257: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 14258: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
14259: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 14260: }
14261: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.337 brouard 14262: printf(" V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]); /* TvarsQ[j] gives the name of the jth quantitative (fixed or time v) */
14263: fprintf(ficreseij,"V%d=%lg ",TvarsQ[j], TinvDoQresult[nres][TvarsQ[j]]);
1.219 brouard 14264: }
14265: fprintf(ficreseij,"******\n");
1.235 brouard 14266: printf("******\n");
1.219 brouard 14267:
14268: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14269: oldm=oldms;savm=savms;
1.330 brouard 14270: /* 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 14271: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 14272:
1.219 brouard 14273: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 14274: }
14275: fclose(ficreseij);
1.208 brouard 14276: printf("done evsij\n");fflush(stdout);
14277: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 14278:
1.218 brouard 14279:
1.227 brouard 14280: /*---------- State-specific expectancies and variances ------------*/
1.336 brouard 14281: /* Should be moved in a function */
1.201 brouard 14282: strcpy(filerest,"T_");
14283: strcat(filerest,fileresu);
1.127 brouard 14284: if((ficrest=fopen(filerest,"w"))==NULL) {
14285: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
14286: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
14287: }
1.208 brouard 14288: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
14289: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 14290: strcpy(fileresstde,"STDE_");
14291: strcat(fileresstde,fileresu);
1.126 brouard 14292: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 14293: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
14294: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 14295: }
1.227 brouard 14296: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
14297: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 14298:
1.201 brouard 14299: strcpy(filerescve,"CVE_");
14300: strcat(filerescve,fileresu);
1.126 brouard 14301: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 14302: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
14303: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 14304: }
1.227 brouard 14305: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
14306: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 14307:
1.201 brouard 14308: strcpy(fileresv,"V_");
14309: strcat(fileresv,fileresu);
1.126 brouard 14310: if((ficresvij=fopen(fileresv,"w"))==NULL) {
14311: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
14312: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
14313: }
1.227 brouard 14314: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
14315: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 14316:
1.235 brouard 14317: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
14318: if (cptcovn < 1){i1=1;}
14319:
1.334 brouard 14320: for(nres=1; nres <= nresult; nres++) /* For each resultline, find the combination and output results according to the values of dummies and then quanti. */
14321: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying. For each nres and each value at position k
14322: * we know Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline
14323: * Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline
14324: * and Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
14325: /* */
14326: 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 14327: continue;
1.321 brouard 14328: printf("\n# model %s \n#****** Result for:", model);
14329: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
14330: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.334 brouard 14331: /* It might not be a good idea to mix dummies and quantitative */
14332: /* 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 *\/ */
14333: 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 */
14334: /* printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /\* Output by variables in the resultline *\/ */
14335: /* Tvaraff[j] is the name of the dummy variable in position j in the equation model:
14336: * Tvaraff[1]@9={4, 3, 0, 0, 0, 0, 0, 0, 0}, in model=V5+V4+V3+V4*V3+V5*age
14337: * (V5 is quanti) V4 and V3 are dummies
14338: * TnsdVar[4] is the position 1 and TnsdVar[3]=2 in codtabm(k,l)(V4 V3)=V4 V3
14339: * l=1 l=2
14340: * k=1 1 1 0 0
14341: * k=2 2 1 1 0
14342: * k=3 [1] [2] 0 1
14343: * k=4 2 2 1 1
14344: * If nres=1 result: V3=1 V4=0 then k=3 and outputs
14345: * If nres=2 result: V4=1 V3=0 then k=2 and outputs
14346: * nres=1 =>k=3 j=1 V4= nbcode[4][codtabm(3,1)=1)=0; j=2 V3= nbcode[3][codtabm(3,2)=2]=1
14347: * nres=2 =>k=2 j=1 V4= nbcode[4][codtabm(2,1)=2)=1; j=2 V3= nbcode[3][codtabm(2,2)=1]=0
14348: */
14349: /* Tvresult[nres][j] Name of the variable at position j in this resultline */
14350: /* Tresult[nres][j] Value of this variable at position j could be a float if quantitative */
14351: /* We give up with the combinations!! */
1.342 brouard 14352: /* if(debugILK) */
14353: /* printf("\n j=%d In computing T_ Dummy[modelresult[%d][%d]]=%d, modelresult[%d][%d]=%d cptcovs=%d, cptcoveff=%d Fixed[modelresult[nres][j]]=%d\n", j, nres, j, Dummy[modelresult[nres][j]],nres,j,modelresult[nres][j],cptcovs, cptcoveff,Fixed[modelresult[nres][j]]); /\* end if dummy or quanti *\/ */
1.334 brouard 14354:
14355: if(Dummy[modelresult[nres][j]]==0){/* Dummy variable of the variable in position modelresult in the model corresponding to j in resultline */
1.344 ! brouard 14356: /* printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][j]); /\* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline *\/ */ /* TinvDoQresult[nres][Name of the variable] */
! 14357: printf("V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); /* Output of each value for the combination TKresult[nres], ordered by the covariate values in the resultline */
! 14358: fprintf(ficlog,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
! 14359: fprintf(ficrest,"V%d=%lg ",Tvresult[nres][j],TinvDoQresult[nres][Tvresult[nres][j]]); /* Output of each value for the combination TKresult[nres], ordere by the covariate values in the resultline */
1.334 brouard 14360: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14361: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14362: }else{
14363: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14364: }
14365: /* fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14366: /* fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14367: }else if(Dummy[modelresult[nres][j]]==1){ /* Quanti variable */
14368: /* For each selected (single) quantitative value */
1.337 brouard 14369: printf(" V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14370: fprintf(ficlog," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
14371: fprintf(ficrest," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][j]);
1.334 brouard 14372: if(Fixed[modelresult[nres][j]]==0){ /* Fixed */
14373: printf("fixed ");fprintf(ficlog,"fixed ");fprintf(ficrest,"fixed ");
14374: }else{
14375: printf("varyi ");fprintf(ficlog,"varyi ");fprintf(ficrest,"varyi ");
14376: }
14377: }else{
14378: 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 */
14379: 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 */
14380: exit(1);
14381: }
1.335 brouard 14382: } /* End loop for each variable in the resultline */
1.334 brouard 14383: /* for (j=1; j<= nsq; j++){ /\* For each selected (single) quantitative value *\/ */
14384: /* printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /\* Wrong j is not in the equation model *\/ */
14385: /* fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14386: /* fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); */
14387: /* } */
1.208 brouard 14388: fprintf(ficrest,"******\n");
1.227 brouard 14389: fprintf(ficlog,"******\n");
14390: printf("******\n");
1.208 brouard 14391:
14392: fprintf(ficresstdeij,"\n#****** ");
14393: fprintf(ficrescveij,"\n#****** ");
1.337 brouard 14394: /* It could have been: for(j=1;j<=cptcoveff;j++) {printf("V=%d=%lg",Tvresult[nres][cpt],TinvDoQresult[nres][Tvresult[nres][cpt]]);} */
14395: /* But it won't be sorted and depends on how the resultline is ordered */
1.225 brouard 14396: for(j=1;j<=cptcoveff;j++) {
1.334 brouard 14397: fprintf(ficresstdeij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14398: /* fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14399: /* fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); */
14400: }
14401: 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 14402: fprintf(ficresstdeij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
14403: fprintf(ficrescveij," V%d=%lg ",Tvar[TvarsQind[j]],Tqresult[nres][resultmodel[nres][TvarsQind[j]]]);
1.235 brouard 14404: }
1.208 brouard 14405: fprintf(ficresstdeij,"******\n");
14406: fprintf(ficrescveij,"******\n");
14407:
14408: fprintf(ficresvij,"\n#****** ");
1.238 brouard 14409: /* pstamp(ficresvij); */
1.225 brouard 14410: for(j=1;j<=cptcoveff;j++)
1.335 brouard 14411: fprintf(ficresvij,"V%d=%d ",Tvresult[nres][j],Tresult[nres][j]);
14412: /* fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]); */
1.235 brouard 14413: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 14414: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
1.337 brouard 14415: fprintf(ficresvij," V%d=%lg ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 14416: }
1.208 brouard 14417: fprintf(ficresvij,"******\n");
14418:
14419: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14420: oldm=oldms;savm=savms;
1.235 brouard 14421: printf(" cvevsij ");
14422: fprintf(ficlog, " cvevsij ");
14423: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 14424: printf(" end cvevsij \n ");
14425: fprintf(ficlog, " end cvevsij \n ");
14426:
14427: /*
14428: */
14429: /* goto endfree; */
14430:
14431: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
14432: pstamp(ficrest);
14433:
1.269 brouard 14434: epj=vector(1,nlstate+1);
1.208 brouard 14435: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 14436: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
14437: cptcod= 0; /* To be deleted */
14438: printf("varevsij vpopbased=%d \n",vpopbased);
14439: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 14440: 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 14441: 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 ");
14442: if(vpopbased==1)
14443: 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);
14444: else
1.288 brouard 14445: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.335 brouard 14446: fprintf(ficrest,"# Age popbased mobilav e.. (std) "); /* Adding covariate values? */
1.227 brouard 14447: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
14448: fprintf(ficrest,"\n");
14449: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 14450: printf("Computing age specific forward period (stable) prevalences in each health state \n");
14451: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 14452: for(age=bage; age <=fage ;age++){
1.235 brouard 14453: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 14454: if (vpopbased==1) {
14455: if(mobilav ==0){
14456: for(i=1; i<=nlstate;i++)
14457: prlim[i][i]=probs[(int)age][i][k];
14458: }else{ /* mobilav */
14459: for(i=1; i<=nlstate;i++)
14460: prlim[i][i]=mobaverage[(int)age][i][k];
14461: }
14462: }
1.219 brouard 14463:
1.227 brouard 14464: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
14465: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
14466: /* printf(" age %4.0f ",age); */
14467: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
14468: for(i=1, epj[j]=0.;i <=nlstate;i++) {
14469: epj[j] += prlim[i][i]*eij[i][j][(int)age];
14470: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
14471: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
14472: }
14473: epj[nlstate+1] +=epj[j];
14474: }
14475: /* printf(" age %4.0f \n",age); */
1.219 brouard 14476:
1.227 brouard 14477: for(i=1, vepp=0.;i <=nlstate;i++)
14478: for(j=1;j <=nlstate;j++)
14479: vepp += vareij[i][j][(int)age];
14480: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
14481: for(j=1;j <=nlstate;j++){
14482: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
14483: }
14484: fprintf(ficrest,"\n");
14485: }
1.208 brouard 14486: } /* End vpopbased */
1.269 brouard 14487: free_vector(epj,1,nlstate+1);
1.208 brouard 14488: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
14489: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 14490: printf("done selection\n");fflush(stdout);
14491: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 14492:
1.335 brouard 14493: } /* End k selection or end covariate selection for nres */
1.227 brouard 14494:
14495: printf("done State-specific expectancies\n");fflush(stdout);
14496: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
14497:
1.335 brouard 14498: /* variance-covariance of forward period prevalence */
1.269 brouard 14499: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 14500:
1.227 brouard 14501:
1.290 brouard 14502: free_vector(weight,firstobs,lastobs);
1.330 brouard 14503: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 14504: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 14505: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
14506: free_matrix(anint,1,maxwav,firstobs,lastobs);
14507: free_matrix(mint,1,maxwav,firstobs,lastobs);
14508: free_ivector(cod,firstobs,lastobs);
1.227 brouard 14509: free_ivector(tab,1,NCOVMAX);
14510: fclose(ficresstdeij);
14511: fclose(ficrescveij);
14512: fclose(ficresvij);
14513: fclose(ficrest);
14514: fclose(ficpar);
14515:
14516:
1.126 brouard 14517: /*---------- End : free ----------------*/
1.219 brouard 14518: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 14519: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
14520: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 14521: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
14522: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 14523: } /* mle==-3 arrives here for freeing */
1.227 brouard 14524: /* endfree:*/
14525: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
14526: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
14527: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.341 brouard 14528: /* if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs); */
14529: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,ncovcol+nqv+1,ncovcol+nqv+ntv+nqtv,firstobs,lastobs);
1.290 brouard 14530: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
14531: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
14532: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 14533: free_matrix(matcov,1,npar,1,npar);
14534: free_matrix(hess,1,npar,1,npar);
14535: /*free_vector(delti,1,npar);*/
14536: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14537: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 14538: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 14539: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
14540:
14541: free_ivector(ncodemax,1,NCOVMAX);
14542: free_ivector(ncodemaxwundef,1,NCOVMAX);
14543: free_ivector(Dummy,-1,NCOVMAX);
14544: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 14545: free_ivector(DummyV,1,NCOVMAX);
14546: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 14547: free_ivector(Typevar,-1,NCOVMAX);
14548: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 14549: free_ivector(TvarsQ,1,NCOVMAX);
14550: free_ivector(TvarsQind,1,NCOVMAX);
14551: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 14552: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 14553: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 14554: free_ivector(TvarFD,1,NCOVMAX);
14555: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 14556: free_ivector(TvarF,1,NCOVMAX);
14557: free_ivector(TvarFind,1,NCOVMAX);
14558: free_ivector(TvarV,1,NCOVMAX);
14559: free_ivector(TvarVind,1,NCOVMAX);
14560: free_ivector(TvarA,1,NCOVMAX);
14561: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 14562: free_ivector(TvarFQ,1,NCOVMAX);
14563: free_ivector(TvarFQind,1,NCOVMAX);
14564: free_ivector(TvarVD,1,NCOVMAX);
14565: free_ivector(TvarVDind,1,NCOVMAX);
14566: free_ivector(TvarVQ,1,NCOVMAX);
14567: free_ivector(TvarVQind,1,NCOVMAX);
1.339 brouard 14568: free_ivector(TvarVV,1,NCOVMAX);
14569: free_ivector(TvarVVind,1,NCOVMAX);
14570:
1.230 brouard 14571: free_ivector(Tvarsel,1,NCOVMAX);
14572: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 14573: free_ivector(Tposprod,1,NCOVMAX);
14574: free_ivector(Tprod,1,NCOVMAX);
14575: free_ivector(Tvaraff,1,NCOVMAX);
1.338 brouard 14576: free_ivector(invalidvarcomb,0,ncovcombmax);
1.227 brouard 14577: free_ivector(Tage,1,NCOVMAX);
14578: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 14579: free_ivector(TmodelInvind,1,NCOVMAX);
14580: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 14581:
14582: free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
14583:
1.227 brouard 14584: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
14585: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 14586: fflush(fichtm);
14587: fflush(ficgp);
14588:
1.227 brouard 14589:
1.126 brouard 14590: if((nberr >0) || (nbwarn>0)){
1.216 brouard 14591: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
14592: 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 14593: }else{
14594: printf("End of Imach\n");
14595: fprintf(ficlog,"End of Imach\n");
14596: }
14597: printf("See log file on %s\n",filelog);
14598: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 14599: /*(void) gettimeofday(&end_time,&tzp);*/
14600: rend_time = time(NULL);
14601: end_time = *localtime(&rend_time);
14602: /* tml = *localtime(&end_time.tm_sec); */
14603: strcpy(strtend,asctime(&end_time));
1.126 brouard 14604: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
14605: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 14606: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 14607:
1.157 brouard 14608: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
14609: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
14610: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 14611: /* printf("Total time was %d uSec.\n", total_usecs);*/
14612: /* if(fileappend(fichtm,optionfilehtm)){ */
14613: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14614: fclose(fichtm);
14615: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
14616: fclose(fichtmcov);
14617: fclose(ficgp);
14618: fclose(ficlog);
14619: /*------ End -----------*/
1.227 brouard 14620:
1.281 brouard 14621:
14622: /* Executes gnuplot */
1.227 brouard 14623:
14624: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 14625: #ifdef WIN32
1.227 brouard 14626: if (_chdir(pathcd) != 0)
14627: printf("Can't move to directory %s!\n",path);
14628: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 14629: #else
1.227 brouard 14630: if(chdir(pathcd) != 0)
14631: printf("Can't move to directory %s!\n", path);
14632: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 14633: #endif
1.126 brouard 14634: printf("Current directory %s!\n",pathcd);
14635: /*strcat(plotcmd,CHARSEPARATOR);*/
14636: sprintf(plotcmd,"gnuplot");
1.157 brouard 14637: #ifdef _WIN32
1.126 brouard 14638: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
14639: #endif
14640: if(!stat(plotcmd,&info)){
1.158 brouard 14641: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14642: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 14643: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 14644: }else
14645: strcpy(pplotcmd,plotcmd);
1.157 brouard 14646: #ifdef __unix
1.126 brouard 14647: strcpy(plotcmd,GNUPLOTPROGRAM);
14648: if(!stat(plotcmd,&info)){
1.158 brouard 14649: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 14650: }else
14651: strcpy(pplotcmd,plotcmd);
14652: #endif
14653: }else
14654: strcpy(pplotcmd,plotcmd);
14655:
14656: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 14657: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 14658: strcpy(pplotcmd,plotcmd);
1.227 brouard 14659:
1.126 brouard 14660: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 14661: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 14662: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 14663: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 14664: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 14665: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 14666: strcpy(plotcmd,pplotcmd);
14667: }
1.126 brouard 14668: }
1.158 brouard 14669: printf(" Successful, please wait...");
1.126 brouard 14670: while (z[0] != 'q') {
14671: /* chdir(path); */
1.154 brouard 14672: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 14673: scanf("%s",z);
14674: /* if (z[0] == 'c') system("./imach"); */
14675: if (z[0] == 'e') {
1.158 brouard 14676: #ifdef __APPLE__
1.152 brouard 14677: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 14678: #elif __linux
14679: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 14680: #else
1.152 brouard 14681: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 14682: #endif
14683: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
14684: system(pplotcmd);
1.126 brouard 14685: }
14686: else if (z[0] == 'g') system(plotcmd);
14687: else if (z[0] == 'q') exit(0);
14688: }
1.227 brouard 14689: end:
1.126 brouard 14690: while (z[0] != 'q') {
1.195 brouard 14691: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 14692: scanf("%s",z);
14693: }
1.283 brouard 14694: printf("End\n");
1.282 brouard 14695: exit(0);
1.126 brouard 14696: }
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