Annotation of imach/src/imach.c, revision 1.324
1.324 ! brouard 1: /* $Id: imach.c,v 1.323 2022/07/22 12:30:08 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.324 ! brouard 4: Revision 1.323 2022/07/22 12:30:08 brouard
! 5: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
! 6:
1.323 brouard 7: Revision 1.322 2022/07/22 12:27:48 brouard
8: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
9:
1.322 brouard 10: Revision 1.321 2022/07/22 12:04:24 brouard
11: Summary: r28
12:
13: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
14:
1.321 brouard 15: Revision 1.320 2022/06/02 05:10:11 brouard
16: *** empty log message ***
17:
1.320 brouard 18: Revision 1.319 2022/06/02 04:45:11 brouard
19: * imach.c (Module): Adding the Wald tests from the log to the main
20: htm for better display of the maximum likelihood estimators.
21:
1.319 brouard 22: Revision 1.318 2022/05/24 08:10:59 brouard
23: * imach.c (Module): Some attempts to find a bug of wrong estimates
24: of confidencce intervals with product in the equation modelC
25:
1.318 brouard 26: Revision 1.317 2022/05/15 15:06:23 brouard
27: * imach.c (Module): Some minor improvements
28:
1.317 brouard 29: Revision 1.316 2022/05/11 15:11:31 brouard
30: Summary: r27
31:
1.316 brouard 32: Revision 1.315 2022/05/11 15:06:32 brouard
33: *** empty log message ***
34:
1.315 brouard 35: Revision 1.314 2022/04/13 17:43:09 brouard
36: * imach.c (Module): Adding link to text data files
37:
1.314 brouard 38: Revision 1.313 2022/04/11 15:57:42 brouard
39: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
40:
1.313 brouard 41: Revision 1.312 2022/04/05 21:24:39 brouard
42: *** empty log message ***
43:
1.312 brouard 44: Revision 1.311 2022/04/05 21:03:51 brouard
45: Summary: Fixed quantitative covariates
46:
47: Fixed covariates (dummy or quantitative)
48: with missing values have never been allowed but are ERRORS and
49: program quits. Standard deviations of fixed covariates were
50: wrongly computed. Mean and standard deviations of time varying
51: covariates are still not computed.
52:
1.311 brouard 53: Revision 1.310 2022/03/17 08:45:53 brouard
54: Summary: 99r25
55:
56: Improving detection of errors: result lines should be compatible with
57: the model.
58:
1.310 brouard 59: Revision 1.309 2021/05/20 12:39:14 brouard
60: Summary: Version 0.99r24
61:
1.309 brouard 62: Revision 1.308 2021/03/31 13:11:57 brouard
63: Summary: Version 0.99r23
64:
65:
66: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
67:
1.308 brouard 68: Revision 1.307 2021/03/08 18:11:32 brouard
69: Summary: 0.99r22 fixed bug on result:
70:
1.307 brouard 71: Revision 1.306 2021/02/20 15:44:02 brouard
72: Summary: Version 0.99r21
73:
74: * imach.c (Module): Fix bug on quitting after result lines!
75: (Module): Version 0.99r21
76:
1.306 brouard 77: Revision 1.305 2021/02/20 15:28:30 brouard
78: * imach.c (Module): Fix bug on quitting after result lines!
79:
1.305 brouard 80: Revision 1.304 2021/02/12 11:34:20 brouard
81: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
82:
1.304 brouard 83: Revision 1.303 2021/02/11 19:50:15 brouard
84: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
85:
1.303 brouard 86: Revision 1.302 2020/02/22 21:00:05 brouard
87: * (Module): imach.c Update mle=-3 (for computing Life expectancy
88: and life table from the data without any state)
89:
1.302 brouard 90: Revision 1.301 2019/06/04 13:51:20 brouard
91: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
92:
1.301 brouard 93: Revision 1.300 2019/05/22 19:09:45 brouard
94: Summary: version 0.99r19 of May 2019
95:
1.300 brouard 96: Revision 1.299 2019/05/22 18:37:08 brouard
97: Summary: Cleaned 0.99r19
98:
1.299 brouard 99: Revision 1.298 2019/05/22 18:19:56 brouard
100: *** empty log message ***
101:
1.298 brouard 102: Revision 1.297 2019/05/22 17:56:10 brouard
103: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
104:
1.297 brouard 105: Revision 1.296 2019/05/20 13:03:18 brouard
106: Summary: Projection syntax simplified
107:
108:
109: We can now start projections, forward or backward, from the mean date
110: of inteviews up to or down to a number of years of projection:
111: prevforecast=1 yearsfproj=15.3 mobil_average=0
112: or
113: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
114: or
115: prevbackcast=1 yearsbproj=12.3 mobil_average=1
116: or
117: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
118:
1.296 brouard 119: Revision 1.295 2019/05/18 09:52:50 brouard
120: Summary: doxygen tex bug
121:
1.295 brouard 122: Revision 1.294 2019/05/16 14:54:33 brouard
123: Summary: There was some wrong lines added
124:
1.294 brouard 125: Revision 1.293 2019/05/09 15:17:34 brouard
126: *** empty log message ***
127:
1.293 brouard 128: Revision 1.292 2019/05/09 14:17:20 brouard
129: Summary: Some updates
130:
1.292 brouard 131: Revision 1.291 2019/05/09 13:44:18 brouard
132: Summary: Before ncovmax
133:
1.291 brouard 134: Revision 1.290 2019/05/09 13:39:37 brouard
135: Summary: 0.99r18 unlimited number of individuals
136:
137: 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.
138:
1.290 brouard 139: Revision 1.289 2018/12/13 09:16:26 brouard
140: Summary: Bug for young ages (<-30) will be in r17
141:
1.289 brouard 142: Revision 1.288 2018/05/02 20:58:27 brouard
143: Summary: Some bugs fixed
144:
1.288 brouard 145: Revision 1.287 2018/05/01 17:57:25 brouard
146: Summary: Bug fixed by providing frequencies only for non missing covariates
147:
1.287 brouard 148: Revision 1.286 2018/04/27 14:27:04 brouard
149: Summary: some minor bugs
150:
1.286 brouard 151: Revision 1.285 2018/04/21 21:02:16 brouard
152: Summary: Some bugs fixed, valgrind tested
153:
1.285 brouard 154: Revision 1.284 2018/04/20 05:22:13 brouard
155: Summary: Computing mean and stdeviation of fixed quantitative variables
156:
1.284 brouard 157: Revision 1.283 2018/04/19 14:49:16 brouard
158: Summary: Some minor bugs fixed
159:
1.283 brouard 160: Revision 1.282 2018/02/27 22:50:02 brouard
161: *** empty log message ***
162:
1.282 brouard 163: Revision 1.281 2018/02/27 19:25:23 brouard
164: Summary: Adding second argument for quitting
165:
1.281 brouard 166: Revision 1.280 2018/02/21 07:58:13 brouard
167: Summary: 0.99r15
168:
169: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
170:
1.280 brouard 171: Revision 1.279 2017/07/20 13:35:01 brouard
172: Summary: temporary working
173:
1.279 brouard 174: Revision 1.278 2017/07/19 14:09:02 brouard
175: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
176:
1.278 brouard 177: Revision 1.277 2017/07/17 08:53:49 brouard
178: Summary: BOM files can be read now
179:
1.277 brouard 180: Revision 1.276 2017/06/30 15:48:31 brouard
181: Summary: Graphs improvements
182:
1.276 brouard 183: Revision 1.275 2017/06/30 13:39:33 brouard
184: Summary: Saito's color
185:
1.275 brouard 186: Revision 1.274 2017/06/29 09:47:08 brouard
187: Summary: Version 0.99r14
188:
1.274 brouard 189: Revision 1.273 2017/06/27 11:06:02 brouard
190: Summary: More documentation on projections
191:
1.273 brouard 192: Revision 1.272 2017/06/27 10:22:40 brouard
193: Summary: Color of backprojection changed from 6 to 5(yellow)
194:
1.272 brouard 195: Revision 1.271 2017/06/27 10:17:50 brouard
196: Summary: Some bug with rint
197:
1.271 brouard 198: Revision 1.270 2017/05/24 05:45:29 brouard
199: *** empty log message ***
200:
1.270 brouard 201: Revision 1.269 2017/05/23 08:39:25 brouard
202: Summary: Code into subroutine, cleanings
203:
1.269 brouard 204: Revision 1.268 2017/05/18 20:09:32 brouard
205: Summary: backprojection and confidence intervals of backprevalence
206:
1.268 brouard 207: Revision 1.267 2017/05/13 10:25:05 brouard
208: Summary: temporary save for backprojection
209:
1.267 brouard 210: Revision 1.266 2017/05/13 07:26:12 brouard
211: Summary: Version 0.99r13 (improvements and bugs fixed)
212:
1.266 brouard 213: Revision 1.265 2017/04/26 16:22:11 brouard
214: Summary: imach 0.99r13 Some bugs fixed
215:
1.265 brouard 216: Revision 1.264 2017/04/26 06:01:29 brouard
217: Summary: Labels in graphs
218:
1.264 brouard 219: Revision 1.263 2017/04/24 15:23:15 brouard
220: Summary: to save
221:
1.263 brouard 222: Revision 1.262 2017/04/18 16:48:12 brouard
223: *** empty log message ***
224:
1.262 brouard 225: Revision 1.261 2017/04/05 10:14:09 brouard
226: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
227:
1.261 brouard 228: Revision 1.260 2017/04/04 17:46:59 brouard
229: Summary: Gnuplot indexations fixed (humm)
230:
1.260 brouard 231: Revision 1.259 2017/04/04 13:01:16 brouard
232: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
233:
1.259 brouard 234: Revision 1.258 2017/04/03 10:17:47 brouard
235: Summary: Version 0.99r12
236:
237: Some cleanings, conformed with updated documentation.
238:
1.258 brouard 239: Revision 1.257 2017/03/29 16:53:30 brouard
240: Summary: Temp
241:
1.257 brouard 242: Revision 1.256 2017/03/27 05:50:23 brouard
243: Summary: Temporary
244:
1.256 brouard 245: Revision 1.255 2017/03/08 16:02:28 brouard
246: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
247:
1.255 brouard 248: Revision 1.254 2017/03/08 07:13:00 brouard
249: Summary: Fixing data parameter line
250:
1.254 brouard 251: Revision 1.253 2016/12/15 11:59:41 brouard
252: Summary: 0.99 in progress
253:
1.253 brouard 254: Revision 1.252 2016/09/15 21:15:37 brouard
255: *** empty log message ***
256:
1.252 brouard 257: Revision 1.251 2016/09/15 15:01:13 brouard
258: Summary: not working
259:
1.251 brouard 260: Revision 1.250 2016/09/08 16:07:27 brouard
261: Summary: continue
262:
1.250 brouard 263: Revision 1.249 2016/09/07 17:14:18 brouard
264: Summary: Starting values from frequencies
265:
1.249 brouard 266: Revision 1.248 2016/09/07 14:10:18 brouard
267: *** empty log message ***
268:
1.248 brouard 269: Revision 1.247 2016/09/02 11:11:21 brouard
270: *** empty log message ***
271:
1.247 brouard 272: Revision 1.246 2016/09/02 08:49:22 brouard
273: *** empty log message ***
274:
1.246 brouard 275: Revision 1.245 2016/09/02 07:25:01 brouard
276: *** empty log message ***
277:
1.245 brouard 278: Revision 1.244 2016/09/02 07:17:34 brouard
279: *** empty log message ***
280:
1.244 brouard 281: Revision 1.243 2016/09/02 06:45:35 brouard
282: *** empty log message ***
283:
1.243 brouard 284: Revision 1.242 2016/08/30 15:01:20 brouard
285: Summary: Fixing a lots
286:
1.242 brouard 287: Revision 1.241 2016/08/29 17:17:25 brouard
288: Summary: gnuplot problem in Back projection to fix
289:
1.241 brouard 290: Revision 1.240 2016/08/29 07:53:18 brouard
291: Summary: Better
292:
1.240 brouard 293: Revision 1.239 2016/08/26 15:51:03 brouard
294: Summary: Improvement in Powell output in order to copy and paste
295:
296: Author:
297:
1.239 brouard 298: Revision 1.238 2016/08/26 14:23:35 brouard
299: Summary: Starting tests of 0.99
300:
1.238 brouard 301: Revision 1.237 2016/08/26 09:20:19 brouard
302: Summary: to valgrind
303:
1.237 brouard 304: Revision 1.236 2016/08/25 10:50:18 brouard
305: *** empty log message ***
306:
1.236 brouard 307: Revision 1.235 2016/08/25 06:59:23 brouard
308: *** empty log message ***
309:
1.235 brouard 310: Revision 1.234 2016/08/23 16:51:20 brouard
311: *** empty log message ***
312:
1.234 brouard 313: Revision 1.233 2016/08/23 07:40:50 brouard
314: Summary: not working
315:
1.233 brouard 316: Revision 1.232 2016/08/22 14:20:21 brouard
317: Summary: not working
318:
1.232 brouard 319: Revision 1.231 2016/08/22 07:17:15 brouard
320: Summary: not working
321:
1.231 brouard 322: Revision 1.230 2016/08/22 06:55:53 brouard
323: Summary: Not working
324:
1.230 brouard 325: Revision 1.229 2016/07/23 09:45:53 brouard
326: Summary: Completing for func too
327:
1.229 brouard 328: Revision 1.228 2016/07/22 17:45:30 brouard
329: Summary: Fixing some arrays, still debugging
330:
1.227 brouard 331: Revision 1.226 2016/07/12 18:42:34 brouard
332: Summary: temp
333:
1.226 brouard 334: Revision 1.225 2016/07/12 08:40:03 brouard
335: Summary: saving but not running
336:
1.225 brouard 337: Revision 1.224 2016/07/01 13:16:01 brouard
338: Summary: Fixes
339:
1.224 brouard 340: Revision 1.223 2016/02/19 09:23:35 brouard
341: Summary: temporary
342:
1.223 brouard 343: Revision 1.222 2016/02/17 08:14:50 brouard
344: Summary: Probably last 0.98 stable version 0.98r6
345:
1.222 brouard 346: Revision 1.221 2016/02/15 23:35:36 brouard
347: Summary: minor bug
348:
1.220 brouard 349: Revision 1.219 2016/02/15 00:48:12 brouard
350: *** empty log message ***
351:
1.219 brouard 352: Revision 1.218 2016/02/12 11:29:23 brouard
353: Summary: 0.99 Back projections
354:
1.218 brouard 355: Revision 1.217 2015/12/23 17:18:31 brouard
356: Summary: Experimental backcast
357:
1.217 brouard 358: Revision 1.216 2015/12/18 17:32:11 brouard
359: Summary: 0.98r4 Warning and status=-2
360:
361: Version 0.98r4 is now:
362: - displaying an error when status is -1, date of interview unknown and date of death known;
363: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
364: Older changes concerning s=-2, dating from 2005 have been supersed.
365:
1.216 brouard 366: Revision 1.215 2015/12/16 08:52:24 brouard
367: Summary: 0.98r4 working
368:
1.215 brouard 369: Revision 1.214 2015/12/16 06:57:54 brouard
370: Summary: temporary not working
371:
1.214 brouard 372: Revision 1.213 2015/12/11 18:22:17 brouard
373: Summary: 0.98r4
374:
1.213 brouard 375: Revision 1.212 2015/11/21 12:47:24 brouard
376: Summary: minor typo
377:
1.212 brouard 378: Revision 1.211 2015/11/21 12:41:11 brouard
379: Summary: 0.98r3 with some graph of projected cross-sectional
380:
381: Author: Nicolas Brouard
382:
1.211 brouard 383: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 384: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 385: Summary: Adding ftolpl parameter
386: Author: N Brouard
387:
388: We had difficulties to get smoothed confidence intervals. It was due
389: to the period prevalence which wasn't computed accurately. The inner
390: parameter ftolpl is now an outer parameter of the .imach parameter
391: file after estepm. If ftolpl is small 1.e-4 and estepm too,
392: computation are long.
393:
1.209 brouard 394: Revision 1.208 2015/11/17 14:31:57 brouard
395: Summary: temporary
396:
1.208 brouard 397: Revision 1.207 2015/10/27 17:36:57 brouard
398: *** empty log message ***
399:
1.207 brouard 400: Revision 1.206 2015/10/24 07:14:11 brouard
401: *** empty log message ***
402:
1.206 brouard 403: Revision 1.205 2015/10/23 15:50:53 brouard
404: Summary: 0.98r3 some clarification for graphs on likelihood contributions
405:
1.205 brouard 406: Revision 1.204 2015/10/01 16:20:26 brouard
407: Summary: Some new graphs of contribution to likelihood
408:
1.204 brouard 409: Revision 1.203 2015/09/30 17:45:14 brouard
410: Summary: looking at better estimation of the hessian
411:
412: Also a better criteria for convergence to the period prevalence And
413: therefore adding the number of years needed to converge. (The
414: prevalence in any alive state shold sum to one
415:
1.203 brouard 416: Revision 1.202 2015/09/22 19:45:16 brouard
417: Summary: Adding some overall graph on contribution to likelihood. Might change
418:
1.202 brouard 419: Revision 1.201 2015/09/15 17:34:58 brouard
420: Summary: 0.98r0
421:
422: - Some new graphs like suvival functions
423: - Some bugs fixed like model=1+age+V2.
424:
1.201 brouard 425: Revision 1.200 2015/09/09 16:53:55 brouard
426: Summary: Big bug thanks to Flavia
427:
428: Even model=1+age+V2. did not work anymore
429:
1.200 brouard 430: Revision 1.199 2015/09/07 14:09:23 brouard
431: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
432:
1.199 brouard 433: Revision 1.198 2015/09/03 07:14:39 brouard
434: Summary: 0.98q5 Flavia
435:
1.198 brouard 436: Revision 1.197 2015/09/01 18:24:39 brouard
437: *** empty log message ***
438:
1.197 brouard 439: Revision 1.196 2015/08/18 23:17:52 brouard
440: Summary: 0.98q5
441:
1.196 brouard 442: Revision 1.195 2015/08/18 16:28:39 brouard
443: Summary: Adding a hack for testing purpose
444:
445: After reading the title, ftol and model lines, if the comment line has
446: a q, starting with #q, the answer at the end of the run is quit. It
447: permits to run test files in batch with ctest. The former workaround was
448: $ echo q | imach foo.imach
449:
1.195 brouard 450: Revision 1.194 2015/08/18 13:32:00 brouard
451: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
452:
1.194 brouard 453: Revision 1.193 2015/08/04 07:17:42 brouard
454: Summary: 0.98q4
455:
1.193 brouard 456: Revision 1.192 2015/07/16 16:49:02 brouard
457: Summary: Fixing some outputs
458:
1.192 brouard 459: Revision 1.191 2015/07/14 10:00:33 brouard
460: Summary: Some fixes
461:
1.191 brouard 462: Revision 1.190 2015/05/05 08:51:13 brouard
463: Summary: Adding digits in output parameters (7 digits instead of 6)
464:
465: Fix 1+age+.
466:
1.190 brouard 467: Revision 1.189 2015/04/30 14:45:16 brouard
468: Summary: 0.98q2
469:
1.189 brouard 470: Revision 1.188 2015/04/30 08:27:53 brouard
471: *** empty log message ***
472:
1.188 brouard 473: Revision 1.187 2015/04/29 09:11:15 brouard
474: *** empty log message ***
475:
1.187 brouard 476: Revision 1.186 2015/04/23 12:01:52 brouard
477: Summary: V1*age is working now, version 0.98q1
478:
479: Some codes had been disabled in order to simplify and Vn*age was
480: working in the optimization phase, ie, giving correct MLE parameters,
481: but, as usual, outputs were not correct and program core dumped.
482:
1.186 brouard 483: Revision 1.185 2015/03/11 13:26:42 brouard
484: Summary: Inclusion of compile and links command line for Intel Compiler
485:
1.185 brouard 486: Revision 1.184 2015/03/11 11:52:39 brouard
487: Summary: Back from Windows 8. Intel Compiler
488:
1.184 brouard 489: Revision 1.183 2015/03/10 20:34:32 brouard
490: Summary: 0.98q0, trying with directest, mnbrak fixed
491:
492: We use directest instead of original Powell test; probably no
493: incidence on the results, but better justifications;
494: We fixed Numerical Recipes mnbrak routine which was wrong and gave
495: wrong results.
496:
1.183 brouard 497: Revision 1.182 2015/02/12 08:19:57 brouard
498: Summary: Trying to keep directest which seems simpler and more general
499: Author: Nicolas Brouard
500:
1.182 brouard 501: Revision 1.181 2015/02/11 23:22:24 brouard
502: Summary: Comments on Powell added
503:
504: Author:
505:
1.181 brouard 506: Revision 1.180 2015/02/11 17:33:45 brouard
507: Summary: Finishing move from main to function (hpijx and prevalence_limit)
508:
1.180 brouard 509: Revision 1.179 2015/01/04 09:57:06 brouard
510: Summary: back to OS/X
511:
1.179 brouard 512: Revision 1.178 2015/01/04 09:35:48 brouard
513: *** empty log message ***
514:
1.178 brouard 515: Revision 1.177 2015/01/03 18:40:56 brouard
516: Summary: Still testing ilc32 on OSX
517:
1.177 brouard 518: Revision 1.176 2015/01/03 16:45:04 brouard
519: *** empty log message ***
520:
1.176 brouard 521: Revision 1.175 2015/01/03 16:33:42 brouard
522: *** empty log message ***
523:
1.175 brouard 524: Revision 1.174 2015/01/03 16:15:49 brouard
525: Summary: Still in cross-compilation
526:
1.174 brouard 527: Revision 1.173 2015/01/03 12:06:26 brouard
528: Summary: trying to detect cross-compilation
529:
1.173 brouard 530: Revision 1.172 2014/12/27 12:07:47 brouard
531: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
532:
1.172 brouard 533: Revision 1.171 2014/12/23 13:26:59 brouard
534: Summary: Back from Visual C
535:
536: Still problem with utsname.h on Windows
537:
1.171 brouard 538: Revision 1.170 2014/12/23 11:17:12 brouard
539: Summary: Cleaning some \%% back to %%
540:
541: The escape was mandatory for a specific compiler (which one?), but too many warnings.
542:
1.170 brouard 543: Revision 1.169 2014/12/22 23:08:31 brouard
544: Summary: 0.98p
545:
546: Outputs some informations on compiler used, OS etc. Testing on different platforms.
547:
1.169 brouard 548: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 549: Summary: update
1.169 brouard 550:
1.168 brouard 551: Revision 1.167 2014/12/22 13:50:56 brouard
552: Summary: Testing uname and compiler version and if compiled 32 or 64
553:
554: Testing on Linux 64
555:
1.167 brouard 556: Revision 1.166 2014/12/22 11:40:47 brouard
557: *** empty log message ***
558:
1.166 brouard 559: Revision 1.165 2014/12/16 11:20:36 brouard
560: Summary: After compiling on Visual C
561:
562: * imach.c (Module): Merging 1.61 to 1.162
563:
1.165 brouard 564: Revision 1.164 2014/12/16 10:52:11 brouard
565: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
566:
567: * imach.c (Module): Merging 1.61 to 1.162
568:
1.164 brouard 569: Revision 1.163 2014/12/16 10:30:11 brouard
570: * imach.c (Module): Merging 1.61 to 1.162
571:
1.163 brouard 572: Revision 1.162 2014/09/25 11:43:39 brouard
573: Summary: temporary backup 0.99!
574:
1.162 brouard 575: Revision 1.1 2014/09/16 11:06:58 brouard
576: Summary: With some code (wrong) for nlopt
577:
578: Author:
579:
580: Revision 1.161 2014/09/15 20:41:41 brouard
581: Summary: Problem with macro SQR on Intel compiler
582:
1.161 brouard 583: Revision 1.160 2014/09/02 09:24:05 brouard
584: *** empty log message ***
585:
1.160 brouard 586: Revision 1.159 2014/09/01 10:34:10 brouard
587: Summary: WIN32
588: Author: Brouard
589:
1.159 brouard 590: Revision 1.158 2014/08/27 17:11:51 brouard
591: *** empty log message ***
592:
1.158 brouard 593: Revision 1.157 2014/08/27 16:26:55 brouard
594: Summary: Preparing windows Visual studio version
595: Author: Brouard
596:
597: In order to compile on Visual studio, time.h is now correct and time_t
598: and tm struct should be used. difftime should be used but sometimes I
599: just make the differences in raw time format (time(&now).
600: Trying to suppress #ifdef LINUX
601: Add xdg-open for __linux in order to open default browser.
602:
1.157 brouard 603: Revision 1.156 2014/08/25 20:10:10 brouard
604: *** empty log message ***
605:
1.156 brouard 606: Revision 1.155 2014/08/25 18:32:34 brouard
607: Summary: New compile, minor changes
608: Author: Brouard
609:
1.155 brouard 610: Revision 1.154 2014/06/20 17:32:08 brouard
611: Summary: Outputs now all graphs of convergence to period prevalence
612:
1.154 brouard 613: Revision 1.153 2014/06/20 16:45:46 brouard
614: Summary: If 3 live state, convergence to period prevalence on same graph
615: Author: Brouard
616:
1.153 brouard 617: Revision 1.152 2014/06/18 17:54:09 brouard
618: Summary: open browser, use gnuplot on same dir than imach if not found in the path
619:
1.152 brouard 620: Revision 1.151 2014/06/18 16:43:30 brouard
621: *** empty log message ***
622:
1.151 brouard 623: Revision 1.150 2014/06/18 16:42:35 brouard
624: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
625: Author: brouard
626:
1.150 brouard 627: Revision 1.149 2014/06/18 15:51:14 brouard
628: Summary: Some fixes in parameter files errors
629: Author: Nicolas Brouard
630:
1.149 brouard 631: Revision 1.148 2014/06/17 17:38:48 brouard
632: Summary: Nothing new
633: Author: Brouard
634:
635: Just a new packaging for OS/X version 0.98nS
636:
1.148 brouard 637: Revision 1.147 2014/06/16 10:33:11 brouard
638: *** empty log message ***
639:
1.147 brouard 640: Revision 1.146 2014/06/16 10:20:28 brouard
641: Summary: Merge
642: Author: Brouard
643:
644: Merge, before building revised version.
645:
1.146 brouard 646: Revision 1.145 2014/06/10 21:23:15 brouard
647: Summary: Debugging with valgrind
648: Author: Nicolas Brouard
649:
650: Lot of changes in order to output the results with some covariates
651: After the Edimburgh REVES conference 2014, it seems mandatory to
652: improve the code.
653: No more memory valgrind error but a lot has to be done in order to
654: continue the work of splitting the code into subroutines.
655: Also, decodemodel has been improved. Tricode is still not
656: optimal. nbcode should be improved. Documentation has been added in
657: the source code.
658:
1.144 brouard 659: Revision 1.143 2014/01/26 09:45:38 brouard
660: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
661:
662: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
663: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
664:
1.143 brouard 665: Revision 1.142 2014/01/26 03:57:36 brouard
666: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
667:
668: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
669:
1.142 brouard 670: Revision 1.141 2014/01/26 02:42:01 brouard
671: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
672:
1.141 brouard 673: Revision 1.140 2011/09/02 10:37:54 brouard
674: Summary: times.h is ok with mingw32 now.
675:
1.140 brouard 676: Revision 1.139 2010/06/14 07:50:17 brouard
677: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
678: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
679:
1.139 brouard 680: Revision 1.138 2010/04/30 18:19:40 brouard
681: *** empty log message ***
682:
1.138 brouard 683: Revision 1.137 2010/04/29 18:11:38 brouard
684: (Module): Checking covariates for more complex models
685: than V1+V2. A lot of change to be done. Unstable.
686:
1.137 brouard 687: Revision 1.136 2010/04/26 20:30:53 brouard
688: (Module): merging some libgsl code. Fixing computation
689: of likelione (using inter/intrapolation if mle = 0) in order to
690: get same likelihood as if mle=1.
691: Some cleaning of code and comments added.
692:
1.136 brouard 693: Revision 1.135 2009/10/29 15:33:14 brouard
694: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
695:
1.135 brouard 696: Revision 1.134 2009/10/29 13:18:53 brouard
697: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
698:
1.134 brouard 699: Revision 1.133 2009/07/06 10:21:25 brouard
700: just nforces
701:
1.133 brouard 702: Revision 1.132 2009/07/06 08:22:05 brouard
703: Many tings
704:
1.132 brouard 705: Revision 1.131 2009/06/20 16:22:47 brouard
706: Some dimensions resccaled
707:
1.131 brouard 708: Revision 1.130 2009/05/26 06:44:34 brouard
709: (Module): Max Covariate is now set to 20 instead of 8. A
710: lot of cleaning with variables initialized to 0. Trying to make
711: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
712:
1.130 brouard 713: Revision 1.129 2007/08/31 13:49:27 lievre
714: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
715:
1.129 lievre 716: Revision 1.128 2006/06/30 13:02:05 brouard
717: (Module): Clarifications on computing e.j
718:
1.128 brouard 719: Revision 1.127 2006/04/28 18:11:50 brouard
720: (Module): Yes the sum of survivors was wrong since
721: imach-114 because nhstepm was no more computed in the age
722: loop. Now we define nhstepma in the age loop.
723: (Module): In order to speed up (in case of numerous covariates) we
724: compute health expectancies (without variances) in a first step
725: and then all the health expectancies with variances or standard
726: deviation (needs data from the Hessian matrices) which slows the
727: computation.
728: In the future we should be able to stop the program is only health
729: expectancies and graph are needed without standard deviations.
730:
1.127 brouard 731: Revision 1.126 2006/04/28 17:23:28 brouard
732: (Module): Yes the sum of survivors was wrong since
733: imach-114 because nhstepm was no more computed in the age
734: loop. Now we define nhstepma in the age loop.
735: Version 0.98h
736:
1.126 brouard 737: Revision 1.125 2006/04/04 15:20:31 lievre
738: Errors in calculation of health expectancies. Age was not initialized.
739: Forecasting file added.
740:
741: Revision 1.124 2006/03/22 17:13:53 lievre
742: Parameters are printed with %lf instead of %f (more numbers after the comma).
743: The log-likelihood is printed in the log file
744:
745: Revision 1.123 2006/03/20 10:52:43 brouard
746: * imach.c (Module): <title> changed, corresponds to .htm file
747: name. <head> headers where missing.
748:
749: * imach.c (Module): Weights can have a decimal point as for
750: English (a comma might work with a correct LC_NUMERIC environment,
751: otherwise the weight is truncated).
752: Modification of warning when the covariates values are not 0 or
753: 1.
754: Version 0.98g
755:
756: Revision 1.122 2006/03/20 09:45:41 brouard
757: (Module): Weights can have a decimal point as for
758: English (a comma might work with a correct LC_NUMERIC environment,
759: otherwise the weight is truncated).
760: Modification of warning when the covariates values are not 0 or
761: 1.
762: Version 0.98g
763:
764: Revision 1.121 2006/03/16 17:45:01 lievre
765: * imach.c (Module): Comments concerning covariates added
766:
767: * imach.c (Module): refinements in the computation of lli if
768: status=-2 in order to have more reliable computation if stepm is
769: not 1 month. Version 0.98f
770:
771: Revision 1.120 2006/03/16 15:10:38 lievre
772: (Module): refinements in the computation of lli if
773: status=-2 in order to have more reliable computation if stepm is
774: not 1 month. Version 0.98f
775:
776: Revision 1.119 2006/03/15 17:42:26 brouard
777: (Module): Bug if status = -2, the loglikelihood was
778: computed as likelihood omitting the logarithm. Version O.98e
779:
780: Revision 1.118 2006/03/14 18:20:07 brouard
781: (Module): varevsij Comments added explaining the second
782: table of variances if popbased=1 .
783: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
784: (Module): Function pstamp added
785: (Module): Version 0.98d
786:
787: Revision 1.117 2006/03/14 17:16:22 brouard
788: (Module): varevsij Comments added explaining the second
789: table of variances if popbased=1 .
790: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
791: (Module): Function pstamp added
792: (Module): Version 0.98d
793:
794: Revision 1.116 2006/03/06 10:29:27 brouard
795: (Module): Variance-covariance wrong links and
796: varian-covariance of ej. is needed (Saito).
797:
798: Revision 1.115 2006/02/27 12:17:45 brouard
799: (Module): One freematrix added in mlikeli! 0.98c
800:
801: Revision 1.114 2006/02/26 12:57:58 brouard
802: (Module): Some improvements in processing parameter
803: filename with strsep.
804:
805: Revision 1.113 2006/02/24 14:20:24 brouard
806: (Module): Memory leaks checks with valgrind and:
807: datafile was not closed, some imatrix were not freed and on matrix
808: allocation too.
809:
810: Revision 1.112 2006/01/30 09:55:26 brouard
811: (Module): Back to gnuplot.exe instead of wgnuplot.exe
812:
813: Revision 1.111 2006/01/25 20:38:18 brouard
814: (Module): Lots of cleaning and bugs added (Gompertz)
815: (Module): Comments can be added in data file. Missing date values
816: can be a simple dot '.'.
817:
818: Revision 1.110 2006/01/25 00:51:50 brouard
819: (Module): Lots of cleaning and bugs added (Gompertz)
820:
821: Revision 1.109 2006/01/24 19:37:15 brouard
822: (Module): Comments (lines starting with a #) are allowed in data.
823:
824: Revision 1.108 2006/01/19 18:05:42 lievre
825: Gnuplot problem appeared...
826: To be fixed
827:
828: Revision 1.107 2006/01/19 16:20:37 brouard
829: Test existence of gnuplot in imach path
830:
831: Revision 1.106 2006/01/19 13:24:36 brouard
832: Some cleaning and links added in html output
833:
834: Revision 1.105 2006/01/05 20:23:19 lievre
835: *** empty log message ***
836:
837: Revision 1.104 2005/09/30 16:11:43 lievre
838: (Module): sump fixed, loop imx fixed, and simplifications.
839: (Module): If the status is missing at the last wave but we know
840: that the person is alive, then we can code his/her status as -2
841: (instead of missing=-1 in earlier versions) and his/her
842: contributions to the likelihood is 1 - Prob of dying from last
843: health status (= 1-p13= p11+p12 in the easiest case of somebody in
844: the healthy state at last known wave). Version is 0.98
845:
846: Revision 1.103 2005/09/30 15:54:49 lievre
847: (Module): sump fixed, loop imx fixed, and simplifications.
848:
849: Revision 1.102 2004/09/15 17:31:30 brouard
850: Add the possibility to read data file including tab characters.
851:
852: Revision 1.101 2004/09/15 10:38:38 brouard
853: Fix on curr_time
854:
855: Revision 1.100 2004/07/12 18:29:06 brouard
856: Add version for Mac OS X. Just define UNIX in Makefile
857:
858: Revision 1.99 2004/06/05 08:57:40 brouard
859: *** empty log message ***
860:
861: Revision 1.98 2004/05/16 15:05:56 brouard
862: New version 0.97 . First attempt to estimate force of mortality
863: directly from the data i.e. without the need of knowing the health
864: state at each age, but using a Gompertz model: log u =a + b*age .
865: This is the basic analysis of mortality and should be done before any
866: other analysis, in order to test if the mortality estimated from the
867: cross-longitudinal survey is different from the mortality estimated
868: from other sources like vital statistic data.
869:
870: The same imach parameter file can be used but the option for mle should be -3.
871:
1.324 ! brouard 872: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 873: former routines in order to include the new code within the former code.
874:
875: The output is very simple: only an estimate of the intercept and of
876: the slope with 95% confident intervals.
877:
878: Current limitations:
879: A) Even if you enter covariates, i.e. with the
880: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
881: B) There is no computation of Life Expectancy nor Life Table.
882:
883: Revision 1.97 2004/02/20 13:25:42 lievre
884: Version 0.96d. Population forecasting command line is (temporarily)
885: suppressed.
886:
887: Revision 1.96 2003/07/15 15:38:55 brouard
888: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
889: rewritten within the same printf. Workaround: many printfs.
890:
891: Revision 1.95 2003/07/08 07:54:34 brouard
892: * imach.c (Repository):
893: (Repository): Using imachwizard code to output a more meaningful covariance
894: matrix (cov(a12,c31) instead of numbers.
895:
896: Revision 1.94 2003/06/27 13:00:02 brouard
897: Just cleaning
898:
899: Revision 1.93 2003/06/25 16:33:55 brouard
900: (Module): On windows (cygwin) function asctime_r doesn't
901: exist so I changed back to asctime which exists.
902: (Module): Version 0.96b
903:
904: Revision 1.92 2003/06/25 16:30:45 brouard
905: (Module): On windows (cygwin) function asctime_r doesn't
906: exist so I changed back to asctime which exists.
907:
908: Revision 1.91 2003/06/25 15:30:29 brouard
909: * imach.c (Repository): Duplicated warning errors corrected.
910: (Repository): Elapsed time after each iteration is now output. It
911: helps to forecast when convergence will be reached. Elapsed time
912: is stamped in powell. We created a new html file for the graphs
913: concerning matrix of covariance. It has extension -cov.htm.
914:
915: Revision 1.90 2003/06/24 12:34:15 brouard
916: (Module): Some bugs corrected for windows. Also, when
917: mle=-1 a template is output in file "or"mypar.txt with the design
918: of the covariance matrix to be input.
919:
920: Revision 1.89 2003/06/24 12:30:52 brouard
921: (Module): Some bugs corrected for windows. Also, when
922: mle=-1 a template is output in file "or"mypar.txt with the design
923: of the covariance matrix to be input.
924:
925: Revision 1.88 2003/06/23 17:54:56 brouard
926: * 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.
927:
928: Revision 1.87 2003/06/18 12:26:01 brouard
929: Version 0.96
930:
931: Revision 1.86 2003/06/17 20:04:08 brouard
932: (Module): Change position of html and gnuplot routines and added
933: routine fileappend.
934:
935: Revision 1.85 2003/06/17 13:12:43 brouard
936: * imach.c (Repository): Check when date of death was earlier that
937: current date of interview. It may happen when the death was just
938: prior to the death. In this case, dh was negative and likelihood
939: was wrong (infinity). We still send an "Error" but patch by
940: assuming that the date of death was just one stepm after the
941: interview.
942: (Repository): Because some people have very long ID (first column)
943: we changed int to long in num[] and we added a new lvector for
944: memory allocation. But we also truncated to 8 characters (left
945: truncation)
946: (Repository): No more line truncation errors.
947:
948: Revision 1.84 2003/06/13 21:44:43 brouard
949: * imach.c (Repository): Replace "freqsummary" at a correct
950: place. It differs from routine "prevalence" which may be called
951: many times. Probs is memory consuming and must be used with
952: parcimony.
953: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
954:
955: Revision 1.83 2003/06/10 13:39:11 lievre
956: *** empty log message ***
957:
958: Revision 1.82 2003/06/05 15:57:20 brouard
959: Add log in imach.c and fullversion number is now printed.
960:
961: */
962: /*
963: Interpolated Markov Chain
964:
965: Short summary of the programme:
966:
1.227 brouard 967: This program computes Healthy Life Expectancies or State-specific
968: (if states aren't health statuses) Expectancies from
969: cross-longitudinal data. Cross-longitudinal data consist in:
970:
971: -1- a first survey ("cross") where individuals from different ages
972: are interviewed on their health status or degree of disability (in
973: the case of a health survey which is our main interest)
974:
975: -2- at least a second wave of interviews ("longitudinal") which
976: measure each change (if any) in individual health status. Health
977: expectancies are computed from the time spent in each health state
978: according to a model. More health states you consider, more time is
979: necessary to reach the Maximum Likelihood of the parameters involved
980: in the model. The simplest model is the multinomial logistic model
981: where pij is the probability to be observed in state j at the second
982: wave conditional to be observed in state i at the first
983: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
984: etc , where 'age' is age and 'sex' is a covariate. If you want to
985: have a more complex model than "constant and age", you should modify
986: the program where the markup *Covariates have to be included here
987: again* invites you to do it. More covariates you add, slower the
1.126 brouard 988: convergence.
989:
990: The advantage of this computer programme, compared to a simple
991: multinomial logistic model, is clear when the delay between waves is not
992: identical for each individual. Also, if a individual missed an
993: intermediate interview, the information is lost, but taken into
994: account using an interpolation or extrapolation.
995:
996: hPijx is the probability to be observed in state i at age x+h
997: conditional to the observed state i at age x. The delay 'h' can be
998: split into an exact number (nh*stepm) of unobserved intermediate
999: states. This elementary transition (by month, quarter,
1000: semester or year) is modelled as a multinomial logistic. The hPx
1001: matrix is simply the matrix product of nh*stepm elementary matrices
1002: and the contribution of each individual to the likelihood is simply
1003: hPijx.
1004:
1005: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1006: of the life expectancies. It also computes the period (stable) prevalence.
1007:
1008: Back prevalence and projections:
1.227 brouard 1009:
1010: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1011: double agemaxpar, double ftolpl, int *ncvyearp, double
1012: dateprev1,double dateprev2, int firstpass, int lastpass, int
1013: mobilavproj)
1014:
1015: Computes the back prevalence limit for any combination of
1016: covariate values k at any age between ageminpar and agemaxpar and
1017: returns it in **bprlim. In the loops,
1018:
1019: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1020: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1021:
1022: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1023: Computes for any combination of covariates k and any age between bage and fage
1024: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1025: oldm=oldms;savm=savms;
1.227 brouard 1026:
1.267 brouard 1027: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1028: Computes the transition matrix starting at age 'age' over
1029: 'nhstepm*hstepm*stepm' months (i.e. until
1030: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1031: nhstepm*hstepm matrices.
1032:
1033: Returns p3mat[i][j][h] after calling
1034: p3mat[i][j][h]=matprod2(newm,
1035: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1036: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1037: oldm);
1.226 brouard 1038:
1039: Important routines
1040:
1041: - func (or funcone), computes logit (pij) distinguishing
1042: o fixed variables (single or product dummies or quantitative);
1043: o varying variables by:
1044: (1) wave (single, product dummies, quantitative),
1045: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1046: % fixed dummy (treated) or quantitative (not done because time-consuming);
1047: % varying dummy (not done) or quantitative (not done);
1048: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1049: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1050: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.324 ! brouard 1051: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1052: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1053:
1.226 brouard 1054:
1055:
1.324 ! brouard 1056: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
! 1057: Institut national d'études démographiques, Paris.
1.126 brouard 1058: This software have been partly granted by Euro-REVES, a concerted action
1059: from the European Union.
1060: It is copyrighted identically to a GNU software product, ie programme and
1061: software can be distributed freely for non commercial use. Latest version
1062: can be accessed at http://euroreves.ined.fr/imach .
1063:
1064: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1065: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1066:
1067: **********************************************************************/
1068: /*
1069: main
1070: read parameterfile
1071: read datafile
1072: concatwav
1073: freqsummary
1074: if (mle >= 1)
1075: mlikeli
1076: print results files
1077: if mle==1
1078: computes hessian
1079: read end of parameter file: agemin, agemax, bage, fage, estepm
1080: begin-prev-date,...
1081: open gnuplot file
1082: open html file
1.145 brouard 1083: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1084: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1085: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1086: freexexit2 possible for memory heap.
1087:
1088: h Pij x | pij_nom ficrestpij
1089: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1090: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1091: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1092:
1093: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1094: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1095: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1096: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1097: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1098:
1.126 brouard 1099: forecasting if prevfcast==1 prevforecast call prevalence()
1100: health expectancies
1101: Variance-covariance of DFLE
1102: prevalence()
1103: movingaverage()
1104: varevsij()
1105: if popbased==1 varevsij(,popbased)
1106: total life expectancies
1107: Variance of period (stable) prevalence
1108: end
1109: */
1110:
1.187 brouard 1111: /* #define DEBUG */
1112: /* #define DEBUGBRENT */
1.203 brouard 1113: /* #define DEBUGLINMIN */
1114: /* #define DEBUGHESS */
1115: #define DEBUGHESSIJ
1.224 brouard 1116: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1117: #define POWELL /* Instead of NLOPT */
1.224 brouard 1118: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1119: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1120: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1121: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1122:
1123: #include <math.h>
1124: #include <stdio.h>
1125: #include <stdlib.h>
1126: #include <string.h>
1.226 brouard 1127: #include <ctype.h>
1.159 brouard 1128:
1129: #ifdef _WIN32
1130: #include <io.h>
1.172 brouard 1131: #include <windows.h>
1132: #include <tchar.h>
1.159 brouard 1133: #else
1.126 brouard 1134: #include <unistd.h>
1.159 brouard 1135: #endif
1.126 brouard 1136:
1137: #include <limits.h>
1138: #include <sys/types.h>
1.171 brouard 1139:
1140: #if defined(__GNUC__)
1141: #include <sys/utsname.h> /* Doesn't work on Windows */
1142: #endif
1143:
1.126 brouard 1144: #include <sys/stat.h>
1145: #include <errno.h>
1.159 brouard 1146: /* extern int errno; */
1.126 brouard 1147:
1.157 brouard 1148: /* #ifdef LINUX */
1149: /* #include <time.h> */
1150: /* #include "timeval.h" */
1151: /* #else */
1152: /* #include <sys/time.h> */
1153: /* #endif */
1154:
1.126 brouard 1155: #include <time.h>
1156:
1.136 brouard 1157: #ifdef GSL
1158: #include <gsl/gsl_errno.h>
1159: #include <gsl/gsl_multimin.h>
1160: #endif
1161:
1.167 brouard 1162:
1.162 brouard 1163: #ifdef NLOPT
1164: #include <nlopt.h>
1165: typedef struct {
1166: double (* function)(double [] );
1167: } myfunc_data ;
1168: #endif
1169:
1.126 brouard 1170: /* #include <libintl.h> */
1171: /* #define _(String) gettext (String) */
1172:
1.251 brouard 1173: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1174:
1175: #define GNUPLOTPROGRAM "gnuplot"
1176: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1177: #define FILENAMELENGTH 132
1178:
1179: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1180: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1181:
1.144 brouard 1182: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1183: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1184:
1185: #define NINTERVMAX 8
1.144 brouard 1186: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1187: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.318 brouard 1188: #define NCOVMAX 30 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1189: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1190: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1191: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1192: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1193: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1194: /* #define AGESUP 130 */
1.288 brouard 1195: /* #define AGESUP 150 */
1196: #define AGESUP 200
1.268 brouard 1197: #define AGEINF 0
1.218 brouard 1198: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1199: #define AGEBASE 40
1.194 brouard 1200: #define AGEOVERFLOW 1.e20
1.164 brouard 1201: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1202: #ifdef _WIN32
1203: #define DIRSEPARATOR '\\'
1204: #define CHARSEPARATOR "\\"
1205: #define ODIRSEPARATOR '/'
1206: #else
1.126 brouard 1207: #define DIRSEPARATOR '/'
1208: #define CHARSEPARATOR "/"
1209: #define ODIRSEPARATOR '\\'
1210: #endif
1211:
1.324 ! brouard 1212: /* $Id: imach.c,v 1.323 2022/07/22 12:30:08 brouard Exp $ */
1.126 brouard 1213: /* $State: Exp $ */
1.196 brouard 1214: #include "version.h"
1215: char version[]=__IMACH_VERSION__;
1.323 brouard 1216: char copyright[]="July 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";
1.324 ! brouard 1217: char fullversion[]="$Revision: 1.323 $ $Date: 2022/07/22 12:30:08 $";
1.126 brouard 1218: char strstart[80];
1219: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1220: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1221: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1222: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1223: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1224: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1225: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1226: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1227: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1228: int cptcovprodnoage=0; /**< Number of covariate products without age */
1229: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1230: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1231: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1232: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1233: int nsd=0; /**< Total number of single dummy variables (output) */
1234: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1235: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1236: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1237: int ntveff=0; /**< ntveff number of effective time varying variables */
1238: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1239: int cptcov=0; /* Working variable */
1.290 brouard 1240: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1241: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1242: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1243: int nlstate=2; /* Number of live states */
1244: int ndeath=1; /* Number of dead states */
1.130 brouard 1245: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1246: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1247: int popbased=0;
1248:
1249: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1250: int maxwav=0; /* Maxim number of waves */
1251: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1252: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1253: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1254: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1255: int mle=1, weightopt=0;
1.126 brouard 1256: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1257: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1258: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1259: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1260: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1261: int selected(int kvar); /* Is covariate kvar selected for printing results */
1262:
1.130 brouard 1263: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1264: double **matprod2(); /* test */
1.126 brouard 1265: double **oldm, **newm, **savm; /* Working pointers to matrices */
1266: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1267: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1268:
1.136 brouard 1269: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1270: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1271: FILE *ficlog, *ficrespow;
1.130 brouard 1272: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1273: double fretone; /* Only one call to likelihood */
1.130 brouard 1274: long ipmx=0; /* Number of contributions */
1.126 brouard 1275: double sw; /* Sum of weights */
1276: char filerespow[FILENAMELENGTH];
1277: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1278: FILE *ficresilk;
1279: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1280: FILE *ficresprobmorprev;
1281: FILE *fichtm, *fichtmcov; /* Html File */
1282: FILE *ficreseij;
1283: char filerese[FILENAMELENGTH];
1284: FILE *ficresstdeij;
1285: char fileresstde[FILENAMELENGTH];
1286: FILE *ficrescveij;
1287: char filerescve[FILENAMELENGTH];
1288: FILE *ficresvij;
1289: char fileresv[FILENAMELENGTH];
1.269 brouard 1290:
1.126 brouard 1291: char title[MAXLINE];
1.234 brouard 1292: char model[MAXLINE]; /**< The model line */
1.217 brouard 1293: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1294: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1295: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1296: char command[FILENAMELENGTH];
1297: int outcmd=0;
1298:
1.217 brouard 1299: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1300: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1301: char filelog[FILENAMELENGTH]; /* Log file */
1302: char filerest[FILENAMELENGTH];
1303: char fileregp[FILENAMELENGTH];
1304: char popfile[FILENAMELENGTH];
1305:
1306: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1307:
1.157 brouard 1308: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1309: /* struct timezone tzp; */
1310: /* extern int gettimeofday(); */
1311: struct tm tml, *gmtime(), *localtime();
1312:
1313: extern time_t time();
1314:
1315: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1316: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1317: struct tm tm;
1318:
1.126 brouard 1319: char strcurr[80], strfor[80];
1320:
1321: char *endptr;
1322: long lval;
1323: double dval;
1324:
1325: #define NR_END 1
1326: #define FREE_ARG char*
1327: #define FTOL 1.0e-10
1328:
1329: #define NRANSI
1.240 brouard 1330: #define ITMAX 200
1331: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1332:
1333: #define TOL 2.0e-4
1334:
1335: #define CGOLD 0.3819660
1336: #define ZEPS 1.0e-10
1337: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1338:
1339: #define GOLD 1.618034
1340: #define GLIMIT 100.0
1341: #define TINY 1.0e-20
1342:
1343: static double maxarg1,maxarg2;
1344: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1345: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1346:
1347: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1348: #define rint(a) floor(a+0.5)
1.166 brouard 1349: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1350: #define mytinydouble 1.0e-16
1.166 brouard 1351: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1352: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1353: /* static double dsqrarg; */
1354: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1355: static double sqrarg;
1356: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1357: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1358: int agegomp= AGEGOMP;
1359:
1360: int imx;
1361: int stepm=1;
1362: /* Stepm, step in month: minimum step interpolation*/
1363:
1364: int estepm;
1365: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1366:
1367: int m,nb;
1368: long *num;
1.197 brouard 1369: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1370: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1371: covariate for which somebody answered excluding
1372: undefined. Usually 2: 0 and 1. */
1373: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1374: covariate for which somebody answered including
1375: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1376: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1377: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1378: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1379: double *ageexmed,*agecens;
1380: double dateintmean=0;
1.296 brouard 1381: double anprojd, mprojd, jprojd; /* For eventual projections */
1382: double anprojf, mprojf, jprojf;
1.126 brouard 1383:
1.296 brouard 1384: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1385: double anbackf, mbackf, jbackf;
1386: double jintmean,mintmean,aintmean;
1.126 brouard 1387: double *weight;
1388: int **s; /* Status */
1.141 brouard 1389: double *agedc;
1.145 brouard 1390: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1391: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1392: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1393: double **coqvar; /* Fixed quantitative covariate nqv */
1394: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1395: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1396: double idx;
1397: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1398: /* Some documentation */
1399: /* Design original data
1400: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1401: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1402: * ntv=3 nqtv=1
1403: * cptcovn number of covariates (not including constant and age) = # of + plus 1 = 10+1=11
1404: * For time varying covariate, quanti or dummies
1405: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1406: * cotvar[wav][ntv+iv][i]= [3+(1 to nqtv)][i]=(V12) quanti
1407: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1408: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1409: * covar[k,i], value of kth fixed covariate dummy or quanti :
1410: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1411: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1412: * k= 1 2 3 4 5 6 7 8 9 10 11
1413: */
1414: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1415: /* 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
1416: # States 1=Coresidence, 2 Living alone, 3 Institution
1417: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1418: */
1.319 brouard 1419: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1420: /* k 1 2 3 4 5 6 7 8 9 */
1421: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1422: /* fixed or varying), 1 for age product, 2 for*/
1423: /* product */
1424: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1425: /*(single or product without age), 2 dummy*/
1426: /* with age product, 3 quant with age product*/
1427: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1428: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1429: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1430: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1431: /* nsq 1 2 */ /* Counting single quantit tv */
1432: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1433: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1434: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1435: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1436: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1437: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1438: /* 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 1439: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1440: /* Type */
1441: /* V 1 2 3 4 5 */
1442: /* F F V V V */
1443: /* D Q D D Q */
1444: /* */
1445: int *TvarsD;
1446: int *TvarsDind;
1447: int *TvarsQ;
1448: int *TvarsQind;
1449:
1.318 brouard 1450: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1451: int nresult=0;
1.258 brouard 1452: int parameterline=0; /* # of the parameter (type) line */
1.318 brouard 1453: int TKresult[MAXRESULTLINESPONE];
1454: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1455: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1456: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For dummy variable , variable # (output) */
1457: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1458: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1459: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , variable # (output) */
1460:
1461: /* 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
1462: # States 1=Coresidence, 2 Living alone, 3 Institution
1463: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1464: */
1.234 brouard 1465: /* 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 1466: 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 */
1467: 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 */
1468: 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 */
1469: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1470: 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 */
1471: 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 1472: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1473: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1474: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1475: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1476: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1477: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1478: 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 */
1479: 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 */
1480:
1.230 brouard 1481: int *Tvarsel; /**< Selected covariates for output */
1482: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1483: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1484: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1485: 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 1486: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1487: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1488: int *Tage;
1.227 brouard 1489: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1490: 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 1491: 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*/
1492: 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 1493: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1494: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1495: int **Tvard;
1496: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1497: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1498: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1499: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1500: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1501: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1502: double *lsurv, *lpop, *tpop;
1503:
1.231 brouard 1504: #define FD 1; /* Fixed dummy covariate */
1505: #define FQ 2; /* Fixed quantitative covariate */
1506: #define FP 3; /* Fixed product covariate */
1507: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1508: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1509: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1510: #define VD 10; /* Varying dummy covariate */
1511: #define VQ 11; /* Varying quantitative covariate */
1512: #define VP 12; /* Varying product covariate */
1513: #define VPDD 13; /* Varying product dummy*dummy covariate */
1514: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1515: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1516: #define APFD 16; /* Age product * fixed dummy covariate */
1517: #define APFQ 17; /* Age product * fixed quantitative covariate */
1518: #define APVD 18; /* Age product * varying dummy covariate */
1519: #define APVQ 19; /* Age product * varying quantitative covariate */
1520:
1521: #define FTYPE 1; /* Fixed covariate */
1522: #define VTYPE 2; /* Varying covariate (loop in wave) */
1523: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1524:
1525: struct kmodel{
1526: int maintype; /* main type */
1527: int subtype; /* subtype */
1528: };
1529: struct kmodel modell[NCOVMAX];
1530:
1.143 brouard 1531: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1532: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1533:
1534: /**************** split *************************/
1535: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1536: {
1537: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1538: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1539: */
1540: char *ss; /* pointer */
1.186 brouard 1541: int l1=0, l2=0; /* length counters */
1.126 brouard 1542:
1543: l1 = strlen(path ); /* length of path */
1544: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1545: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1546: if ( ss == NULL ) { /* no directory, so determine current directory */
1547: strcpy( name, path ); /* we got the fullname name because no directory */
1548: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1549: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1550: /* get current working directory */
1551: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1552: #ifdef WIN32
1553: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1554: #else
1555: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1556: #endif
1.126 brouard 1557: return( GLOCK_ERROR_GETCWD );
1558: }
1559: /* got dirc from getcwd*/
1560: printf(" DIRC = %s \n",dirc);
1.205 brouard 1561: } else { /* strip directory from path */
1.126 brouard 1562: ss++; /* after this, the filename */
1563: l2 = strlen( ss ); /* length of filename */
1564: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1565: strcpy( name, ss ); /* save file name */
1566: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1567: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1568: printf(" DIRC2 = %s \n",dirc);
1569: }
1570: /* We add a separator at the end of dirc if not exists */
1571: l1 = strlen( dirc ); /* length of directory */
1572: if( dirc[l1-1] != DIRSEPARATOR ){
1573: dirc[l1] = DIRSEPARATOR;
1574: dirc[l1+1] = 0;
1575: printf(" DIRC3 = %s \n",dirc);
1576: }
1577: ss = strrchr( name, '.' ); /* find last / */
1578: if (ss >0){
1579: ss++;
1580: strcpy(ext,ss); /* save extension */
1581: l1= strlen( name);
1582: l2= strlen(ss)+1;
1583: strncpy( finame, name, l1-l2);
1584: finame[l1-l2]= 0;
1585: }
1586:
1587: return( 0 ); /* we're done */
1588: }
1589:
1590:
1591: /******************************************/
1592:
1593: void replace_back_to_slash(char *s, char*t)
1594: {
1595: int i;
1596: int lg=0;
1597: i=0;
1598: lg=strlen(t);
1599: for(i=0; i<= lg; i++) {
1600: (s[i] = t[i]);
1601: if (t[i]== '\\') s[i]='/';
1602: }
1603: }
1604:
1.132 brouard 1605: char *trimbb(char *out, char *in)
1.137 brouard 1606: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1607: char *s;
1608: s=out;
1609: while (*in != '\0'){
1.137 brouard 1610: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1611: in++;
1612: }
1613: *out++ = *in++;
1614: }
1615: *out='\0';
1616: return s;
1617: }
1618:
1.187 brouard 1619: /* char *substrchaine(char *out, char *in, char *chain) */
1620: /* { */
1621: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1622: /* char *s, *t; */
1623: /* t=in;s=out; */
1624: /* while ((*in != *chain) && (*in != '\0')){ */
1625: /* *out++ = *in++; */
1626: /* } */
1627:
1628: /* /\* *in matches *chain *\/ */
1629: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1630: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1631: /* } */
1632: /* in--; chain--; */
1633: /* while ( (*in != '\0')){ */
1634: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1635: /* *out++ = *in++; */
1636: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1637: /* } */
1638: /* *out='\0'; */
1639: /* out=s; */
1640: /* return out; */
1641: /* } */
1642: char *substrchaine(char *out, char *in, char *chain)
1643: {
1644: /* Substract chain 'chain' from 'in', return and output 'out' */
1645: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1646:
1647: char *strloc;
1648:
1649: strcpy (out, in);
1650: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1651: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1652: if(strloc != NULL){
1653: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1654: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1655: /* strcpy (strloc, strloc +strlen(chain));*/
1656: }
1657: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1658: return out;
1659: }
1660:
1661:
1.145 brouard 1662: char *cutl(char *blocc, char *alocc, char *in, char occ)
1663: {
1.187 brouard 1664: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1665: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1666: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1667: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1668: */
1.160 brouard 1669: char *s, *t;
1.145 brouard 1670: t=in;s=in;
1671: while ((*in != occ) && (*in != '\0')){
1672: *alocc++ = *in++;
1673: }
1674: if( *in == occ){
1675: *(alocc)='\0';
1676: s=++in;
1677: }
1678:
1679: if (s == t) {/* occ not found */
1680: *(alocc-(in-s))='\0';
1681: in=s;
1682: }
1683: while ( *in != '\0'){
1684: *blocc++ = *in++;
1685: }
1686:
1687: *blocc='\0';
1688: return t;
1689: }
1.137 brouard 1690: char *cutv(char *blocc, char *alocc, char *in, char occ)
1691: {
1.187 brouard 1692: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1693: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1694: gives blocc="abcdef2ghi" and alocc="j".
1695: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1696: */
1697: char *s, *t;
1698: t=in;s=in;
1699: while (*in != '\0'){
1700: while( *in == occ){
1701: *blocc++ = *in++;
1702: s=in;
1703: }
1704: *blocc++ = *in++;
1705: }
1706: if (s == t) /* occ not found */
1707: *(blocc-(in-s))='\0';
1708: else
1709: *(blocc-(in-s)-1)='\0';
1710: in=s;
1711: while ( *in != '\0'){
1712: *alocc++ = *in++;
1713: }
1714:
1715: *alocc='\0';
1716: return s;
1717: }
1718:
1.126 brouard 1719: int nbocc(char *s, char occ)
1720: {
1721: int i,j=0;
1722: int lg=20;
1723: i=0;
1724: lg=strlen(s);
1725: for(i=0; i<= lg; i++) {
1.234 brouard 1726: if (s[i] == occ ) j++;
1.126 brouard 1727: }
1728: return j;
1729: }
1730:
1.137 brouard 1731: /* void cutv(char *u,char *v, char*t, char occ) */
1732: /* { */
1733: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1734: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1735: /* gives u="abcdef2ghi" and v="j" *\/ */
1736: /* int i,lg,j,p=0; */
1737: /* i=0; */
1738: /* lg=strlen(t); */
1739: /* for(j=0; j<=lg-1; j++) { */
1740: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1741: /* } */
1.126 brouard 1742:
1.137 brouard 1743: /* for(j=0; j<p; j++) { */
1744: /* (u[j] = t[j]); */
1745: /* } */
1746: /* u[p]='\0'; */
1.126 brouard 1747:
1.137 brouard 1748: /* for(j=0; j<= lg; j++) { */
1749: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1750: /* } */
1751: /* } */
1.126 brouard 1752:
1.160 brouard 1753: #ifdef _WIN32
1754: char * strsep(char **pp, const char *delim)
1755: {
1756: char *p, *q;
1757:
1758: if ((p = *pp) == NULL)
1759: return 0;
1760: if ((q = strpbrk (p, delim)) != NULL)
1761: {
1762: *pp = q + 1;
1763: *q = '\0';
1764: }
1765: else
1766: *pp = 0;
1767: return p;
1768: }
1769: #endif
1770:
1.126 brouard 1771: /********************** nrerror ********************/
1772:
1773: void nrerror(char error_text[])
1774: {
1775: fprintf(stderr,"ERREUR ...\n");
1776: fprintf(stderr,"%s\n",error_text);
1777: exit(EXIT_FAILURE);
1778: }
1779: /*********************** vector *******************/
1780: double *vector(int nl, int nh)
1781: {
1782: double *v;
1783: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1784: if (!v) nrerror("allocation failure in vector");
1785: return v-nl+NR_END;
1786: }
1787:
1788: /************************ free vector ******************/
1789: void free_vector(double*v, int nl, int nh)
1790: {
1791: free((FREE_ARG)(v+nl-NR_END));
1792: }
1793:
1794: /************************ivector *******************************/
1795: int *ivector(long nl,long nh)
1796: {
1797: int *v;
1798: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1799: if (!v) nrerror("allocation failure in ivector");
1800: return v-nl+NR_END;
1801: }
1802:
1803: /******************free ivector **************************/
1804: void free_ivector(int *v, long nl, long nh)
1805: {
1806: free((FREE_ARG)(v+nl-NR_END));
1807: }
1808:
1809: /************************lvector *******************************/
1810: long *lvector(long nl,long nh)
1811: {
1812: long *v;
1813: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1814: if (!v) nrerror("allocation failure in ivector");
1815: return v-nl+NR_END;
1816: }
1817:
1818: /******************free lvector **************************/
1819: void free_lvector(long *v, long nl, long nh)
1820: {
1821: free((FREE_ARG)(v+nl-NR_END));
1822: }
1823:
1824: /******************* imatrix *******************************/
1825: int **imatrix(long nrl, long nrh, long ncl, long nch)
1826: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1827: {
1828: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1829: int **m;
1830:
1831: /* allocate pointers to rows */
1832: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1833: if (!m) nrerror("allocation failure 1 in matrix()");
1834: m += NR_END;
1835: m -= nrl;
1836:
1837:
1838: /* allocate rows and set pointers to them */
1839: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1840: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1841: m[nrl] += NR_END;
1842: m[nrl] -= ncl;
1843:
1844: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1845:
1846: /* return pointer to array of pointers to rows */
1847: return m;
1848: }
1849:
1850: /****************** free_imatrix *************************/
1851: void free_imatrix(m,nrl,nrh,ncl,nch)
1852: int **m;
1853: long nch,ncl,nrh,nrl;
1854: /* free an int matrix allocated by imatrix() */
1855: {
1856: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1857: free((FREE_ARG) (m+nrl-NR_END));
1858: }
1859:
1860: /******************* matrix *******************************/
1861: double **matrix(long nrl, long nrh, long ncl, long nch)
1862: {
1863: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1864: double **m;
1865:
1866: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1867: if (!m) nrerror("allocation failure 1 in matrix()");
1868: m += NR_END;
1869: m -= nrl;
1870:
1871: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1872: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1873: m[nrl] += NR_END;
1874: m[nrl] -= ncl;
1875:
1876: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1877: return m;
1.145 brouard 1878: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1879: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1880: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1881: */
1882: }
1883:
1884: /*************************free matrix ************************/
1885: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1886: {
1887: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1888: free((FREE_ARG)(m+nrl-NR_END));
1889: }
1890:
1891: /******************* ma3x *******************************/
1892: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1893: {
1894: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1895: double ***m;
1896:
1897: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1898: if (!m) nrerror("allocation failure 1 in matrix()");
1899: m += NR_END;
1900: m -= nrl;
1901:
1902: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1903: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1904: m[nrl] += NR_END;
1905: m[nrl] -= ncl;
1906:
1907: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1908:
1909: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1910: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1911: m[nrl][ncl] += NR_END;
1912: m[nrl][ncl] -= nll;
1913: for (j=ncl+1; j<=nch; j++)
1914: m[nrl][j]=m[nrl][j-1]+nlay;
1915:
1916: for (i=nrl+1; i<=nrh; i++) {
1917: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1918: for (j=ncl+1; j<=nch; j++)
1919: m[i][j]=m[i][j-1]+nlay;
1920: }
1921: return m;
1922: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1923: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1924: */
1925: }
1926:
1927: /*************************free ma3x ************************/
1928: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1929: {
1930: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1931: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1932: free((FREE_ARG)(m+nrl-NR_END));
1933: }
1934:
1935: /*************** function subdirf ***********/
1936: char *subdirf(char fileres[])
1937: {
1938: /* Caution optionfilefiname is hidden */
1939: strcpy(tmpout,optionfilefiname);
1940: strcat(tmpout,"/"); /* Add to the right */
1941: strcat(tmpout,fileres);
1942: return tmpout;
1943: }
1944:
1945: /*************** function subdirf2 ***********/
1946: char *subdirf2(char fileres[], char *preop)
1947: {
1.314 brouard 1948: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
1949: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 1950: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 1951: /* Caution optionfilefiname is hidden */
1952: strcpy(tmpout,optionfilefiname);
1953: strcat(tmpout,"/");
1954: strcat(tmpout,preop);
1955: strcat(tmpout,fileres);
1956: return tmpout;
1957: }
1958:
1959: /*************** function subdirf3 ***********/
1960: char *subdirf3(char fileres[], char *preop, char *preop2)
1961: {
1962:
1963: /* Caution optionfilefiname is hidden */
1964: strcpy(tmpout,optionfilefiname);
1965: strcat(tmpout,"/");
1966: strcat(tmpout,preop);
1967: strcat(tmpout,preop2);
1968: strcat(tmpout,fileres);
1969: return tmpout;
1970: }
1.213 brouard 1971:
1972: /*************** function subdirfext ***********/
1973: char *subdirfext(char fileres[], char *preop, char *postop)
1974: {
1975:
1976: strcpy(tmpout,preop);
1977: strcat(tmpout,fileres);
1978: strcat(tmpout,postop);
1979: return tmpout;
1980: }
1.126 brouard 1981:
1.213 brouard 1982: /*************** function subdirfext3 ***********/
1983: char *subdirfext3(char fileres[], char *preop, char *postop)
1984: {
1985:
1986: /* Caution optionfilefiname is hidden */
1987: strcpy(tmpout,optionfilefiname);
1988: strcat(tmpout,"/");
1989: strcat(tmpout,preop);
1990: strcat(tmpout,fileres);
1991: strcat(tmpout,postop);
1992: return tmpout;
1993: }
1994:
1.162 brouard 1995: char *asc_diff_time(long time_sec, char ascdiff[])
1996: {
1997: long sec_left, days, hours, minutes;
1998: days = (time_sec) / (60*60*24);
1999: sec_left = (time_sec) % (60*60*24);
2000: hours = (sec_left) / (60*60) ;
2001: sec_left = (sec_left) %(60*60);
2002: minutes = (sec_left) /60;
2003: sec_left = (sec_left) % (60);
2004: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2005: return ascdiff;
2006: }
2007:
1.126 brouard 2008: /***************** f1dim *************************/
2009: extern int ncom;
2010: extern double *pcom,*xicom;
2011: extern double (*nrfunc)(double []);
2012:
2013: double f1dim(double x)
2014: {
2015: int j;
2016: double f;
2017: double *xt;
2018:
2019: xt=vector(1,ncom);
2020: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2021: f=(*nrfunc)(xt);
2022: free_vector(xt,1,ncom);
2023: return f;
2024: }
2025:
2026: /*****************brent *************************/
2027: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2028: {
2029: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2030: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2031: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2032: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2033: * returned function value.
2034: */
1.126 brouard 2035: int iter;
2036: double a,b,d,etemp;
1.159 brouard 2037: double fu=0,fv,fw,fx;
1.164 brouard 2038: double ftemp=0.;
1.126 brouard 2039: double p,q,r,tol1,tol2,u,v,w,x,xm;
2040: double e=0.0;
2041:
2042: a=(ax < cx ? ax : cx);
2043: b=(ax > cx ? ax : cx);
2044: x=w=v=bx;
2045: fw=fv=fx=(*f)(x);
2046: for (iter=1;iter<=ITMAX;iter++) {
2047: xm=0.5*(a+b);
2048: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2049: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2050: printf(".");fflush(stdout);
2051: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2052: #ifdef DEBUGBRENT
1.126 brouard 2053: 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);
2054: 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);
2055: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2056: #endif
2057: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2058: *xmin=x;
2059: return fx;
2060: }
2061: ftemp=fu;
2062: if (fabs(e) > tol1) {
2063: r=(x-w)*(fx-fv);
2064: q=(x-v)*(fx-fw);
2065: p=(x-v)*q-(x-w)*r;
2066: q=2.0*(q-r);
2067: if (q > 0.0) p = -p;
2068: q=fabs(q);
2069: etemp=e;
2070: e=d;
2071: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2072: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2073: else {
1.224 brouard 2074: d=p/q;
2075: u=x+d;
2076: if (u-a < tol2 || b-u < tol2)
2077: d=SIGN(tol1,xm-x);
1.126 brouard 2078: }
2079: } else {
2080: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2081: }
2082: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2083: fu=(*f)(u);
2084: if (fu <= fx) {
2085: if (u >= x) a=x; else b=x;
2086: SHFT(v,w,x,u)
1.183 brouard 2087: SHFT(fv,fw,fx,fu)
2088: } else {
2089: if (u < x) a=u; else b=u;
2090: if (fu <= fw || w == x) {
1.224 brouard 2091: v=w;
2092: w=u;
2093: fv=fw;
2094: fw=fu;
1.183 brouard 2095: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2096: v=u;
2097: fv=fu;
1.183 brouard 2098: }
2099: }
1.126 brouard 2100: }
2101: nrerror("Too many iterations in brent");
2102: *xmin=x;
2103: return fx;
2104: }
2105:
2106: /****************** mnbrak ***********************/
2107:
2108: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2109: double (*func)(double))
1.183 brouard 2110: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2111: the downhill direction (defined by the function as evaluated at the initial points) and returns
2112: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2113: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2114: */
1.126 brouard 2115: double ulim,u,r,q, dum;
2116: double fu;
1.187 brouard 2117:
2118: double scale=10.;
2119: int iterscale=0;
2120:
2121: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2122: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2123:
2124:
2125: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2126: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2127: /* *bx = *ax - (*ax - *bx)/scale; */
2128: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2129: /* } */
2130:
1.126 brouard 2131: if (*fb > *fa) {
2132: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2133: SHFT(dum,*fb,*fa,dum)
2134: }
1.126 brouard 2135: *cx=(*bx)+GOLD*(*bx-*ax);
2136: *fc=(*func)(*cx);
1.183 brouard 2137: #ifdef DEBUG
1.224 brouard 2138: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2139: 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 2140: #endif
1.224 brouard 2141: 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 2142: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2143: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2144: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2145: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2146: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2147: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2148: fu=(*func)(u);
1.163 brouard 2149: #ifdef DEBUG
2150: /* f(x)=A(x-u)**2+f(u) */
2151: double A, fparabu;
2152: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2153: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2154: 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);
2155: 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 2156: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2157: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2158: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2159: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2160: #endif
1.184 brouard 2161: #ifdef MNBRAKORIGINAL
1.183 brouard 2162: #else
1.191 brouard 2163: /* if (fu > *fc) { */
2164: /* #ifdef DEBUG */
2165: /* printf("mnbrak4 fu > fc \n"); */
2166: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2167: /* #endif */
2168: /* /\* 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 *\\/ *\/ */
2169: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2170: /* dum=u; /\* Shifting c and u *\/ */
2171: /* u = *cx; */
2172: /* *cx = dum; */
2173: /* dum = fu; */
2174: /* fu = *fc; */
2175: /* *fc =dum; */
2176: /* } else { /\* end *\/ */
2177: /* #ifdef DEBUG */
2178: /* printf("mnbrak3 fu < fc \n"); */
2179: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2180: /* #endif */
2181: /* dum=u; /\* Shifting c and u *\/ */
2182: /* u = *cx; */
2183: /* *cx = dum; */
2184: /* dum = fu; */
2185: /* fu = *fc; */
2186: /* *fc =dum; */
2187: /* } */
1.224 brouard 2188: #ifdef DEBUGMNBRAK
2189: double A, fparabu;
2190: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2191: fparabu= *fa - A*(*ax-u)*(*ax-u);
2192: 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);
2193: 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 2194: #endif
1.191 brouard 2195: dum=u; /* Shifting c and u */
2196: u = *cx;
2197: *cx = dum;
2198: dum = fu;
2199: fu = *fc;
2200: *fc =dum;
1.183 brouard 2201: #endif
1.162 brouard 2202: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2203: #ifdef DEBUG
1.224 brouard 2204: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2205: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2206: #endif
1.126 brouard 2207: fu=(*func)(u);
2208: if (fu < *fc) {
1.183 brouard 2209: #ifdef DEBUG
1.224 brouard 2210: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2211: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2212: #endif
2213: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2214: SHFT(*fb,*fc,fu,(*func)(u))
2215: #ifdef DEBUG
2216: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2217: #endif
2218: }
1.162 brouard 2219: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2220: #ifdef DEBUG
1.224 brouard 2221: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2222: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2223: #endif
1.126 brouard 2224: u=ulim;
2225: fu=(*func)(u);
1.183 brouard 2226: } else { /* u could be left to b (if r > q parabola has a maximum) */
2227: #ifdef DEBUG
1.224 brouard 2228: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2229: 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 2230: #endif
1.126 brouard 2231: u=(*cx)+GOLD*(*cx-*bx);
2232: fu=(*func)(u);
1.224 brouard 2233: #ifdef DEBUG
2234: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2235: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2236: #endif
1.183 brouard 2237: } /* end tests */
1.126 brouard 2238: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2239: SHFT(*fa,*fb,*fc,fu)
2240: #ifdef DEBUG
1.224 brouard 2241: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2242: 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 2243: #endif
2244: } /* 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 2245: }
2246:
2247: /*************** linmin ************************/
1.162 brouard 2248: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2249: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2250: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2251: the value of func at the returned location p . This is actually all accomplished by calling the
2252: routines mnbrak and brent .*/
1.126 brouard 2253: int ncom;
2254: double *pcom,*xicom;
2255: double (*nrfunc)(double []);
2256:
1.224 brouard 2257: #ifdef LINMINORIGINAL
1.126 brouard 2258: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2259: #else
2260: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2261: #endif
1.126 brouard 2262: {
2263: double brent(double ax, double bx, double cx,
2264: double (*f)(double), double tol, double *xmin);
2265: double f1dim(double x);
2266: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2267: double *fc, double (*func)(double));
2268: int j;
2269: double xx,xmin,bx,ax;
2270: double fx,fb,fa;
1.187 brouard 2271:
1.203 brouard 2272: #ifdef LINMINORIGINAL
2273: #else
2274: double scale=10., axs, xxs; /* Scale added for infinity */
2275: #endif
2276:
1.126 brouard 2277: ncom=n;
2278: pcom=vector(1,n);
2279: xicom=vector(1,n);
2280: nrfunc=func;
2281: for (j=1;j<=n;j++) {
2282: pcom[j]=p[j];
1.202 brouard 2283: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2284: }
1.187 brouard 2285:
1.203 brouard 2286: #ifdef LINMINORIGINAL
2287: xx=1.;
2288: #else
2289: axs=0.0;
2290: xxs=1.;
2291: do{
2292: xx= xxs;
2293: #endif
1.187 brouard 2294: ax=0.;
2295: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2296: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2297: /* 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)) */
2298: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2299: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2300: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2301: /* 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 2302: #ifdef LINMINORIGINAL
2303: #else
2304: if (fx != fx){
1.224 brouard 2305: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2306: printf("|");
2307: fprintf(ficlog,"|");
1.203 brouard 2308: #ifdef DEBUGLINMIN
1.224 brouard 2309: 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 2310: #endif
2311: }
1.224 brouard 2312: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2313: #endif
2314:
1.191 brouard 2315: #ifdef DEBUGLINMIN
2316: 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 2317: 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 2318: #endif
1.224 brouard 2319: #ifdef LINMINORIGINAL
2320: #else
1.317 brouard 2321: if(fb == fx){ /* Flat function in the direction */
2322: xmin=xx;
1.224 brouard 2323: *flat=1;
1.317 brouard 2324: }else{
1.224 brouard 2325: *flat=0;
2326: #endif
2327: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2328: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2329: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2330: /* fmin = f(p[j] + xmin * xi[j]) */
2331: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2332: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2333: #ifdef DEBUG
1.224 brouard 2334: 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);
2335: 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);
2336: #endif
2337: #ifdef LINMINORIGINAL
2338: #else
2339: }
1.126 brouard 2340: #endif
1.191 brouard 2341: #ifdef DEBUGLINMIN
2342: printf("linmin end ");
1.202 brouard 2343: fprintf(ficlog,"linmin end ");
1.191 brouard 2344: #endif
1.126 brouard 2345: for (j=1;j<=n;j++) {
1.203 brouard 2346: #ifdef LINMINORIGINAL
2347: xi[j] *= xmin;
2348: #else
2349: #ifdef DEBUGLINMIN
2350: if(xxs <1.0)
2351: printf(" before xi[%d]=%12.8f", j,xi[j]);
2352: #endif
2353: 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) */
2354: #ifdef DEBUGLINMIN
2355: if(xxs <1.0)
2356: 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 );
2357: #endif
2358: #endif
1.187 brouard 2359: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2360: }
1.191 brouard 2361: #ifdef DEBUGLINMIN
1.203 brouard 2362: printf("\n");
1.191 brouard 2363: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2364: 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 2365: for (j=1;j<=n;j++) {
1.202 brouard 2366: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2367: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2368: if(j % ncovmodel == 0){
1.191 brouard 2369: printf("\n");
1.202 brouard 2370: fprintf(ficlog,"\n");
2371: }
1.191 brouard 2372: }
1.203 brouard 2373: #else
1.191 brouard 2374: #endif
1.126 brouard 2375: free_vector(xicom,1,n);
2376: free_vector(pcom,1,n);
2377: }
2378:
2379:
2380: /*************** powell ************************/
1.162 brouard 2381: /*
1.317 brouard 2382: Minimization of a function func of n variables. Input consists in an initial starting point
2383: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2384: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2385: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2386: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2387: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2388: */
1.224 brouard 2389: #ifdef LINMINORIGINAL
2390: #else
2391: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2392: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2393: #endif
1.126 brouard 2394: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2395: double (*func)(double []))
2396: {
1.224 brouard 2397: #ifdef LINMINORIGINAL
2398: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2399: double (*func)(double []));
1.224 brouard 2400: #else
1.241 brouard 2401: void linmin(double p[], double xi[], int n, double *fret,
2402: double (*func)(double []),int *flat);
1.224 brouard 2403: #endif
1.239 brouard 2404: int i,ibig,j,jk,k;
1.126 brouard 2405: double del,t,*pt,*ptt,*xit;
1.181 brouard 2406: double directest;
1.126 brouard 2407: double fp,fptt;
2408: double *xits;
2409: int niterf, itmp;
2410:
2411: pt=vector(1,n);
2412: ptt=vector(1,n);
2413: xit=vector(1,n);
2414: xits=vector(1,n);
2415: *fret=(*func)(p);
2416: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2417: rcurr_time = time(NULL);
1.126 brouard 2418: for (*iter=1;;++(*iter)) {
2419: ibig=0;
2420: del=0.0;
1.157 brouard 2421: rlast_time=rcurr_time;
2422: /* (void) gettimeofday(&curr_time,&tzp); */
2423: rcurr_time = time(NULL);
2424: curr_time = *localtime(&rcurr_time);
1.324 ! brouard 2425: 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);
! 2426: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f gain=%.12f=%.3g %ld sec. %ld sec.",*iter,*fret, fp-*fret,fp-*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
1.157 brouard 2427: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 ! brouard 2428: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2429: for (i=1;i<=n;i++) {
1.126 brouard 2430: fprintf(ficrespow," %.12lf", p[i]);
2431: }
1.239 brouard 2432: fprintf(ficrespow,"\n");fflush(ficrespow);
2433: printf("\n#model= 1 + age ");
2434: fprintf(ficlog,"\n#model= 1 + age ");
2435: if(nagesqr==1){
1.241 brouard 2436: printf(" + age*age ");
2437: fprintf(ficlog," + age*age ");
1.239 brouard 2438: }
2439: for(j=1;j <=ncovmodel-2;j++){
2440: if(Typevar[j]==0) {
2441: printf(" + V%d ",Tvar[j]);
2442: fprintf(ficlog," + V%d ",Tvar[j]);
2443: }else if(Typevar[j]==1) {
2444: printf(" + V%d*age ",Tvar[j]);
2445: fprintf(ficlog," + V%d*age ",Tvar[j]);
2446: }else if(Typevar[j]==2) {
2447: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2448: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2449: }
2450: }
1.126 brouard 2451: printf("\n");
1.239 brouard 2452: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2453: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2454: fprintf(ficlog,"\n");
1.239 brouard 2455: for(i=1,jk=1; i <=nlstate; i++){
2456: for(k=1; k <=(nlstate+ndeath); k++){
2457: if (k != i) {
2458: printf("%d%d ",i,k);
2459: fprintf(ficlog,"%d%d ",i,k);
2460: for(j=1; j <=ncovmodel; j++){
2461: printf("%12.7f ",p[jk]);
2462: fprintf(ficlog,"%12.7f ",p[jk]);
2463: jk++;
2464: }
2465: printf("\n");
2466: fprintf(ficlog,"\n");
2467: }
2468: }
2469: }
1.241 brouard 2470: if(*iter <=3 && *iter >1){
1.157 brouard 2471: tml = *localtime(&rcurr_time);
2472: strcpy(strcurr,asctime(&tml));
2473: rforecast_time=rcurr_time;
1.126 brouard 2474: itmp = strlen(strcurr);
2475: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2476: strcurr[itmp-1]='\0';
1.162 brouard 2477: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2478: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2479: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2480: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2481: forecast_time = *localtime(&rforecast_time);
2482: strcpy(strfor,asctime(&forecast_time));
2483: itmp = strlen(strfor);
2484: if(strfor[itmp-1]=='\n')
2485: strfor[itmp-1]='\0';
2486: 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);
2487: 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 2488: }
2489: }
1.187 brouard 2490: for (i=1;i<=n;i++) { /* For each direction i */
2491: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2492: fptt=(*fret);
2493: #ifdef DEBUG
1.203 brouard 2494: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2495: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2496: #endif
1.203 brouard 2497: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2498: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2499: #ifdef LINMINORIGINAL
1.188 brouard 2500: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2501: #else
2502: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2503: flatdir[i]=flat; /* Function is vanishing in that direction i */
2504: #endif
2505: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2506: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2507: /* because that direction will be replaced unless the gain del is small */
2508: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2509: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2510: /* with the new direction. */
2511: del=fabs(fptt-(*fret));
2512: ibig=i;
1.126 brouard 2513: }
2514: #ifdef DEBUG
2515: printf("%d %.12e",i,(*fret));
2516: fprintf(ficlog,"%d %.12e",i,(*fret));
2517: for (j=1;j<=n;j++) {
1.224 brouard 2518: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2519: printf(" x(%d)=%.12e",j,xit[j]);
2520: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2521: }
2522: for(j=1;j<=n;j++) {
1.225 brouard 2523: printf(" p(%d)=%.12e",j,p[j]);
2524: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2525: }
2526: printf("\n");
2527: fprintf(ficlog,"\n");
2528: #endif
1.187 brouard 2529: } /* end loop on each direction i */
2530: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2531: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2532: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2533: for(j=1;j<=n;j++) {
2534: if(flatdir[j] >0){
2535: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2536: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2537: }
1.319 brouard 2538: /* printf("\n"); */
2539: /* fprintf(ficlog,"\n"); */
2540: }
1.243 brouard 2541: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2542: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2543: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2544: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2545: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2546: /* decreased of more than 3.84 */
2547: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2548: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2549: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2550:
1.188 brouard 2551: /* Starting the program with initial values given by a former maximization will simply change */
2552: /* the scales of the directions and the directions, because the are reset to canonical directions */
2553: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2554: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2555: #ifdef DEBUG
2556: int k[2],l;
2557: k[0]=1;
2558: k[1]=-1;
2559: printf("Max: %.12e",(*func)(p));
2560: fprintf(ficlog,"Max: %.12e",(*func)(p));
2561: for (j=1;j<=n;j++) {
2562: printf(" %.12e",p[j]);
2563: fprintf(ficlog," %.12e",p[j]);
2564: }
2565: printf("\n");
2566: fprintf(ficlog,"\n");
2567: for(l=0;l<=1;l++) {
2568: for (j=1;j<=n;j++) {
2569: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2570: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2571: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2572: }
2573: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2574: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2575: }
2576: #endif
2577:
2578: free_vector(xit,1,n);
2579: free_vector(xits,1,n);
2580: free_vector(ptt,1,n);
2581: free_vector(pt,1,n);
2582: return;
1.192 brouard 2583: } /* enough precision */
1.240 brouard 2584: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2585: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2586: ptt[j]=2.0*p[j]-pt[j];
2587: xit[j]=p[j]-pt[j];
2588: pt[j]=p[j];
2589: }
1.181 brouard 2590: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2591: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2592: if (*iter <=4) {
1.225 brouard 2593: #else
2594: #endif
1.224 brouard 2595: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2596: #else
1.161 brouard 2597: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2598: #endif
1.162 brouard 2599: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2600: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2601: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2602: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2603: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2604: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2605: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2606: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2607: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2608: /* Even if f3 <f1, directest can be negative and t >0 */
2609: /* mu² and del² are equal when f3=f1 */
2610: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2611: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2612: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2613: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2614: #ifdef NRCORIGINAL
2615: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2616: #else
2617: 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 2618: t= t- del*SQR(fp-fptt);
1.183 brouard 2619: #endif
1.202 brouard 2620: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2621: #ifdef DEBUG
1.181 brouard 2622: 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);
2623: 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 2624: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2625: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2626: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2627: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2628: 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);
2629: 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);
2630: #endif
1.183 brouard 2631: #ifdef POWELLORIGINAL
2632: if (t < 0.0) { /* Then we use it for new direction */
2633: #else
1.182 brouard 2634: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2635: 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 2636: 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 2637: 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 2638: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2639: }
1.181 brouard 2640: if (directest < 0.0) { /* Then we use it for new direction */
2641: #endif
1.191 brouard 2642: #ifdef DEBUGLINMIN
1.234 brouard 2643: printf("Before linmin in direction P%d-P0\n",n);
2644: for (j=1;j<=n;j++) {
2645: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2646: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2647: if(j % ncovmodel == 0){
2648: printf("\n");
2649: fprintf(ficlog,"\n");
2650: }
2651: }
1.224 brouard 2652: #endif
2653: #ifdef LINMINORIGINAL
1.234 brouard 2654: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2655: #else
1.234 brouard 2656: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2657: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2658: #endif
1.234 brouard 2659:
1.191 brouard 2660: #ifdef DEBUGLINMIN
1.234 brouard 2661: for (j=1;j<=n;j++) {
2662: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2663: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2664: if(j % ncovmodel == 0){
2665: printf("\n");
2666: fprintf(ficlog,"\n");
2667: }
2668: }
1.224 brouard 2669: #endif
1.234 brouard 2670: for (j=1;j<=n;j++) {
2671: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2672: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2673: }
1.224 brouard 2674: #ifdef LINMINORIGINAL
2675: #else
1.234 brouard 2676: for (j=1, flatd=0;j<=n;j++) {
2677: if(flatdir[j]>0)
2678: flatd++;
2679: }
2680: if(flatd >0){
1.255 brouard 2681: printf("%d flat directions: ",flatd);
2682: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2683: for (j=1;j<=n;j++) {
2684: if(flatdir[j]>0){
2685: printf("%d ",j);
2686: fprintf(ficlog,"%d ",j);
2687: }
2688: }
2689: printf("\n");
2690: fprintf(ficlog,"\n");
1.319 brouard 2691: #ifdef FLATSUP
2692: free_vector(xit,1,n);
2693: free_vector(xits,1,n);
2694: free_vector(ptt,1,n);
2695: free_vector(pt,1,n);
2696: return;
2697: #endif
1.234 brouard 2698: }
1.191 brouard 2699: #endif
1.234 brouard 2700: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2701: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2702:
1.126 brouard 2703: #ifdef DEBUG
1.234 brouard 2704: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2705: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2706: for(j=1;j<=n;j++){
2707: printf(" %lf",xit[j]);
2708: fprintf(ficlog," %lf",xit[j]);
2709: }
2710: printf("\n");
2711: fprintf(ficlog,"\n");
1.126 brouard 2712: #endif
1.192 brouard 2713: } /* end of t or directest negative */
1.224 brouard 2714: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2715: #else
1.234 brouard 2716: } /* end if (fptt < fp) */
1.192 brouard 2717: #endif
1.225 brouard 2718: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2719: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2720: #else
1.224 brouard 2721: #endif
1.234 brouard 2722: } /* loop iteration */
1.126 brouard 2723: }
1.234 brouard 2724:
1.126 brouard 2725: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2726:
1.235 brouard 2727: 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 2728: {
1.279 brouard 2729: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2730: * (and selected quantitative values in nres)
2731: * by left multiplying the unit
2732: * matrix by transitions matrix until convergence is reached with precision ftolpl
2733: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2734: * Wx is row vector: population in state 1, population in state 2, population dead
2735: * or prevalence in state 1, prevalence in state 2, 0
2736: * newm is the matrix after multiplications, its rows are identical at a factor.
2737: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2738: * Output is prlim.
2739: * Initial matrix pimij
2740: */
1.206 brouard 2741: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2742: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2743: /* 0, 0 , 1} */
2744: /*
2745: * and after some iteration: */
2746: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2747: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2748: /* 0, 0 , 1} */
2749: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2750: /* {0.51571254859325999, 0.4842874514067399, */
2751: /* 0.51326036147820708, 0.48673963852179264} */
2752: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2753:
1.126 brouard 2754: int i, ii,j,k;
1.209 brouard 2755: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2756: /* double **matprod2(); */ /* test */
1.218 brouard 2757: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2758: double **newm;
1.209 brouard 2759: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2760: int ncvloop=0;
1.288 brouard 2761: int first=0;
1.169 brouard 2762:
1.209 brouard 2763: min=vector(1,nlstate);
2764: max=vector(1,nlstate);
2765: meandiff=vector(1,nlstate);
2766:
1.218 brouard 2767: /* Starting with matrix unity */
1.126 brouard 2768: for (ii=1;ii<=nlstate+ndeath;ii++)
2769: for (j=1;j<=nlstate+ndeath;j++){
2770: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2771: }
1.169 brouard 2772:
2773: cov[1]=1.;
2774:
2775: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2776: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2777: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2778: ncvloop++;
1.126 brouard 2779: newm=savm;
2780: /* Covariates have to be included here again */
1.138 brouard 2781: cov[2]=agefin;
1.319 brouard 2782: if(nagesqr==1){
2783: cov[3]= agefin*agefin;
2784: }
1.234 brouard 2785: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2786: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2787: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.319 brouard 2788: /* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; */
1.235 brouard 2789: /* 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)); */
1.234 brouard 2790: }
2791: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2792: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.319 brouard 2793: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2794: /* cov[++k1]=Tqresult[nres][k]; */
1.235 brouard 2795: /* 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]); */
1.138 brouard 2796: }
1.237 brouard 2797: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 2798: if(Dummy[Tage[k]]==2){ /* dummy with age */
1.234 brouard 2799: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 2800: /* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
2801: } else if(Dummy[Tage[k]]==3){ /* quantitative with age */
2802: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2803: /* cov[++k1]=Tqresult[nres][k]; */
1.234 brouard 2804: }
1.235 brouard 2805: /* 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]); */
1.234 brouard 2806: }
1.237 brouard 2807: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2808: /* 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]); */
1.237 brouard 2809: if(Dummy[Tvard[k][1]==0]){
2810: if(Dummy[Tvard[k][2]==0]){
2811: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
1.319 brouard 2812: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.237 brouard 2813: }else{
2814: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
1.319 brouard 2815: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; */
1.237 brouard 2816: }
2817: }else{
2818: if(Dummy[Tvard[k][2]==0]){
2819: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
1.319 brouard 2820: /* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; */
1.237 brouard 2821: }else{
2822: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
1.319 brouard 2823: /* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
1.237 brouard 2824: }
2825: }
1.234 brouard 2826: }
1.138 brouard 2827: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2828: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2829: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2830: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2831: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 2832: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2833: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2834:
1.126 brouard 2835: savm=oldm;
2836: oldm=newm;
1.209 brouard 2837:
2838: for(j=1; j<=nlstate; j++){
2839: max[j]=0.;
2840: min[j]=1.;
2841: }
2842: for(i=1;i<=nlstate;i++){
2843: sumnew=0;
2844: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2845: for(j=1; j<=nlstate; j++){
2846: prlim[i][j]= newm[i][j]/(1-sumnew);
2847: max[j]=FMAX(max[j],prlim[i][j]);
2848: min[j]=FMIN(min[j],prlim[i][j]);
2849: }
2850: }
2851:
1.126 brouard 2852: maxmax=0.;
1.209 brouard 2853: for(j=1; j<=nlstate; j++){
2854: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2855: maxmax=FMAX(maxmax,meandiff[j]);
2856: /* 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 2857: } /* j loop */
1.203 brouard 2858: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2859: /* 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 2860: if(maxmax < ftolpl){
1.209 brouard 2861: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2862: free_vector(min,1,nlstate);
2863: free_vector(max,1,nlstate);
2864: free_vector(meandiff,1,nlstate);
1.126 brouard 2865: return prlim;
2866: }
1.288 brouard 2867: } /* agefin loop */
1.208 brouard 2868: /* After some age loop it doesn't converge */
1.288 brouard 2869: if(!first){
2870: first=1;
2871: 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 2872: 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);
2873: }else if (first >=1 && first <10){
2874: 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);
2875: first++;
2876: }else if (first ==10){
2877: 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);
2878: 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");
2879: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
2880: first++;
1.288 brouard 2881: }
2882:
1.209 brouard 2883: /* 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); */
2884: free_vector(min,1,nlstate);
2885: free_vector(max,1,nlstate);
2886: free_vector(meandiff,1,nlstate);
1.208 brouard 2887:
1.169 brouard 2888: return prlim; /* should not reach here */
1.126 brouard 2889: }
2890:
1.217 brouard 2891:
2892: /**** Back Prevalence limit (stable or period prevalence) ****************/
2893:
1.218 brouard 2894: /* 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) */
2895: /* 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 2896: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2897: {
1.264 brouard 2898: /* 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 2899: matrix by transitions matrix until convergence is reached with precision ftolpl */
2900: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2901: /* Wx is row vector: population in state 1, population in state 2, population dead */
2902: /* or prevalence in state 1, prevalence in state 2, 0 */
2903: /* newm is the matrix after multiplications, its rows are identical at a factor */
2904: /* Initial matrix pimij */
2905: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2906: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2907: /* 0, 0 , 1} */
2908: /*
2909: * and after some iteration: */
2910: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2911: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2912: /* 0, 0 , 1} */
2913: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2914: /* {0.51571254859325999, 0.4842874514067399, */
2915: /* 0.51326036147820708, 0.48673963852179264} */
2916: /* If we start from prlim again, prlim tends to a constant matrix */
2917:
2918: int i, ii,j,k;
1.247 brouard 2919: int first=0;
1.217 brouard 2920: double *min, *max, *meandiff, maxmax,sumnew=0.;
2921: /* double **matprod2(); */ /* test */
2922: double **out, cov[NCOVMAX+1], **bmij();
2923: double **newm;
1.218 brouard 2924: double **dnewm, **doldm, **dsavm; /* for use */
2925: double **oldm, **savm; /* for use */
2926:
1.217 brouard 2927: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2928: int ncvloop=0;
2929:
2930: min=vector(1,nlstate);
2931: max=vector(1,nlstate);
2932: meandiff=vector(1,nlstate);
2933:
1.266 brouard 2934: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2935: oldm=oldms; savm=savms;
2936:
2937: /* Starting with matrix unity */
2938: for (ii=1;ii<=nlstate+ndeath;ii++)
2939: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2940: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2941: }
2942:
2943: cov[1]=1.;
2944:
2945: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2946: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2947: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2948: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2949: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2950: ncvloop++;
1.218 brouard 2951: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2952: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2953: /* Covariates have to be included here again */
2954: cov[2]=agefin;
1.319 brouard 2955: if(nagesqr==1){
1.217 brouard 2956: cov[3]= agefin*agefin;;
1.319 brouard 2957: }
1.242 brouard 2958: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2959: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2960: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2961: /* 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)); */
1.242 brouard 2962: }
2963: /* for (k=1; k<=cptcovn;k++) { */
2964: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2965: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2966: /* /\* 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])]); *\/ */
2967: /* } */
2968: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2969: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2970: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2971: /* 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]); */
2972: }
2973: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2974: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2975: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2976: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2977: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 2978: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age *\/ ERROR ???*/
2979: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.242 brouard 2980: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 2981: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
2982: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.242 brouard 2983: }
2984: /* 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]); */
2985: }
2986: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2987: /* 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]); */
2988: if(Dummy[Tvard[k][1]==0]){
2989: if(Dummy[Tvard[k][2]==0]){
2990: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2991: }else{
2992: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2993: }
2994: }else{
2995: if(Dummy[Tvard[k][2]==0]){
2996: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2997: }else{
2998: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2999: }
3000: }
1.217 brouard 3001: }
3002:
3003: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3004: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3005: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3006: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3007: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3008: /* ij should be linked to the correct index of cov */
3009: /* age and covariate values ij are in 'cov', but we need to pass
3010: * ij for the observed prevalence at age and status and covariate
3011: * number: prevacurrent[(int)agefin][ii][ij]
3012: */
3013: /* 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 *\/ */
3014: /* 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 *\/ */
3015: 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 3016: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3017: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3018: /* for(i=1; i<=nlstate+ndeath; i++) { */
3019: /* printf("%d newm= ",i); */
3020: /* for(j=1;j<=nlstate+ndeath;j++) { */
3021: /* printf("%f ",newm[i][j]); */
3022: /* } */
3023: /* printf("oldm * "); */
3024: /* for(j=1;j<=nlstate+ndeath;j++) { */
3025: /* printf("%f ",oldm[i][j]); */
3026: /* } */
1.268 brouard 3027: /* printf(" bmmij "); */
1.266 brouard 3028: /* for(j=1;j<=nlstate+ndeath;j++) { */
3029: /* printf("%f ",pmmij[i][j]); */
3030: /* } */
3031: /* printf("\n"); */
3032: /* } */
3033: /* } */
1.217 brouard 3034: savm=oldm;
3035: oldm=newm;
1.266 brouard 3036:
1.217 brouard 3037: for(j=1; j<=nlstate; j++){
3038: max[j]=0.;
3039: min[j]=1.;
3040: }
3041: for(j=1; j<=nlstate; j++){
3042: for(i=1;i<=nlstate;i++){
1.234 brouard 3043: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3044: bprlim[i][j]= newm[i][j];
3045: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3046: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3047: }
3048: }
1.218 brouard 3049:
1.217 brouard 3050: maxmax=0.;
3051: for(i=1; i<=nlstate; i++){
1.318 brouard 3052: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3053: maxmax=FMAX(maxmax,meandiff[i]);
3054: /* 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 3055: } /* i loop */
1.217 brouard 3056: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3057: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3058: if(maxmax < ftolpl){
1.220 brouard 3059: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3060: free_vector(min,1,nlstate);
3061: free_vector(max,1,nlstate);
3062: free_vector(meandiff,1,nlstate);
3063: return bprlim;
3064: }
1.288 brouard 3065: } /* agefin loop */
1.217 brouard 3066: /* After some age loop it doesn't converge */
1.288 brouard 3067: if(!first){
1.247 brouard 3068: first=1;
3069: 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\
3070: 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);
3071: }
3072: 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 3073: 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);
3074: /* 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); */
3075: free_vector(min,1,nlstate);
3076: free_vector(max,1,nlstate);
3077: free_vector(meandiff,1,nlstate);
3078:
3079: return bprlim; /* should not reach here */
3080: }
3081:
1.126 brouard 3082: /*************** transition probabilities ***************/
3083:
3084: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3085: {
1.138 brouard 3086: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3087: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3088: model to the ncovmodel covariates (including constant and age).
3089: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3090: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3091: ncth covariate in the global vector x is given by the formula:
3092: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3093: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3094: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3095: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3096: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3097: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3098: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3099: */
3100: double s1, lnpijopii;
1.126 brouard 3101: /*double t34;*/
1.164 brouard 3102: int i,j, nc, ii, jj;
1.126 brouard 3103:
1.223 brouard 3104: for(i=1; i<= nlstate; i++){
3105: for(j=1; j<i;j++){
3106: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3107: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3108: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3109: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3110: }
3111: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3112: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3113: }
3114: for(j=i+1; j<=nlstate+ndeath;j++){
3115: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3116: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3117: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3118: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3119: }
3120: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3121: }
3122: }
1.218 brouard 3123:
1.223 brouard 3124: for(i=1; i<= nlstate; i++){
3125: s1=0;
3126: for(j=1; j<i; j++){
3127: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3128: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3129: }
3130: for(j=i+1; j<=nlstate+ndeath; j++){
3131: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3132: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3133: }
3134: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3135: ps[i][i]=1./(s1+1.);
3136: /* Computing other pijs */
3137: for(j=1; j<i; j++)
3138: ps[i][j]= exp(ps[i][j])*ps[i][i];
3139: for(j=i+1; j<=nlstate+ndeath; j++)
3140: ps[i][j]= exp(ps[i][j])*ps[i][i];
3141: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3142: } /* end i */
1.218 brouard 3143:
1.223 brouard 3144: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3145: for(jj=1; jj<= nlstate+ndeath; jj++){
3146: ps[ii][jj]=0;
3147: ps[ii][ii]=1;
3148: }
3149: }
1.294 brouard 3150:
3151:
1.223 brouard 3152: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3153: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3154: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3155: /* } */
3156: /* printf("\n "); */
3157: /* } */
3158: /* printf("\n ");printf("%lf ",cov[2]);*/
3159: /*
3160: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3161: goto end;*/
1.266 brouard 3162: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3163: }
3164:
1.218 brouard 3165: /*************** backward transition probabilities ***************/
3166:
3167: /* 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 ) */
3168: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3169: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3170: {
1.302 brouard 3171: /* 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 3172: * 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 3173: */
1.218 brouard 3174: int i, ii, j,k;
1.222 brouard 3175:
3176: double **out, **pmij();
3177: double sumnew=0.;
1.218 brouard 3178: double agefin;
1.292 brouard 3179: 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 3180: double **dnewm, **dsavm, **doldm;
3181: double **bbmij;
3182:
1.218 brouard 3183: doldm=ddoldms; /* global pointers */
1.222 brouard 3184: dnewm=ddnewms;
3185: dsavm=ddsavms;
1.318 brouard 3186:
3187: /* Debug */
3188: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3189: agefin=cov[2];
1.268 brouard 3190: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3191: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3192: the observed prevalence (with this covariate ij) at beginning of transition */
3193: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3194:
3195: /* P_x */
1.266 brouard 3196: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3197: /* outputs pmmij which is a stochastic matrix in row */
3198:
3199: /* Diag(w_x) */
1.292 brouard 3200: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3201: sumnew=0.;
1.269 brouard 3202: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3203: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3204: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3205: sumnew+=prevacurrent[(int)agefin][ii][ij];
3206: }
3207: if(sumnew >0.01){ /* At least some value in the prevalence */
3208: for (ii=1;ii<=nlstate+ndeath;ii++){
3209: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3210: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3211: }
3212: }else{
3213: for (ii=1;ii<=nlstate+ndeath;ii++){
3214: for (j=1;j<=nlstate+ndeath;j++)
3215: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3216: }
3217: /* if(sumnew <0.9){ */
3218: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3219: /* } */
3220: }
3221: k3=0.0; /* We put the last diagonal to 0 */
3222: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3223: doldm[ii][ii]= k3;
3224: }
3225: /* End doldm, At the end doldm is diag[(w_i)] */
3226:
1.292 brouard 3227: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3228: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3229:
1.292 brouard 3230: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3231: /* 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 3232: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3233: sumnew=0.;
1.222 brouard 3234: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3235: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3236: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3237: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3238: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3239: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3240: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3241: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3242: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3243: /* }else */
1.268 brouard 3244: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3245: } /*End ii */
3246: } /* 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 */
3247:
1.292 brouard 3248: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3249: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3250: /* end bmij */
1.266 brouard 3251: return ps; /*pointer is unchanged */
1.218 brouard 3252: }
1.217 brouard 3253: /*************** transition probabilities ***************/
3254:
1.218 brouard 3255: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3256: {
3257: /* According to parameters values stored in x and the covariate's values stored in cov,
3258: computes the probability to be observed in state j being in state i by appying the
3259: model to the ncovmodel covariates (including constant and age).
3260: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3261: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3262: ncth covariate in the global vector x is given by the formula:
3263: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3264: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3265: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3266: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3267: Outputs ps[i][j] the probability to be observed in j being in j according to
3268: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3269: */
3270: double s1, lnpijopii;
3271: /*double t34;*/
3272: int i,j, nc, ii, jj;
3273:
1.234 brouard 3274: for(i=1; i<= nlstate; i++){
3275: for(j=1; j<i;j++){
3276: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3277: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3278: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3279: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3280: }
3281: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3282: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3283: }
3284: for(j=i+1; j<=nlstate+ndeath;j++){
3285: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3286: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3287: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3288: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3289: }
3290: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3291: }
3292: }
3293:
3294: for(i=1; i<= nlstate; i++){
3295: s1=0;
3296: for(j=1; j<i; j++){
3297: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3298: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3299: }
3300: for(j=i+1; j<=nlstate+ndeath; j++){
3301: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3302: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3303: }
3304: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3305: ps[i][i]=1./(s1+1.);
3306: /* Computing other pijs */
3307: for(j=1; j<i; j++)
3308: ps[i][j]= exp(ps[i][j])*ps[i][i];
3309: for(j=i+1; j<=nlstate+ndeath; j++)
3310: ps[i][j]= exp(ps[i][j])*ps[i][i];
3311: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3312: } /* end i */
3313:
3314: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3315: for(jj=1; jj<= nlstate+ndeath; jj++){
3316: ps[ii][jj]=0;
3317: ps[ii][ii]=1;
3318: }
3319: }
1.296 brouard 3320: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3321: for(jj=1; jj<= nlstate+ndeath; jj++){
3322: s1=0.;
3323: for(ii=1; ii<= nlstate+ndeath; ii++){
3324: s1+=ps[ii][jj];
3325: }
3326: for(ii=1; ii<= nlstate; ii++){
3327: ps[ii][jj]=ps[ii][jj]/s1;
3328: }
3329: }
3330: /* Transposition */
3331: for(jj=1; jj<= nlstate+ndeath; jj++){
3332: for(ii=jj; ii<= nlstate+ndeath; ii++){
3333: s1=ps[ii][jj];
3334: ps[ii][jj]=ps[jj][ii];
3335: ps[jj][ii]=s1;
3336: }
3337: }
3338: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3339: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3340: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3341: /* } */
3342: /* printf("\n "); */
3343: /* } */
3344: /* printf("\n ");printf("%lf ",cov[2]);*/
3345: /*
3346: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3347: goto end;*/
3348: return ps;
1.217 brouard 3349: }
3350:
3351:
1.126 brouard 3352: /**************** Product of 2 matrices ******************/
3353:
1.145 brouard 3354: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3355: {
3356: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3357: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3358: /* in, b, out are matrice of pointers which should have been initialized
3359: before: only the contents of out is modified. The function returns
3360: a pointer to pointers identical to out */
1.145 brouard 3361: int i, j, k;
1.126 brouard 3362: for(i=nrl; i<= nrh; i++)
1.145 brouard 3363: for(k=ncolol; k<=ncoloh; k++){
3364: out[i][k]=0.;
3365: for(j=ncl; j<=nch; j++)
3366: out[i][k] +=in[i][j]*b[j][k];
3367: }
1.126 brouard 3368: return out;
3369: }
3370:
3371:
3372: /************* Higher Matrix Product ***************/
3373:
1.235 brouard 3374: 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 3375: {
1.218 brouard 3376: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3377: 'nhstepm*hstepm*stepm' months (i.e. until
3378: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3379: nhstepm*hstepm matrices.
3380: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3381: (typically every 2 years instead of every month which is too big
3382: for the memory).
3383: Model is determined by parameters x and covariates have to be
3384: included manually here.
3385:
3386: */
3387:
3388: int i, j, d, h, k;
1.131 brouard 3389: double **out, cov[NCOVMAX+1];
1.126 brouard 3390: double **newm;
1.187 brouard 3391: double agexact;
1.214 brouard 3392: double agebegin, ageend;
1.126 brouard 3393:
3394: /* Hstepm could be zero and should return the unit matrix */
3395: for (i=1;i<=nlstate+ndeath;i++)
3396: for (j=1;j<=nlstate+ndeath;j++){
3397: oldm[i][j]=(i==j ? 1.0 : 0.0);
3398: po[i][j][0]=(i==j ? 1.0 : 0.0);
3399: }
3400: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3401: for(h=1; h <=nhstepm; h++){
3402: for(d=1; d <=hstepm; d++){
3403: newm=savm;
3404: /* Covariates have to be included here again */
3405: cov[1]=1.;
1.214 brouard 3406: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3407: cov[2]=agexact;
1.319 brouard 3408: if(nagesqr==1){
1.227 brouard 3409: cov[3]= agexact*agexact;
1.319 brouard 3410: }
1.235 brouard 3411: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
1.319 brouard 3412: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3413: /* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 */
3414: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3415: /* k 1 2 3 4 5 6 7 8 9 */
3416: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
3417: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
3418: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
3419: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1.235 brouard 3420: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3421: /* printf("hpxij 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)); */
3422: }
3423: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3424: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.319 brouard 3425: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
1.235 brouard 3426: /* 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]); */
3427: }
1.319 brouard 3428: for (k=1; k<=cptcovage;k++){ /* For product with age V1+V1*age +V4 +age*V3 */
3429: /* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*/
3430: /* */
3431: if(Dummy[Tage[k]]== 2){ /* dummy with age */
3432: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age *\/ */
1.235 brouard 3433: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 3434: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
3435: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.235 brouard 3436: }
3437: /* 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]); */
3438: }
1.319 brouard 3439: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 3440: /* printf("hPxij 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]); */
1.319 brouard 3441: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
3442: if(Dummy[Tvard[k][1]==0]){
3443: if(Dummy[Tvard[k][2]==0]){
3444: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3445: }else{
3446: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
3447: }
3448: }else{
3449: if(Dummy[Tvard[k][2]==0]){
3450: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
3451: }else{
3452: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3453: }
3454: }
1.235 brouard 3455: }
3456: /* for (k=1; k<=cptcovn;k++) */
3457: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3458: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3459: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3460: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3461: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3462:
3463:
1.126 brouard 3464: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3465: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3466: /* right multiplication of oldm by the current matrix */
1.126 brouard 3467: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3468: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3469: /* if((int)age == 70){ */
3470: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3471: /* for(i=1; i<=nlstate+ndeath; i++) { */
3472: /* printf("%d pmmij ",i); */
3473: /* for(j=1;j<=nlstate+ndeath;j++) { */
3474: /* printf("%f ",pmmij[i][j]); */
3475: /* } */
3476: /* printf(" oldm "); */
3477: /* for(j=1;j<=nlstate+ndeath;j++) { */
3478: /* printf("%f ",oldm[i][j]); */
3479: /* } */
3480: /* printf("\n"); */
3481: /* } */
3482: /* } */
1.126 brouard 3483: savm=oldm;
3484: oldm=newm;
3485: }
3486: for(i=1; i<=nlstate+ndeath; i++)
3487: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3488: po[i][j][h]=newm[i][j];
3489: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3490: }
1.128 brouard 3491: /*printf("h=%d ",h);*/
1.126 brouard 3492: } /* end h */
1.267 brouard 3493: /* printf("\n H=%d \n",h); */
1.126 brouard 3494: return po;
3495: }
3496:
1.217 brouard 3497: /************* Higher Back Matrix Product ***************/
1.218 brouard 3498: /* 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 3499: 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 3500: {
1.266 brouard 3501: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3502: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3503: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3504: nhstepm*hstepm matrices.
3505: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3506: (typically every 2 years instead of every month which is too big
1.217 brouard 3507: for the memory).
1.218 brouard 3508: Model is determined by parameters x and covariates have to be
1.266 brouard 3509: included manually here. Then we use a call to bmij(x and cov)
3510: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3511: */
1.217 brouard 3512:
3513: int i, j, d, h, k;
1.266 brouard 3514: double **out, cov[NCOVMAX+1], **bmij();
3515: double **newm, ***newmm;
1.217 brouard 3516: double agexact;
3517: double agebegin, ageend;
1.222 brouard 3518: double **oldm, **savm;
1.217 brouard 3519:
1.266 brouard 3520: newmm=po; /* To be saved */
3521: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3522: /* Hstepm could be zero and should return the unit matrix */
3523: for (i=1;i<=nlstate+ndeath;i++)
3524: for (j=1;j<=nlstate+ndeath;j++){
3525: oldm[i][j]=(i==j ? 1.0 : 0.0);
3526: po[i][j][0]=(i==j ? 1.0 : 0.0);
3527: }
3528: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3529: for(h=1; h <=nhstepm; h++){
3530: for(d=1; d <=hstepm; d++){
3531: newm=savm;
3532: /* Covariates have to be included here again */
3533: cov[1]=1.;
1.271 brouard 3534: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3535: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3536: /* Debug */
3537: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3538: cov[2]=agexact;
3539: if(nagesqr==1)
1.222 brouard 3540: cov[3]= agexact*agexact;
1.266 brouard 3541: for (k=1; k<=cptcovn;k++){
3542: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3543: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3544: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3545: /* 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)); */
3546: }
1.267 brouard 3547: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3548: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3549: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3550: /* 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]); */
3551: }
1.319 brouard 3552: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 *//* For product with age */
3553: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age error!!!*\/ */
3554: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.267 brouard 3555: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 3556: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
1.267 brouard 3557: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3558: }
3559: /* 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]); */
3560: }
3561: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3562: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3563: }
1.217 brouard 3564: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3565: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3566:
1.218 brouard 3567: /* Careful transposed matrix */
1.266 brouard 3568: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3569: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3570: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3571: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3572: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3573: /* if((int)age == 70){ */
3574: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3575: /* for(i=1; i<=nlstate+ndeath; i++) { */
3576: /* printf("%d pmmij ",i); */
3577: /* for(j=1;j<=nlstate+ndeath;j++) { */
3578: /* printf("%f ",pmmij[i][j]); */
3579: /* } */
3580: /* printf(" oldm "); */
3581: /* for(j=1;j<=nlstate+ndeath;j++) { */
3582: /* printf("%f ",oldm[i][j]); */
3583: /* } */
3584: /* printf("\n"); */
3585: /* } */
3586: /* } */
3587: savm=oldm;
3588: oldm=newm;
3589: }
3590: for(i=1; i<=nlstate+ndeath; i++)
3591: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3592: po[i][j][h]=newm[i][j];
1.268 brouard 3593: /* if(h==nhstepm) */
3594: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3595: }
1.268 brouard 3596: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3597: } /* end h */
1.268 brouard 3598: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3599: return po;
3600: }
3601:
3602:
1.162 brouard 3603: #ifdef NLOPT
3604: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3605: double fret;
3606: double *xt;
3607: int j;
3608: myfunc_data *d2 = (myfunc_data *) pd;
3609: /* xt = (p1-1); */
3610: xt=vector(1,n);
3611: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3612:
3613: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3614: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3615: printf("Function = %.12lf ",fret);
3616: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3617: printf("\n");
3618: free_vector(xt,1,n);
3619: return fret;
3620: }
3621: #endif
1.126 brouard 3622:
3623: /*************** log-likelihood *************/
3624: double func( double *x)
3625: {
1.226 brouard 3626: int i, ii, j, k, mi, d, kk;
3627: int ioffset=0;
3628: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3629: double **out;
3630: double lli; /* Individual log likelihood */
3631: int s1, s2;
1.228 brouard 3632: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
1.226 brouard 3633: double bbh, survp;
3634: long ipmx;
3635: double agexact;
3636: /*extern weight */
3637: /* We are differentiating ll according to initial status */
3638: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3639: /*for(i=1;i<imx;i++)
3640: printf(" %d\n",s[4][i]);
3641: */
1.162 brouard 3642:
1.226 brouard 3643: ++countcallfunc;
1.162 brouard 3644:
1.226 brouard 3645: cov[1]=1.;
1.126 brouard 3646:
1.226 brouard 3647: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3648: ioffset=0;
1.226 brouard 3649: if(mle==1){
3650: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3651: /* Computes the values of the ncovmodel covariates of the model
3652: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3653: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3654: to be observed in j being in i according to the model.
3655: */
1.243 brouard 3656: ioffset=2+nagesqr ;
1.233 brouard 3657: /* Fixed */
1.319 brouard 3658: for (k=1; k<=ncovf;k++){ /* For each fixed covariate dummu or quant or prod */
3659: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3660: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3661: /* 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 3662: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.319 brouard 3663: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (TvarFind[1]=6)*/
3664: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3665: }
1.226 brouard 3666: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3667: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3668: has been calculated etc */
3669: /* For an individual i, wav[i] gives the number of effective waves */
3670: /* We compute the contribution to Likelihood of each effective transition
3671: mw[mi][i] is real wave of the mi th effectve wave */
3672: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3673: s2=s[mw[mi+1][i]][i];
3674: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3675: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3676: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3677: */
3678: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3679: 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*/
3680: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? */
1.242 brouard 3681: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3682: }
3683: for (ii=1;ii<=nlstate+ndeath;ii++)
3684: for (j=1;j<=nlstate+ndeath;j++){
3685: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3686: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3687: }
3688: for(d=0; d<dh[mi][i]; d++){
3689: newm=savm;
3690: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3691: cov[2]=agexact;
3692: if(nagesqr==1)
3693: cov[3]= agexact*agexact; /* Should be changed here */
3694: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3695: if(!FixedV[Tvar[Tage[kk]]])
3696: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3697: else
3698: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3699: }
3700: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3701: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3702: savm=oldm;
3703: oldm=newm;
3704: } /* end mult */
3705:
3706: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3707: /* But now since version 0.9 we anticipate for bias at large stepm.
3708: * If stepm is larger than one month (smallest stepm) and if the exact delay
3709: * (in months) between two waves is not a multiple of stepm, we rounded to
3710: * the nearest (and in case of equal distance, to the lowest) interval but now
3711: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3712: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3713: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3714: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3715: * -stepm/2 to stepm/2 .
3716: * For stepm=1 the results are the same as for previous versions of Imach.
3717: * For stepm > 1 the results are less biased than in previous versions.
3718: */
1.234 brouard 3719: s1=s[mw[mi][i]][i];
3720: s2=s[mw[mi+1][i]][i];
3721: bbh=(double)bh[mi][i]/(double)stepm;
3722: /* bias bh is positive if real duration
3723: * is higher than the multiple of stepm and negative otherwise.
3724: */
3725: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3726: if( s2 > nlstate){
3727: /* i.e. if s2 is a death state and if the date of death is known
3728: then the contribution to the likelihood is the probability to
3729: die between last step unit time and current step unit time,
3730: which is also equal to probability to die before dh
3731: minus probability to die before dh-stepm .
3732: In version up to 0.92 likelihood was computed
3733: as if date of death was unknown. Death was treated as any other
3734: health state: the date of the interview describes the actual state
3735: and not the date of a change in health state. The former idea was
3736: to consider that at each interview the state was recorded
3737: (healthy, disable or death) and IMaCh was corrected; but when we
3738: introduced the exact date of death then we should have modified
3739: the contribution of an exact death to the likelihood. This new
3740: contribution is smaller and very dependent of the step unit
3741: stepm. It is no more the probability to die between last interview
3742: and month of death but the probability to survive from last
3743: interview up to one month before death multiplied by the
3744: probability to die within a month. Thanks to Chris
3745: Jackson for correcting this bug. Former versions increased
3746: mortality artificially. The bad side is that we add another loop
3747: which slows down the processing. The difference can be up to 10%
3748: lower mortality.
3749: */
3750: /* If, at the beginning of the maximization mostly, the
3751: cumulative probability or probability to be dead is
3752: constant (ie = 1) over time d, the difference is equal to
3753: 0. out[s1][3] = savm[s1][3]: probability, being at state
3754: s1 at precedent wave, to be dead a month before current
3755: wave is equal to probability, being at state s1 at
3756: precedent wave, to be dead at mont of the current
3757: wave. Then the observed probability (that this person died)
3758: is null according to current estimated parameter. In fact,
3759: it should be very low but not zero otherwise the log go to
3760: infinity.
3761: */
1.183 brouard 3762: /* #ifdef INFINITYORIGINAL */
3763: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3764: /* #else */
3765: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3766: /* lli=log(mytinydouble); */
3767: /* else */
3768: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3769: /* #endif */
1.226 brouard 3770: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3771:
1.226 brouard 3772: } else if ( s2==-1 ) { /* alive */
3773: for (j=1,survp=0. ; j<=nlstate; j++)
3774: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3775: /*survp += out[s1][j]; */
3776: lli= log(survp);
3777: }
3778: else if (s2==-4) {
3779: for (j=3,survp=0. ; j<=nlstate; j++)
3780: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3781: lli= log(survp);
3782: }
3783: else if (s2==-5) {
3784: for (j=1,survp=0. ; j<=2; j++)
3785: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3786: lli= log(survp);
3787: }
3788: else{
3789: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3790: /* 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 */
3791: }
3792: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3793: /*if(lli ==000.0)*/
3794: /*printf("bbh= %f lli=%f savm=%f out=%f %d\n",bbh,lli,savm[s1][s2], out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]],i); */
3795: ipmx +=1;
3796: sw += weight[i];
3797: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3798: /* if (lli < log(mytinydouble)){ */
3799: /* 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); */
3800: /* 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]); */
3801: /* } */
3802: } /* end of wave */
3803: } /* end of individual */
3804: } else if(mle==2){
3805: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 3806: ioffset=2+nagesqr ;
3807: for (k=1; k<=ncovf;k++)
3808: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 3809: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3810: for(k=1; k <= ncovv ; k++){
3811: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3812: }
1.226 brouard 3813: for (ii=1;ii<=nlstate+ndeath;ii++)
3814: for (j=1;j<=nlstate+ndeath;j++){
3815: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3816: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3817: }
3818: for(d=0; d<=dh[mi][i]; d++){
3819: newm=savm;
3820: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3821: cov[2]=agexact;
3822: if(nagesqr==1)
3823: cov[3]= agexact*agexact;
3824: for (kk=1; kk<=cptcovage;kk++) {
3825: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3826: }
3827: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3828: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3829: savm=oldm;
3830: oldm=newm;
3831: } /* end mult */
3832:
3833: s1=s[mw[mi][i]][i];
3834: s2=s[mw[mi+1][i]][i];
3835: bbh=(double)bh[mi][i]/(double)stepm;
3836: 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 */
3837: ipmx +=1;
3838: sw += weight[i];
3839: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3840: } /* end of wave */
3841: } /* end of individual */
3842: } else if(mle==3){ /* exponential inter-extrapolation */
3843: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3844: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3845: for(mi=1; mi<= wav[i]-1; mi++){
3846: for (ii=1;ii<=nlstate+ndeath;ii++)
3847: for (j=1;j<=nlstate+ndeath;j++){
3848: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3849: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3850: }
3851: for(d=0; d<dh[mi][i]; d++){
3852: newm=savm;
3853: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3854: cov[2]=agexact;
3855: if(nagesqr==1)
3856: cov[3]= agexact*agexact;
3857: for (kk=1; kk<=cptcovage;kk++) {
3858: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3859: }
3860: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3861: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3862: savm=oldm;
3863: oldm=newm;
3864: } /* end mult */
3865:
3866: s1=s[mw[mi][i]][i];
3867: s2=s[mw[mi+1][i]][i];
3868: bbh=(double)bh[mi][i]/(double)stepm;
3869: 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 */
3870: ipmx +=1;
3871: sw += weight[i];
3872: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3873: } /* end of wave */
3874: } /* end of individual */
3875: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3876: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3877: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3878: for(mi=1; mi<= wav[i]-1; mi++){
3879: for (ii=1;ii<=nlstate+ndeath;ii++)
3880: for (j=1;j<=nlstate+ndeath;j++){
3881: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3882: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3883: }
3884: for(d=0; d<dh[mi][i]; d++){
3885: newm=savm;
3886: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3887: cov[2]=agexact;
3888: if(nagesqr==1)
3889: cov[3]= agexact*agexact;
3890: for (kk=1; kk<=cptcovage;kk++) {
3891: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3892: }
1.126 brouard 3893:
1.226 brouard 3894: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3895: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3896: savm=oldm;
3897: oldm=newm;
3898: } /* end mult */
3899:
3900: s1=s[mw[mi][i]][i];
3901: s2=s[mw[mi+1][i]][i];
3902: if( s2 > nlstate){
3903: lli=log(out[s1][s2] - savm[s1][s2]);
3904: } else if ( s2==-1 ) { /* alive */
3905: for (j=1,survp=0. ; j<=nlstate; j++)
3906: survp += out[s1][j];
3907: lli= log(survp);
3908: }else{
3909: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3910: }
3911: ipmx +=1;
3912: sw += weight[i];
3913: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3914: /* printf("i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],out[s1][s2],savm[s1][s2]); */
1.226 brouard 3915: } /* end of wave */
3916: } /* end of individual */
3917: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3918: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3919: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3920: for(mi=1; mi<= wav[i]-1; mi++){
3921: for (ii=1;ii<=nlstate+ndeath;ii++)
3922: for (j=1;j<=nlstate+ndeath;j++){
3923: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3924: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3925: }
3926: for(d=0; d<dh[mi][i]; d++){
3927: newm=savm;
3928: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3929: cov[2]=agexact;
3930: if(nagesqr==1)
3931: cov[3]= agexact*agexact;
3932: for (kk=1; kk<=cptcovage;kk++) {
3933: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3934: }
1.126 brouard 3935:
1.226 brouard 3936: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3937: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3938: savm=oldm;
3939: oldm=newm;
3940: } /* end mult */
3941:
3942: s1=s[mw[mi][i]][i];
3943: s2=s[mw[mi+1][i]][i];
3944: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3945: ipmx +=1;
3946: sw += weight[i];
3947: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3948: /*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]);*/
3949: } /* end of wave */
3950: } /* end of individual */
3951: } /* End of if */
3952: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3953: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3954: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3955: return -l;
1.126 brouard 3956: }
3957:
3958: /*************** log-likelihood *************/
3959: double funcone( double *x)
3960: {
1.228 brouard 3961: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3962: int i, ii, j, k, mi, d, kk;
1.228 brouard 3963: int ioffset=0;
1.131 brouard 3964: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3965: double **out;
3966: double lli; /* Individual log likelihood */
3967: double llt;
3968: int s1, s2;
1.228 brouard 3969: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3970:
1.126 brouard 3971: double bbh, survp;
1.187 brouard 3972: double agexact;
1.214 brouard 3973: double agebegin, ageend;
1.126 brouard 3974: /*extern weight */
3975: /* We are differentiating ll according to initial status */
3976: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3977: /*for(i=1;i<imx;i++)
3978: printf(" %d\n",s[4][i]);
3979: */
3980: cov[1]=1.;
3981:
3982: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3983: ioffset=0;
3984: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3985: /* ioffset=2+nagesqr+cptcovage; */
3986: ioffset=2+nagesqr;
1.232 brouard 3987: /* Fixed */
1.224 brouard 3988: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3989: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.311 brouard 3990: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products *//* Missing values are set to -1 but should be dropped */
1.232 brouard 3991: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, only V1 is fixed (k=6)*/
3992: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3993: /* cov[2+6]=covar[Tvar[6]][i]; */
3994: /* cov[2+6]=covar[2][i]; V2 */
3995: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3996: /* cov[2+7]=covar[Tvar[7]][i]; */
3997: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3998: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3999: /* cov[2+9]=covar[Tvar[9]][i]; */
4000: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4001: }
1.232 brouard 4002: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4003: /* 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?)*\/ */
4004: /* } */
1.231 brouard 4005: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4006: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4007: /* } */
1.225 brouard 4008:
1.233 brouard 4009:
4010: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 4011: /* Wave varying (but not age varying) */
4012: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 4013: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
4014: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4015: }
1.232 brouard 4016: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 4017: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4018: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4019: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4020: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4021: /* printf(" i=%d,mi=%d,itv=%d,TmodelInvind[itv]=%d,cotvar[mw[mi][i]][TmodelInvind[itv]][i]=%f\n", i, mi, itv, TmodelInvind[itv],cotvar[mw[mi][i]][TmodelInvind[itv]][i]); */
1.232 brouard 4022: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4023: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4024: /* /\* 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]); *\/ */
4025: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4026: /* } */
1.126 brouard 4027: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4028: for (j=1;j<=nlstate+ndeath;j++){
4029: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4030: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4031: }
1.214 brouard 4032:
4033: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4034: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4035: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4036: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4037: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4038: and mw[mi+1][i]. dh depends on stepm.*/
4039: newm=savm;
1.247 brouard 4040: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4041: cov[2]=agexact;
4042: if(nagesqr==1)
4043: cov[3]= agexact*agexact;
4044: for (kk=1; kk<=cptcovage;kk++) {
4045: if(!FixedV[Tvar[Tage[kk]]])
4046: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4047: else
4048: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
4049: }
4050: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4051: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4052: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4053: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4054: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4055: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4056: savm=oldm;
4057: oldm=newm;
1.126 brouard 4058: } /* end mult */
4059:
4060: s1=s[mw[mi][i]][i];
4061: s2=s[mw[mi+1][i]][i];
1.217 brouard 4062: /* if(s2==-1){ */
1.268 brouard 4063: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4064: /* /\* exit(1); *\/ */
4065: /* } */
1.126 brouard 4066: bbh=(double)bh[mi][i]/(double)stepm;
4067: /* bias is positive if real duration
4068: * is higher than the multiple of stepm and negative otherwise.
4069: */
4070: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4071: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4072: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4073: for (j=1,survp=0. ; j<=nlstate; j++)
4074: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4075: lli= log(survp);
1.126 brouard 4076: }else if (mle==1){
1.242 brouard 4077: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4078: } else if(mle==2){
1.242 brouard 4079: 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 4080: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4081: 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 4082: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4083: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4084: } else{ /* mle=0 back to 1 */
1.242 brouard 4085: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4086: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4087: } /* End of if */
4088: ipmx +=1;
4089: sw += weight[i];
4090: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 4091: /*printf("i=%6d s1=%1d s2=%1d mi=%1d mw=%1d dh=%3d prob=%10.6f w=%6.4f out=%10.6f sav=%10.6f\n",i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],out[s1][s2],savm[s1][s2]); */
1.126 brouard 4092: if(globpr){
1.246 brouard 4093: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4094: %11.6f %11.6f %11.6f ", \
1.242 brouard 4095: 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 4096: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 4097: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4098: llt +=ll[k]*gipmx/gsw;
4099: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
4100: }
4101: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 4102: }
1.232 brouard 4103: } /* end of wave */
4104: } /* end of individual */
4105: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4106: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4107: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4108: if(globpr==0){ /* First time we count the contributions and weights */
4109: gipmx=ipmx;
4110: gsw=sw;
4111: }
4112: return -l;
1.126 brouard 4113: }
4114:
4115:
4116: /*************** function likelione ***********/
1.292 brouard 4117: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4118: {
4119: /* This routine should help understanding what is done with
4120: the selection of individuals/waves and
4121: to check the exact contribution to the likelihood.
4122: Plotting could be done.
4123: */
4124: int k;
4125:
4126: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4127: strcpy(fileresilk,"ILK_");
1.202 brouard 4128: strcat(fileresilk,fileresu);
1.126 brouard 4129: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4130: printf("Problem with resultfile: %s\n", fileresilk);
4131: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4132: }
1.214 brouard 4133: 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");
4134: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4135: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4136: for(k=1; k<=nlstate; k++)
4137: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4138: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4139: }
4140:
1.292 brouard 4141: *fretone=(*func)(p);
1.126 brouard 4142: if(*globpri !=0){
4143: fclose(ficresilk);
1.205 brouard 4144: if (mle ==0)
4145: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4146: else if(mle >=1)
4147: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4148: 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 4149: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4150:
4151: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4152: fprintf(fichtm,"<br>- Probability p<sub>%dj</sub> by origin %d and destination j. Dot's sizes are related to corresponding weight: <a href=\"%s-p%dj.png\">%s-p%dj.png</a><br> \
1.208 brouard 4153: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4154: }
1.207 brouard 4155: fprintf(fichtm,"<br>- The function drawn is -2Log(L) in Log scale: by state of origin <a href=\"%s-ori.png\">%s-ori.png</a><br> \
1.204 brouard 4156: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4157: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4158: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4159: fflush(fichtm);
1.205 brouard 4160: }
1.126 brouard 4161: return;
4162: }
4163:
4164:
4165: /*********** Maximum Likelihood Estimation ***************/
4166:
4167: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4168: {
1.319 brouard 4169: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4170: double **xi;
4171: double fret;
4172: double fretone; /* Only one call to likelihood */
4173: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4174:
4175: #ifdef NLOPT
4176: int creturn;
4177: nlopt_opt opt;
4178: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4179: double *lb;
4180: double minf; /* the minimum objective value, upon return */
4181: double * p1; /* Shifted parameters from 0 instead of 1 */
4182: myfunc_data dinst, *d = &dinst;
4183: #endif
4184:
4185:
1.126 brouard 4186: xi=matrix(1,npar,1,npar);
4187: for (i=1;i<=npar;i++)
4188: for (j=1;j<=npar;j++)
4189: xi[i][j]=(i==j ? 1.0 : 0.0);
4190: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4191: strcpy(filerespow,"POW_");
1.126 brouard 4192: strcat(filerespow,fileres);
4193: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4194: printf("Problem with resultfile: %s\n", filerespow);
4195: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4196: }
4197: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4198: for (i=1;i<=nlstate;i++)
4199: for(j=1;j<=nlstate+ndeath;j++)
4200: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4201: fprintf(ficrespow,"\n");
1.162 brouard 4202: #ifdef POWELL
1.319 brouard 4203: #ifdef LINMINORIGINAL
4204: #else /* LINMINORIGINAL */
4205:
4206: flatdir=ivector(1,npar);
4207: for (j=1;j<=npar;j++) flatdir[j]=0;
4208: #endif /*LINMINORIGINAL */
4209:
4210: #ifdef FLATSUP
4211: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4212: /* reorganizing p by suppressing flat directions */
4213: for(i=1, jk=1; i <=nlstate; i++){
4214: for(k=1; k <=(nlstate+ndeath); k++){
4215: if (k != i) {
4216: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4217: if(flatdir[jk]==1){
4218: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4219: }
4220: for(j=1; j <=ncovmodel; j++){
4221: printf("%12.7f ",p[jk]);
4222: jk++;
4223: }
4224: printf("\n");
4225: }
4226: }
4227: }
4228: /* skipping */
4229: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4230: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4231: for(k=1; k <=(nlstate+ndeath); k++){
4232: if (k != i) {
4233: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4234: if(flatdir[jk]==1){
4235: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4236: for(j=1; j <=ncovmodel; jk++,j++){
4237: printf(" p[%d]=%12.7f",jk, p[jk]);
4238: /*q[jjk]=p[jk];*/
4239: }
4240: }else{
4241: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4242: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4243: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4244: /*q[jjk]=p[jk];*/
4245: }
4246: }
4247: printf("\n");
4248: }
4249: fflush(stdout);
4250: }
4251: }
4252: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4253: #else /* FLATSUP */
1.126 brouard 4254: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4255: #endif /* FLATSUP */
4256:
4257: #ifdef LINMINORIGINAL
4258: #else
4259: free_ivector(flatdir,1,npar);
4260: #endif /* LINMINORIGINAL*/
4261: #endif /* POWELL */
1.126 brouard 4262:
1.162 brouard 4263: #ifdef NLOPT
4264: #ifdef NEWUOA
4265: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4266: #else
4267: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4268: #endif
4269: lb=vector(0,npar-1);
4270: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4271: nlopt_set_lower_bounds(opt, lb);
4272: nlopt_set_initial_step1(opt, 0.1);
4273:
4274: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4275: d->function = func;
4276: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4277: nlopt_set_min_objective(opt, myfunc, d);
4278: nlopt_set_xtol_rel(opt, ftol);
4279: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4280: printf("nlopt failed! %d\n",creturn);
4281: }
4282: else {
4283: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4284: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4285: iter=1; /* not equal */
4286: }
4287: nlopt_destroy(opt);
4288: #endif
1.319 brouard 4289: #ifdef FLATSUP
4290: /* npared = npar -flatd/ncovmodel; */
4291: /* xired= matrix(1,npared,1,npared); */
4292: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4293: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4294: /* free_matrix(xire,1,npared,1,npared); */
4295: #else /* FLATSUP */
4296: #endif /* FLATSUP */
1.126 brouard 4297: free_matrix(xi,1,npar,1,npar);
4298: fclose(ficrespow);
1.203 brouard 4299: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4300: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4301: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4302:
4303: }
4304:
4305: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4306: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4307: {
4308: double **a,**y,*x,pd;
1.203 brouard 4309: /* double **hess; */
1.164 brouard 4310: int i, j;
1.126 brouard 4311: int *indx;
4312:
4313: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4314: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4315: void lubksb(double **a, int npar, int *indx, double b[]) ;
4316: void ludcmp(double **a, int npar, int *indx, double *d) ;
4317: double gompertz(double p[]);
1.203 brouard 4318: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4319:
4320: printf("\nCalculation of the hessian matrix. Wait...\n");
4321: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4322: for (i=1;i<=npar;i++){
1.203 brouard 4323: printf("%d-",i);fflush(stdout);
4324: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4325:
4326: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4327:
4328: /* printf(" %f ",p[i]);
4329: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4330: }
4331:
4332: for (i=1;i<=npar;i++) {
4333: for (j=1;j<=npar;j++) {
4334: if (j>i) {
1.203 brouard 4335: printf(".%d-%d",i,j);fflush(stdout);
4336: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4337: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4338:
4339: hess[j][i]=hess[i][j];
4340: /*printf(" %lf ",hess[i][j]);*/
4341: }
4342: }
4343: }
4344: printf("\n");
4345: fprintf(ficlog,"\n");
4346:
4347: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4348: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4349:
4350: a=matrix(1,npar,1,npar);
4351: y=matrix(1,npar,1,npar);
4352: x=vector(1,npar);
4353: indx=ivector(1,npar);
4354: for (i=1;i<=npar;i++)
4355: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4356: ludcmp(a,npar,indx,&pd);
4357:
4358: for (j=1;j<=npar;j++) {
4359: for (i=1;i<=npar;i++) x[i]=0;
4360: x[j]=1;
4361: lubksb(a,npar,indx,x);
4362: for (i=1;i<=npar;i++){
4363: matcov[i][j]=x[i];
4364: }
4365: }
4366:
4367: printf("\n#Hessian matrix#\n");
4368: fprintf(ficlog,"\n#Hessian matrix#\n");
4369: for (i=1;i<=npar;i++) {
4370: for (j=1;j<=npar;j++) {
1.203 brouard 4371: printf("%.6e ",hess[i][j]);
4372: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4373: }
4374: printf("\n");
4375: fprintf(ficlog,"\n");
4376: }
4377:
1.203 brouard 4378: /* printf("\n#Covariance matrix#\n"); */
4379: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4380: /* for (i=1;i<=npar;i++) { */
4381: /* for (j=1;j<=npar;j++) { */
4382: /* printf("%.6e ",matcov[i][j]); */
4383: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4384: /* } */
4385: /* printf("\n"); */
4386: /* fprintf(ficlog,"\n"); */
4387: /* } */
4388:
1.126 brouard 4389: /* Recompute Inverse */
1.203 brouard 4390: /* for (i=1;i<=npar;i++) */
4391: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4392: /* ludcmp(a,npar,indx,&pd); */
4393:
4394: /* printf("\n#Hessian matrix recomputed#\n"); */
4395:
4396: /* for (j=1;j<=npar;j++) { */
4397: /* for (i=1;i<=npar;i++) x[i]=0; */
4398: /* x[j]=1; */
4399: /* lubksb(a,npar,indx,x); */
4400: /* for (i=1;i<=npar;i++){ */
4401: /* y[i][j]=x[i]; */
4402: /* printf("%.3e ",y[i][j]); */
4403: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4404: /* } */
4405: /* printf("\n"); */
4406: /* fprintf(ficlog,"\n"); */
4407: /* } */
4408:
4409: /* Verifying the inverse matrix */
4410: #ifdef DEBUGHESS
4411: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4412:
1.203 brouard 4413: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4414: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4415:
4416: for (j=1;j<=npar;j++) {
4417: for (i=1;i<=npar;i++){
1.203 brouard 4418: printf("%.2f ",y[i][j]);
4419: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4420: }
4421: printf("\n");
4422: fprintf(ficlog,"\n");
4423: }
1.203 brouard 4424: #endif
1.126 brouard 4425:
4426: free_matrix(a,1,npar,1,npar);
4427: free_matrix(y,1,npar,1,npar);
4428: free_vector(x,1,npar);
4429: free_ivector(indx,1,npar);
1.203 brouard 4430: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4431:
4432:
4433: }
4434:
4435: /*************** hessian matrix ****************/
4436: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4437: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4438: int i;
4439: int l=1, lmax=20;
1.203 brouard 4440: double k1,k2, res, fx;
1.132 brouard 4441: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4442: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4443: int k=0,kmax=10;
4444: double l1;
4445:
4446: fx=func(x);
4447: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4448: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4449: l1=pow(10,l);
4450: delts=delt;
4451: for(k=1 ; k <kmax; k=k+1){
4452: delt = delta*(l1*k);
4453: p2[theta]=x[theta] +delt;
1.145 brouard 4454: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4455: p2[theta]=x[theta]-delt;
4456: k2=func(p2)-fx;
4457: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4458: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4459:
1.203 brouard 4460: #ifdef DEBUGHESSII
1.126 brouard 4461: 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);
4462: 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);
4463: #endif
4464: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4465: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4466: k=kmax;
4467: }
4468: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4469: k=kmax; l=lmax*10;
1.126 brouard 4470: }
4471: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4472: delts=delt;
4473: }
1.203 brouard 4474: } /* End loop k */
1.126 brouard 4475: }
4476: delti[theta]=delts;
4477: return res;
4478:
4479: }
4480:
1.203 brouard 4481: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4482: {
4483: int i;
1.164 brouard 4484: int l=1, lmax=20;
1.126 brouard 4485: double k1,k2,k3,k4,res,fx;
1.132 brouard 4486: double p2[MAXPARM+1];
1.203 brouard 4487: int k, kmax=1;
4488: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4489:
4490: int firstime=0;
1.203 brouard 4491:
1.126 brouard 4492: fx=func(x);
1.203 brouard 4493: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4494: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4495: p2[thetai]=x[thetai]+delti[thetai]*k;
4496: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4497: k1=func(p2)-fx;
4498:
1.203 brouard 4499: p2[thetai]=x[thetai]+delti[thetai]*k;
4500: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4501: k2=func(p2)-fx;
4502:
1.203 brouard 4503: p2[thetai]=x[thetai]-delti[thetai]*k;
4504: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4505: k3=func(p2)-fx;
4506:
1.203 brouard 4507: p2[thetai]=x[thetai]-delti[thetai]*k;
4508: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4509: k4=func(p2)-fx;
1.203 brouard 4510: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4511: if(k1*k2*k3*k4 <0.){
1.208 brouard 4512: firstime=1;
1.203 brouard 4513: kmax=kmax+10;
1.208 brouard 4514: }
4515: if(kmax >=10 || firstime ==1){
1.246 brouard 4516: 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);
4517: 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 4518: 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);
4519: 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);
4520: }
4521: #ifdef DEBUGHESSIJ
4522: v1=hess[thetai][thetai];
4523: v2=hess[thetaj][thetaj];
4524: cv12=res;
4525: /* Computing eigen value of Hessian matrix */
4526: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4527: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4528: if ((lc2 <0) || (lc1 <0) ){
4529: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4530: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4531: 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);
4532: 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);
4533: }
1.126 brouard 4534: #endif
4535: }
4536: return res;
4537: }
4538:
1.203 brouard 4539: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4540: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4541: /* { */
4542: /* int i; */
4543: /* int l=1, lmax=20; */
4544: /* double k1,k2,k3,k4,res,fx; */
4545: /* double p2[MAXPARM+1]; */
4546: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4547: /* int k=0,kmax=10; */
4548: /* double l1; */
4549:
4550: /* fx=func(x); */
4551: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4552: /* l1=pow(10,l); */
4553: /* delts=delt; */
4554: /* for(k=1 ; k <kmax; k=k+1){ */
4555: /* delt = delti*(l1*k); */
4556: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4557: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4558: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4559: /* k1=func(p2)-fx; */
4560:
4561: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4562: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4563: /* k2=func(p2)-fx; */
4564:
4565: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4566: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4567: /* k3=func(p2)-fx; */
4568:
4569: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4570: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4571: /* k4=func(p2)-fx; */
4572: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4573: /* #ifdef DEBUGHESSIJ */
4574: /* 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); */
4575: /* 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); */
4576: /* #endif */
4577: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4578: /* k=kmax; */
4579: /* } */
4580: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4581: /* k=kmax; l=lmax*10; */
4582: /* } */
4583: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4584: /* delts=delt; */
4585: /* } */
4586: /* } /\* End loop k *\/ */
4587: /* } */
4588: /* delti[theta]=delts; */
4589: /* return res; */
4590: /* } */
4591:
4592:
1.126 brouard 4593: /************** Inverse of matrix **************/
4594: void ludcmp(double **a, int n, int *indx, double *d)
4595: {
4596: int i,imax,j,k;
4597: double big,dum,sum,temp;
4598: double *vv;
4599:
4600: vv=vector(1,n);
4601: *d=1.0;
4602: for (i=1;i<=n;i++) {
4603: big=0.0;
4604: for (j=1;j<=n;j++)
4605: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4606: if (big == 0.0){
4607: printf(" Singular Hessian matrix at row %d:\n",i);
4608: for (j=1;j<=n;j++) {
4609: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4610: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4611: }
4612: fflush(ficlog);
4613: fclose(ficlog);
4614: nrerror("Singular matrix in routine ludcmp");
4615: }
1.126 brouard 4616: vv[i]=1.0/big;
4617: }
4618: for (j=1;j<=n;j++) {
4619: for (i=1;i<j;i++) {
4620: sum=a[i][j];
4621: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4622: a[i][j]=sum;
4623: }
4624: big=0.0;
4625: for (i=j;i<=n;i++) {
4626: sum=a[i][j];
4627: for (k=1;k<j;k++)
4628: sum -= a[i][k]*a[k][j];
4629: a[i][j]=sum;
4630: if ( (dum=vv[i]*fabs(sum)) >= big) {
4631: big=dum;
4632: imax=i;
4633: }
4634: }
4635: if (j != imax) {
4636: for (k=1;k<=n;k++) {
4637: dum=a[imax][k];
4638: a[imax][k]=a[j][k];
4639: a[j][k]=dum;
4640: }
4641: *d = -(*d);
4642: vv[imax]=vv[j];
4643: }
4644: indx[j]=imax;
4645: if (a[j][j] == 0.0) a[j][j]=TINY;
4646: if (j != n) {
4647: dum=1.0/(a[j][j]);
4648: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4649: }
4650: }
4651: free_vector(vv,1,n); /* Doesn't work */
4652: ;
4653: }
4654:
4655: void lubksb(double **a, int n, int *indx, double b[])
4656: {
4657: int i,ii=0,ip,j;
4658: double sum;
4659:
4660: for (i=1;i<=n;i++) {
4661: ip=indx[i];
4662: sum=b[ip];
4663: b[ip]=b[i];
4664: if (ii)
4665: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4666: else if (sum) ii=i;
4667: b[i]=sum;
4668: }
4669: for (i=n;i>=1;i--) {
4670: sum=b[i];
4671: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4672: b[i]=sum/a[i][i];
4673: }
4674: }
4675:
4676: void pstamp(FILE *fichier)
4677: {
1.196 brouard 4678: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4679: }
4680:
1.297 brouard 4681: void date2dmy(double date,double *day, double *month, double *year){
4682: double yp=0., yp1=0., yp2=0.;
4683:
4684: yp1=modf(date,&yp);/* extracts integral of date in yp and
4685: fractional in yp1 */
4686: *year=yp;
4687: yp2=modf((yp1*12),&yp);
4688: *month=yp;
4689: yp1=modf((yp2*30.5),&yp);
4690: *day=yp;
4691: if(*day==0) *day=1;
4692: if(*month==0) *month=1;
4693: }
4694:
1.253 brouard 4695:
4696:
1.126 brouard 4697: /************ Frequencies ********************/
1.251 brouard 4698: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4699: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4700: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4701: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4702:
1.265 brouard 4703: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4704: int iind=0, iage=0;
4705: int mi; /* Effective wave */
4706: int first;
4707: double ***freq; /* Frequencies */
1.268 brouard 4708: 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 */
4709: 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 4710: double *meanq, *stdq, *idq;
1.226 brouard 4711: double **meanqt;
4712: double *pp, **prop, *posprop, *pospropt;
4713: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4714: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4715: double agebegin, ageend;
4716:
4717: pp=vector(1,nlstate);
1.251 brouard 4718: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4719: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4720: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4721: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4722: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4723: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4724: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4725: meanqt=matrix(1,lastpass,1,nqtveff);
4726: strcpy(fileresp,"P_");
4727: strcat(fileresp,fileresu);
4728: /*strcat(fileresphtm,fileresu);*/
4729: if((ficresp=fopen(fileresp,"w"))==NULL) {
4730: printf("Problem with prevalence resultfile: %s\n", fileresp);
4731: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4732: exit(0);
4733: }
1.240 brouard 4734:
1.226 brouard 4735: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4736: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4737: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4738: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4739: fflush(ficlog);
4740: exit(70);
4741: }
4742: else{
4743: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4744: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4745: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4746: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4747: }
1.319 brouard 4748: 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 4749:
1.226 brouard 4750: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4751: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4752: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4753: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4754: fflush(ficlog);
4755: exit(70);
1.240 brouard 4756: } else{
1.226 brouard 4757: 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 4758: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4759: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4760: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4761: }
1.319 brouard 4762: 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 4763:
1.253 brouard 4764: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4765: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4766: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4767: j1=0;
1.126 brouard 4768:
1.227 brouard 4769: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4770: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4771: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4772:
4773:
1.226 brouard 4774: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4775: reference=low_education V1=0,V2=0
4776: med_educ V1=1 V2=0,
4777: high_educ V1=0 V2=1
4778: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4779: */
1.249 brouard 4780: dateintsum=0;
4781: k2cpt=0;
4782:
1.253 brouard 4783: if(cptcoveff == 0 )
1.265 brouard 4784: nl=1; /* Constant and age model only */
1.253 brouard 4785: else
4786: nl=2;
1.265 brouard 4787:
4788: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4789: /* Loop on nj=1 or 2 if dummy covariates j!=0
4790: * Loop on j1(1 to 2**cptcoveff) covariate combination
4791: * freq[s1][s2][iage] =0.
4792: * Loop on iind
4793: * ++freq[s1][s2][iage] weighted
4794: * end iind
4795: * if covariate and j!0
4796: * headers Variable on one line
4797: * endif cov j!=0
4798: * header of frequency table by age
4799: * Loop on age
4800: * pp[s1]+=freq[s1][s2][iage] weighted
4801: * pos+=freq[s1][s2][iage] weighted
4802: * Loop on s1 initial state
4803: * fprintf(ficresp
4804: * end s1
4805: * end age
4806: * if j!=0 computes starting values
4807: * end compute starting values
4808: * end j1
4809: * end nl
4810: */
1.253 brouard 4811: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4812: if(nj==1)
4813: j=0; /* First pass for the constant */
1.265 brouard 4814: else{
1.253 brouard 4815: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4816: }
1.251 brouard 4817: first=1;
1.265 brouard 4818: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on all covariates combination of the model, excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
1.251 brouard 4819: posproptt=0.;
4820: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4821: scanf("%d", i);*/
4822: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4823: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4824: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4825: freq[i][s2][m]=0;
1.251 brouard 4826:
4827: for (i=1; i<=nlstate; i++) {
1.240 brouard 4828: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4829: prop[i][m]=0;
4830: posprop[i]=0;
4831: pospropt[i]=0;
4832: }
1.283 brouard 4833: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4834: idq[z1]=0.;
4835: meanq[z1]=0.;
4836: stdq[z1]=0.;
1.283 brouard 4837: }
4838: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4839: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4840: /* meanqt[m][z1]=0.; */
4841: /* } */
4842: /* } */
1.251 brouard 4843: /* dateintsum=0; */
4844: /* k2cpt=0; */
4845:
1.265 brouard 4846: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4847: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4848: bool=1;
4849: if(j !=0){
4850: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4851: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4852: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4853: /* if(Tvaraff[z1] ==-20){ */
4854: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4855: /* }else if(Tvaraff[z1] ==-10){ */
4856: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4857: /* }else */
4858: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4859: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4860: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4861: /* 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",
4862: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4863: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4864: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4865: } /* Onlyf fixed */
4866: } /* end z1 */
4867: } /* cptcovn > 0 */
4868: } /* end any */
4869: }/* end j==0 */
1.265 brouard 4870: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4871: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4872: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4873: m=mw[mi][iind];
4874: if(j!=0){
4875: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4876: for (z1=1; z1<=cptcoveff; z1++) {
4877: if( Fixed[Tmodelind[z1]]==1){
4878: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4879: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4880: value is -1, we don't select. It differs from the
4881: constant and age model which counts them. */
4882: bool=0; /* not selected */
4883: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4884: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4885: bool=0;
4886: }
4887: }
4888: }
4889: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4890: } /* end j==0 */
4891: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4892: if(bool==1){ /*Selected */
1.251 brouard 4893: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4894: and mw[mi+1][iind]. dh depends on stepm. */
4895: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4896: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4897: if(m >=firstpass && m <=lastpass){
4898: k2=anint[m][iind]+(mint[m][iind]/12.);
4899: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4900: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4901: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4902: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4903: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4904: if (m<lastpass) {
4905: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4906: /* 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]); */
4907: if(s[m][iind]==-1)
4908: 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.));
4909: 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 4910: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
4911: if(!isnan(covar[ncovcol+z1][iind])){
4912: idq[z1]=idq[z1]+weight[iind];
4913: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4914: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
4915: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4916: }
1.284 brouard 4917: }
1.251 brouard 4918: /* if((int)agev[m][iind] == 55) */
4919: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4920: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4921: 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 4922: }
1.251 brouard 4923: } /* end if between passes */
4924: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4925: dateintsum=dateintsum+k2; /* on all covariates ?*/
4926: k2cpt++;
4927: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4928: }
1.251 brouard 4929: }else{
4930: bool=1;
4931: }/* end bool 2 */
4932: } /* end m */
1.284 brouard 4933: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4934: /* idq[z1]=idq[z1]+weight[iind]; */
4935: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4936: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4937: /* } */
1.251 brouard 4938: } /* end bool */
4939: } /* end iind = 1 to imx */
1.319 brouard 4940: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 4941: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4942:
4943:
4944: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4945: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4946: pstamp(ficresp);
1.251 brouard 4947: if (cptcoveff>0 && j!=0){
1.265 brouard 4948: pstamp(ficresp);
1.251 brouard 4949: printf( "\n#********** Variable ");
4950: fprintf(ficresp, "\n#********** Variable ");
4951: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4952: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4953: fprintf(ficlog, "\n#********** Variable ");
4954: for (z1=1; z1<=cptcoveff; z1++){
4955: if(!FixedV[Tvaraff[z1]]){
4956: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4957: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4958: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4959: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4960: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4961: }else{
1.251 brouard 4962: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4963: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4964: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4965: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4966: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4967: }
4968: }
4969: printf( "**********\n#");
4970: fprintf(ficresp, "**********\n#");
4971: fprintf(ficresphtm, "**********</h3>\n");
4972: fprintf(ficresphtmfr, "**********</h3>\n");
4973: fprintf(ficlog, "**********\n");
4974: }
1.284 brouard 4975: /*
4976: Printing means of quantitative variables if any
4977: */
4978: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 4979: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 4980: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 4981: if(weightopt==1){
4982: printf(" Weighted mean and standard deviation of");
4983: fprintf(ficlog," Weighted mean and standard deviation of");
4984: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4985: }
1.311 brouard 4986: /* mu = \frac{w x}{\sum w}
4987: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
4988: */
4989: 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]));
4990: 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]));
4991: 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 4992: }
4993: /* for (z1=1; z1<= nqtveff; z1++) { */
4994: /* for(m=1;m<=lastpass;m++){ */
4995: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4996: /* } */
4997: /* } */
1.283 brouard 4998:
1.251 brouard 4999: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 5000: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
5001: fprintf(ficresp, " Age");
5002: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.251 brouard 5003: for(i=1; i<=nlstate;i++) {
1.265 brouard 5004: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5005: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5006: }
1.265 brouard 5007: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5008: fprintf(ficresphtm, "\n");
5009:
5010: /* Header of frequency table by age */
5011: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5012: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5013: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5014: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5015: if(s2!=0 && m!=0)
5016: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5017: }
1.226 brouard 5018: }
1.251 brouard 5019: fprintf(ficresphtmfr, "\n");
5020:
5021: /* For each age */
5022: for(iage=iagemin; iage <= iagemax+3; iage++){
5023: fprintf(ficresphtm,"<tr>");
5024: if(iage==iagemax+1){
5025: fprintf(ficlog,"1");
5026: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5027: }else if(iage==iagemax+2){
5028: fprintf(ficlog,"0");
5029: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5030: }else if(iage==iagemax+3){
5031: fprintf(ficlog,"Total");
5032: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5033: }else{
1.240 brouard 5034: if(first==1){
1.251 brouard 5035: first=0;
5036: printf("See log file for details...\n");
5037: }
5038: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5039: fprintf(ficlog,"Age %d", iage);
5040: }
1.265 brouard 5041: for(s1=1; s1 <=nlstate ; s1++){
5042: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5043: pp[s1] += freq[s1][m][iage];
1.251 brouard 5044: }
1.265 brouard 5045: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5046: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5047: pos += freq[s1][m][iage];
5048: if(pp[s1]>=1.e-10){
1.251 brouard 5049: if(first==1){
1.265 brouard 5050: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5051: }
1.265 brouard 5052: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5053: }else{
5054: if(first==1)
1.265 brouard 5055: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5056: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5057: }
5058: }
5059:
1.265 brouard 5060: for(s1=1; s1 <=nlstate ; s1++){
5061: /* posprop[s1]=0; */
5062: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5063: pp[s1] += freq[s1][m][iage];
5064: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5065:
5066: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5067: pos += pp[s1]; /* pos is the total number of transitions until this age */
5068: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5069: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5070: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5071: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5072: }
5073:
5074: /* Writing ficresp */
5075: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
5076: if( iage <= iagemax){
5077: fprintf(ficresp," %d",iage);
5078: }
5079: }else if( nj==2){
5080: if( iage <= iagemax){
5081: fprintf(ficresp," %d",iage);
5082: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
5083: }
1.240 brouard 5084: }
1.265 brouard 5085: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5086: if(pos>=1.e-5){
1.251 brouard 5087: if(first==1)
1.265 brouard 5088: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5089: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5090: }else{
5091: if(first==1)
1.265 brouard 5092: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5093: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5094: }
5095: if( iage <= iagemax){
5096: if(pos>=1.e-5){
1.265 brouard 5097: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
5098: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5099: }else if( nj==2){
5100: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5101: }
5102: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5103: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5104: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5105: } else{
5106: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
5107: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5108: }
1.240 brouard 5109: }
1.265 brouard 5110: pospropt[s1] +=posprop[s1];
5111: } /* end loop s1 */
1.251 brouard 5112: /* pospropt=0.; */
1.265 brouard 5113: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5114: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5115: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5116: if(first==1){
1.265 brouard 5117: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5118: }
1.265 brouard 5119: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5120: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5121: }
1.265 brouard 5122: if(s1!=0 && m!=0)
5123: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5124: }
1.265 brouard 5125: } /* end loop s1 */
1.251 brouard 5126: posproptt=0.;
1.265 brouard 5127: for(s1=1; s1 <=nlstate; s1++){
5128: posproptt += pospropt[s1];
1.251 brouard 5129: }
5130: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5131: fprintf(ficresphtm,"</tr>\n");
5132: if((cptcoveff==0 && nj==1)|| nj==2 ) {
5133: if(iage <= iagemax)
5134: fprintf(ficresp,"\n");
1.240 brouard 5135: }
1.251 brouard 5136: if(first==1)
5137: printf("Others in log...\n");
5138: fprintf(ficlog,"\n");
5139: } /* end loop age iage */
1.265 brouard 5140:
1.251 brouard 5141: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5142: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5143: if(posproptt < 1.e-5){
1.265 brouard 5144: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5145: }else{
1.265 brouard 5146: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5147: }
1.226 brouard 5148: }
1.251 brouard 5149: fprintf(ficresphtm,"</tr>\n");
5150: fprintf(ficresphtm,"</table>\n");
5151: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5152: if(posproptt < 1.e-5){
1.251 brouard 5153: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5154: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5155: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5156: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5157: invalidvarcomb[j1]=1;
1.226 brouard 5158: }else{
1.251 brouard 5159: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
5160: invalidvarcomb[j1]=0;
1.226 brouard 5161: }
1.251 brouard 5162: fprintf(ficresphtmfr,"</table>\n");
5163: fprintf(ficlog,"\n");
5164: if(j!=0){
5165: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5166: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5167: for(k=1; k <=(nlstate+ndeath); k++){
5168: if (k != i) {
1.265 brouard 5169: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5170: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5171: if(j1==1){ /* All dummy covariates to zero */
5172: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5173: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5174: printf("%d%d ",i,k);
5175: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5176: 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]));
5177: 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]));
5178: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5179: }
1.253 brouard 5180: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5181: for(iage=iagemin; iage <= iagemax+3; iage++){
5182: x[iage]= (double)iage;
5183: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5184: /* 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 5185: }
1.268 brouard 5186: /* Some are not finite, but linreg will ignore these ages */
5187: no=0;
1.253 brouard 5188: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5189: pstart[s1]=b;
5190: pstart[s1-1]=a;
1.252 brouard 5191: }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 */
5192: 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]);
5193: 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 5194: 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 5195: printf("%d%d ",i,k);
5196: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5197: 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 5198: }else{ /* Other cases, like quantitative fixed or varying covariates */
5199: ;
5200: }
5201: /* printf("%12.7f )", param[i][jj][k]); */
5202: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5203: s1++;
1.251 brouard 5204: } /* end jj */
5205: } /* end k!= i */
5206: } /* end k */
1.265 brouard 5207: } /* end i, s1 */
1.251 brouard 5208: } /* end j !=0 */
5209: } /* end selected combination of covariate j1 */
5210: if(j==0){ /* We can estimate starting values from the occurences in each case */
5211: printf("#Freqsummary: Starting values for the constants:\n");
5212: fprintf(ficlog,"\n");
1.265 brouard 5213: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5214: for(k=1; k <=(nlstate+ndeath); k++){
5215: if (k != i) {
5216: printf("%d%d ",i,k);
5217: fprintf(ficlog,"%d%d ",i,k);
5218: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5219: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5220: if(jj==1){ /* Age has to be done */
1.265 brouard 5221: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5222: 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]));
5223: 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 5224: }
5225: /* printf("%12.7f )", param[i][jj][k]); */
5226: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5227: s1++;
1.250 brouard 5228: }
1.251 brouard 5229: printf("\n");
5230: fprintf(ficlog,"\n");
1.250 brouard 5231: }
5232: }
1.284 brouard 5233: } /* end of state i */
1.251 brouard 5234: printf("#Freqsummary\n");
5235: fprintf(ficlog,"\n");
1.265 brouard 5236: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5237: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5238: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5239: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5240: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5241: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5242: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5243: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5244: /* } */
5245: }
1.265 brouard 5246: } /* end loop s1 */
1.251 brouard 5247:
5248: printf("\n");
5249: fprintf(ficlog,"\n");
5250: } /* end j=0 */
1.249 brouard 5251: } /* end j */
1.252 brouard 5252:
1.253 brouard 5253: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5254: for(i=1, jk=1; i <=nlstate; i++){
5255: for(j=1; j <=nlstate+ndeath; j++){
5256: if(j!=i){
5257: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5258: printf("%1d%1d",i,j);
5259: fprintf(ficparo,"%1d%1d",i,j);
5260: for(k=1; k<=ncovmodel;k++){
5261: /* printf(" %lf",param[i][j][k]); */
5262: /* fprintf(ficparo," %lf",param[i][j][k]); */
5263: p[jk]=pstart[jk];
5264: printf(" %f ",pstart[jk]);
5265: fprintf(ficparo," %f ",pstart[jk]);
5266: jk++;
5267: }
5268: printf("\n");
5269: fprintf(ficparo,"\n");
5270: }
5271: }
5272: }
5273: } /* end mle=-2 */
1.226 brouard 5274: dateintmean=dateintsum/k2cpt;
1.296 brouard 5275: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5276:
1.226 brouard 5277: fclose(ficresp);
5278: fclose(ficresphtm);
5279: fclose(ficresphtmfr);
1.283 brouard 5280: free_vector(idq,1,nqfveff);
1.226 brouard 5281: free_vector(meanq,1,nqfveff);
1.284 brouard 5282: free_vector(stdq,1,nqfveff);
1.226 brouard 5283: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5284: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5285: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5286: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5287: free_vector(pospropt,1,nlstate);
5288: free_vector(posprop,1,nlstate);
1.251 brouard 5289: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5290: free_vector(pp,1,nlstate);
5291: /* End of freqsummary */
5292: }
1.126 brouard 5293:
1.268 brouard 5294: /* Simple linear regression */
5295: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5296:
5297: /* y=a+bx regression */
5298: double sumx = 0.0; /* sum of x */
5299: double sumx2 = 0.0; /* sum of x**2 */
5300: double sumxy = 0.0; /* sum of x * y */
5301: double sumy = 0.0; /* sum of y */
5302: double sumy2 = 0.0; /* sum of y**2 */
5303: double sume2 = 0.0; /* sum of square or residuals */
5304: double yhat;
5305:
5306: double denom=0;
5307: int i;
5308: int ne=*no;
5309:
5310: for ( i=ifi, ne=0;i<=ila;i++) {
5311: if(!isfinite(x[i]) || !isfinite(y[i])){
5312: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5313: continue;
5314: }
5315: ne=ne+1;
5316: sumx += x[i];
5317: sumx2 += x[i]*x[i];
5318: sumxy += x[i] * y[i];
5319: sumy += y[i];
5320: sumy2 += y[i]*y[i];
5321: denom = (ne * sumx2 - sumx*sumx);
5322: /* 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); */
5323: }
5324:
5325: denom = (ne * sumx2 - sumx*sumx);
5326: if (denom == 0) {
5327: // vertical, slope m is infinity
5328: *b = INFINITY;
5329: *a = 0;
5330: if (r) *r = 0;
5331: return 1;
5332: }
5333:
5334: *b = (ne * sumxy - sumx * sumy) / denom;
5335: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5336: if (r!=NULL) {
5337: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5338: sqrt((sumx2 - sumx*sumx/ne) *
5339: (sumy2 - sumy*sumy/ne));
5340: }
5341: *no=ne;
5342: for ( i=ifi, ne=0;i<=ila;i++) {
5343: if(!isfinite(x[i]) || !isfinite(y[i])){
5344: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5345: continue;
5346: }
5347: ne=ne+1;
5348: yhat = y[i] - *a -*b* x[i];
5349: sume2 += yhat * yhat ;
5350:
5351: denom = (ne * sumx2 - sumx*sumx);
5352: /* 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); */
5353: }
5354: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5355: *sa= *sb * sqrt(sumx2/ne);
5356:
5357: return 0;
5358: }
5359:
1.126 brouard 5360: /************ Prevalence ********************/
1.227 brouard 5361: 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)
5362: {
5363: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5364: in each health status at the date of interview (if between dateprev1 and dateprev2).
5365: We still use firstpass and lastpass as another selection.
5366: */
1.126 brouard 5367:
1.227 brouard 5368: int i, m, jk, j1, bool, z1,j, iv;
5369: int mi; /* Effective wave */
5370: int iage;
5371: double agebegin, ageend;
5372:
5373: double **prop;
5374: double posprop;
5375: double y2; /* in fractional years */
5376: int iagemin, iagemax;
5377: int first; /** to stop verbosity which is redirected to log file */
5378:
5379: iagemin= (int) agemin;
5380: iagemax= (int) agemax;
5381: /*pp=vector(1,nlstate);*/
1.251 brouard 5382: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5383: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5384: j1=0;
1.222 brouard 5385:
1.227 brouard 5386: /*j=cptcoveff;*/
5387: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5388:
1.288 brouard 5389: first=0;
1.227 brouard 5390: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5391: for (i=1; i<=nlstate; i++)
1.251 brouard 5392: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5393: prop[i][iage]=0.0;
5394: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5395: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5396: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5397:
5398: for (i=1; i<=imx; i++) { /* Each individual */
5399: bool=1;
5400: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5401: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5402: m=mw[mi][i];
5403: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5404: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5405: for (z1=1; z1<=cptcoveff; z1++){
5406: if( Fixed[Tmodelind[z1]]==1){
5407: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5408: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5409: bool=0;
5410: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5411: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5412: bool=0;
5413: }
5414: }
5415: if(bool==1){ /* Otherwise we skip that wave/person */
5416: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5417: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5418: if(m >=firstpass && m <=lastpass){
5419: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5420: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5421: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5422: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5423: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5424: 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);
5425: exit(1);
5426: }
5427: if (s[m][i]>0 && s[m][i]<=nlstate) {
5428: /*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]]);*/
5429: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5430: prop[s[m][i]][iagemax+3] += weight[i];
5431: } /* end valid statuses */
5432: } /* end selection of dates */
5433: } /* end selection of waves */
5434: } /* end bool */
5435: } /* end wave */
5436: } /* end individual */
5437: for(i=iagemin; i <= iagemax+3; i++){
5438: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5439: posprop += prop[jk][i];
5440: }
5441:
5442: for(jk=1; jk <=nlstate ; jk++){
5443: if( i <= iagemax){
5444: if(posprop>=1.e-5){
5445: probs[i][jk][j1]= prop[jk][i]/posprop;
5446: } else{
1.288 brouard 5447: if(!first){
5448: first=1;
1.266 brouard 5449: 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]);
5450: }else{
1.288 brouard 5451: 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 5452: }
5453: }
5454: }
5455: }/* end jk */
5456: }/* end i */
1.222 brouard 5457: /*} *//* end i1 */
1.227 brouard 5458: } /* end j1 */
1.222 brouard 5459:
1.227 brouard 5460: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5461: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5462: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5463: } /* End of prevalence */
1.126 brouard 5464:
5465: /************* Waves Concatenation ***************/
5466:
5467: 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)
5468: {
1.298 brouard 5469: /* 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 5470: Death is a valid wave (if date is known).
5471: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5472: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5473: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5474: */
1.126 brouard 5475:
1.224 brouard 5476: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5477: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5478: double sum=0., jmean=0.;*/
1.224 brouard 5479: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5480: int j, k=0,jk, ju, jl;
5481: double sum=0.;
5482: first=0;
1.214 brouard 5483: firstwo=0;
1.217 brouard 5484: firsthree=0;
1.218 brouard 5485: firstfour=0;
1.164 brouard 5486: jmin=100000;
1.126 brouard 5487: jmax=-1;
5488: jmean=0.;
1.224 brouard 5489:
5490: /* Treating live states */
1.214 brouard 5491: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5492: mi=0; /* First valid wave */
1.227 brouard 5493: mli=0; /* Last valid wave */
1.309 brouard 5494: m=firstpass; /* Loop on waves */
5495: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5496: 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 */
5497: mli=m-1;/* mw[++mi][i]=m-1; */
5498: }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 5499: 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 5500: mli=m;
1.224 brouard 5501: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5502: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5503: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5504: }
1.309 brouard 5505: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5506: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5507: break;
1.224 brouard 5508: #else
1.317 brouard 5509: 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 5510: if(firsthree == 0){
1.302 brouard 5511: 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 5512: firsthree=1;
1.317 brouard 5513: }else if(firsthree >=1 && firsthree < 10){
5514: 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);
5515: firsthree++;
5516: }else if(firsthree == 10){
5517: printf("Information, too many Information flags: no more reported to log either\n");
5518: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5519: firsthree++;
5520: }else{
5521: firsthree++;
1.227 brouard 5522: }
1.309 brouard 5523: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5524: mli=m;
5525: }
5526: if(s[m][i]==-2){ /* Vital status is really unknown */
5527: nbwarn++;
1.309 brouard 5528: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5529: 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);
5530: 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);
5531: }
5532: break;
5533: }
5534: break;
1.224 brouard 5535: #endif
1.227 brouard 5536: }/* End m >= lastpass */
1.126 brouard 5537: }/* end while */
1.224 brouard 5538:
1.227 brouard 5539: /* 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 5540: /* After last pass */
1.224 brouard 5541: /* Treating death states */
1.214 brouard 5542: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5543: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5544: /* } */
1.126 brouard 5545: mi++; /* Death is another wave */
5546: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5547: /* Only death is a correct wave */
1.126 brouard 5548: mw[mi][i]=m;
1.257 brouard 5549: } /* else not in a death state */
1.224 brouard 5550: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5551: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5552: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5553: 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 5554: nbwarn++;
5555: if(firstfiv==0){
1.309 brouard 5556: 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 5557: firstfiv=1;
5558: }else{
1.309 brouard 5559: 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 5560: }
1.309 brouard 5561: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5562: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5563: nberr++;
5564: if(firstwo==0){
1.309 brouard 5565: 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 5566: firstwo=1;
5567: }
1.309 brouard 5568: 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 5569: }
1.257 brouard 5570: }else{ /* if date of interview is unknown */
1.227 brouard 5571: /* death is known but not confirmed by death status at any wave */
5572: if(firstfour==0){
1.309 brouard 5573: 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 5574: firstfour=1;
5575: }
1.309 brouard 5576: 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 5577: }
1.224 brouard 5578: } /* end if date of death is known */
5579: #endif
1.309 brouard 5580: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5581: /* wav[i]=mw[mi][i]; */
1.126 brouard 5582: if(mi==0){
5583: nbwarn++;
5584: if(first==0){
1.227 brouard 5585: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5586: first=1;
1.126 brouard 5587: }
5588: if(first==1){
1.227 brouard 5589: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5590: }
5591: } /* end mi==0 */
5592: } /* End individuals */
1.214 brouard 5593: /* wav and mw are no more changed */
1.223 brouard 5594:
1.317 brouard 5595: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
5596: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
5597:
5598:
1.126 brouard 5599: for(i=1; i<=imx; i++){
5600: for(mi=1; mi<wav[i];mi++){
5601: if (stepm <=0)
1.227 brouard 5602: dh[mi][i]=1;
1.126 brouard 5603: else{
1.260 brouard 5604: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5605: if (agedc[i] < 2*AGESUP) {
5606: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5607: if(j==0) j=1; /* Survives at least one month after exam */
5608: else if(j<0){
5609: nberr++;
5610: 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]);
5611: j=1; /* Temporary Dangerous patch */
5612: 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);
5613: 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]);
5614: 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);
5615: }
5616: k=k+1;
5617: if (j >= jmax){
5618: jmax=j;
5619: ijmax=i;
5620: }
5621: if (j <= jmin){
5622: jmin=j;
5623: ijmin=i;
5624: }
5625: sum=sum+j;
5626: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5627: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5628: }
5629: }
5630: else{
5631: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5632: /* 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 5633:
1.227 brouard 5634: k=k+1;
5635: if (j >= jmax) {
5636: jmax=j;
5637: ijmax=i;
5638: }
5639: else if (j <= jmin){
5640: jmin=j;
5641: ijmin=i;
5642: }
5643: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5644: /*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]);*/
5645: if(j<0){
5646: nberr++;
5647: 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]);
5648: 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]);
5649: }
5650: sum=sum+j;
5651: }
5652: jk= j/stepm;
5653: jl= j -jk*stepm;
5654: ju= j -(jk+1)*stepm;
5655: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5656: if(jl==0){
5657: dh[mi][i]=jk;
5658: bh[mi][i]=0;
5659: }else{ /* We want a negative bias in order to only have interpolation ie
5660: * to avoid the price of an extra matrix product in likelihood */
5661: dh[mi][i]=jk+1;
5662: bh[mi][i]=ju;
5663: }
5664: }else{
5665: if(jl <= -ju){
5666: dh[mi][i]=jk;
5667: bh[mi][i]=jl; /* bias is positive if real duration
5668: * is higher than the multiple of stepm and negative otherwise.
5669: */
5670: }
5671: else{
5672: dh[mi][i]=jk+1;
5673: bh[mi][i]=ju;
5674: }
5675: if(dh[mi][i]==0){
5676: dh[mi][i]=1; /* At least one step */
5677: bh[mi][i]=ju; /* At least one step */
5678: /* 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);*/
5679: }
5680: } /* end if mle */
1.126 brouard 5681: }
5682: } /* end wave */
5683: }
5684: jmean=sum/k;
5685: 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 5686: 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 5687: }
1.126 brouard 5688:
5689: /*********** Tricode ****************************/
1.220 brouard 5690: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5691: {
5692: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5693: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5694: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5695: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5696: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5697: */
1.130 brouard 5698:
1.242 brouard 5699: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5700: int modmaxcovj=0; /* Modality max of covariates j */
5701: int cptcode=0; /* Modality max of covariates j */
5702: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5703:
5704:
1.242 brouard 5705: /* cptcoveff=0; */
5706: /* *cptcov=0; */
1.126 brouard 5707:
1.242 brouard 5708: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5709: for (k=1; k <= maxncov; k++)
5710: for(j=1; j<=2; j++)
5711: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5712:
1.242 brouard 5713: /* Loop on covariates without age and products and no quantitative variable */
5714: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5715: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5716: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5717: switch(Fixed[k]) {
5718: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5719: modmaxcovj=0;
5720: modmincovj=0;
1.242 brouard 5721: 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*/
5722: ij=(int)(covar[Tvar[k]][i]);
5723: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5724: * If product of Vn*Vm, still boolean *:
5725: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5726: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5727: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5728: modality of the nth covariate of individual i. */
5729: if (ij > modmaxcovj)
5730: modmaxcovj=ij;
5731: else if (ij < modmincovj)
5732: modmincovj=ij;
1.287 brouard 5733: if (ij <0 || ij >1 ){
1.311 brouard 5734: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5735: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5736: fflush(ficlog);
5737: exit(1);
1.287 brouard 5738: }
5739: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5740: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5741: exit(1);
5742: }else
5743: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5744: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5745: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5746: /* getting the maximum value of the modality of the covariate
5747: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5748: female ies 1, then modmaxcovj=1.
5749: */
5750: } /* end for loop on individuals i */
5751: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5752: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5753: cptcode=modmaxcovj;
5754: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5755: /*for (i=0; i<=cptcode; i++) {*/
5756: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5757: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5758: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5759: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5760: if( j != -1){
5761: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5762: covariate for which somebody answered excluding
5763: undefined. Usually 2: 0 and 1. */
5764: }
5765: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5766: covariate for which somebody answered including
5767: undefined. Usually 3: -1, 0 and 1. */
5768: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5769: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5770: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5771:
1.242 brouard 5772: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5773: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5774: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5775: /* modmincovj=3; modmaxcovj = 7; */
5776: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5777: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5778: /* defining two dummy variables: variables V1_1 and V1_2.*/
5779: /* nbcode[Tvar[j]][ij]=k; */
5780: /* nbcode[Tvar[j]][1]=0; */
5781: /* nbcode[Tvar[j]][2]=1; */
5782: /* nbcode[Tvar[j]][3]=2; */
5783: /* To be continued (not working yet). */
5784: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5785:
5786: /* 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*/
5787: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5788: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5789: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5790: /*, could be restored in the future */
5791: 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 5792: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5793: break;
5794: }
5795: ij++;
1.287 brouard 5796: 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 5797: cptcode = ij; /* New max modality for covar j */
5798: } /* end of loop on modality i=-1 to 1 or more */
5799: break;
5800: case 1: /* Testing on varying covariate, could be simple and
5801: * should look at waves or product of fixed *
5802: * varying. No time to test -1, assuming 0 and 1 only */
5803: ij=0;
5804: for(i=0; i<=1;i++){
5805: nbcode[Tvar[k]][++ij]=i;
5806: }
5807: break;
5808: default:
5809: break;
5810: } /* end switch */
5811: } /* end dummy test */
1.311 brouard 5812: if(Dummy[k]==1 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5813: 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*/
5814: if(isnan(covar[Tvar[k]][i])){
5815: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5816: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5817: fflush(ficlog);
5818: exit(1);
5819: }
5820: }
5821: }
1.287 brouard 5822: } /* 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 5823:
5824: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5825: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5826: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5827: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5828: 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 */
5829: 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 */
5830: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5831: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5832:
5833: ij=0;
5834: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5835: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5836: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5837: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5838: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5839: /* If product not in single variable we don't print results */
5840: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5841: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5842: 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*/
5843: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5844: 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 */
5845: if(Fixed[k]!=0)
5846: anyvaryingduminmodel=1;
5847: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5848: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5849: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5850: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5851: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5852: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5853: }
5854: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5855: /* ij--; */
5856: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5857: *cptcov=ij; /*Number of total real effective covariates: effective
5858: * because they can be excluded from the model and real
5859: * if in the model but excluded because missing values, but how to get k from ij?*/
5860: for(j=ij+1; j<= cptcovt; j++){
5861: Tvaraff[j]=0;
5862: Tmodelind[j]=0;
5863: }
5864: for(j=ntveff+1; j<= cptcovt; j++){
5865: TmodelInvind[j]=0;
5866: }
5867: /* To be sorted */
5868: ;
5869: }
1.126 brouard 5870:
1.145 brouard 5871:
1.126 brouard 5872: /*********** Health Expectancies ****************/
5873:
1.235 brouard 5874: 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 5875:
5876: {
5877: /* Health expectancies, no variances */
1.164 brouard 5878: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5879: int nhstepma, nstepma; /* Decreasing with age */
5880: double age, agelim, hf;
5881: double ***p3mat;
5882: double eip;
5883:
1.238 brouard 5884: /* pstamp(ficreseij); */
1.126 brouard 5885: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5886: fprintf(ficreseij,"# Age");
5887: for(i=1; i<=nlstate;i++){
5888: for(j=1; j<=nlstate;j++){
5889: fprintf(ficreseij," e%1d%1d ",i,j);
5890: }
5891: fprintf(ficreseij," e%1d. ",i);
5892: }
5893: fprintf(ficreseij,"\n");
5894:
5895:
5896: if(estepm < stepm){
5897: printf ("Problem %d lower than %d\n",estepm, stepm);
5898: }
5899: else hstepm=estepm;
5900: /* We compute the life expectancy from trapezoids spaced every estepm months
5901: * This is mainly to measure the difference between two models: for example
5902: * if stepm=24 months pijx are given only every 2 years and by summing them
5903: * we are calculating an estimate of the Life Expectancy assuming a linear
5904: * progression in between and thus overestimating or underestimating according
5905: * to the curvature of the survival function. If, for the same date, we
5906: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5907: * to compare the new estimate of Life expectancy with the same linear
5908: * hypothesis. A more precise result, taking into account a more precise
5909: * curvature will be obtained if estepm is as small as stepm. */
5910:
5911: /* For example we decided to compute the life expectancy with the smallest unit */
5912: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5913: nhstepm is the number of hstepm from age to agelim
5914: nstepm is the number of stepm from age to agelin.
1.270 brouard 5915: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5916: and note for a fixed period like estepm months */
5917: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5918: survival function given by stepm (the optimization length). Unfortunately it
5919: means that if the survival funtion is printed only each two years of age and if
5920: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5921: results. So we changed our mind and took the option of the best precision.
5922: */
5923: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5924:
5925: agelim=AGESUP;
5926: /* If stepm=6 months */
5927: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5928: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5929:
5930: /* nhstepm age range expressed in number of stepm */
5931: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5932: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5933: /* if (stepm >= YEARM) hstepm=1;*/
5934: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5935: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5936:
5937: for (age=bage; age<=fage; age ++){
5938: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5939: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5940: /* if (stepm >= YEARM) hstepm=1;*/
5941: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5942:
5943: /* If stepm=6 months */
5944: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5945: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5946:
1.235 brouard 5947: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5948:
5949: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5950:
5951: printf("%d|",(int)age);fflush(stdout);
5952: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5953:
5954: /* Computing expectancies */
5955: for(i=1; i<=nlstate;i++)
5956: for(j=1; j<=nlstate;j++)
5957: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5958: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5959:
5960: /* 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]);*/
5961:
5962: }
5963:
5964: fprintf(ficreseij,"%3.0f",age );
5965: for(i=1; i<=nlstate;i++){
5966: eip=0;
5967: for(j=1; j<=nlstate;j++){
5968: eip +=eij[i][j][(int)age];
5969: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5970: }
5971: fprintf(ficreseij,"%9.4f", eip );
5972: }
5973: fprintf(ficreseij,"\n");
5974:
5975: }
5976: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5977: printf("\n");
5978: fprintf(ficlog,"\n");
5979:
5980: }
5981:
1.235 brouard 5982: 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 5983:
5984: {
5985: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5986: to initial status i, ei. .
1.126 brouard 5987: */
5988: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5989: int nhstepma, nstepma; /* Decreasing with age */
5990: double age, agelim, hf;
5991: double ***p3matp, ***p3matm, ***varhe;
5992: double **dnewm,**doldm;
5993: double *xp, *xm;
5994: double **gp, **gm;
5995: double ***gradg, ***trgradg;
5996: int theta;
5997:
5998: double eip, vip;
5999:
6000: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6001: xp=vector(1,npar);
6002: xm=vector(1,npar);
6003: dnewm=matrix(1,nlstate*nlstate,1,npar);
6004: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6005:
6006: pstamp(ficresstdeij);
6007: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6008: fprintf(ficresstdeij,"# Age");
6009: for(i=1; i<=nlstate;i++){
6010: for(j=1; j<=nlstate;j++)
6011: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6012: fprintf(ficresstdeij," e%1d. ",i);
6013: }
6014: fprintf(ficresstdeij,"\n");
6015:
6016: pstamp(ficrescveij);
6017: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6018: fprintf(ficrescveij,"# Age");
6019: for(i=1; i<=nlstate;i++)
6020: for(j=1; j<=nlstate;j++){
6021: cptj= (j-1)*nlstate+i;
6022: for(i2=1; i2<=nlstate;i2++)
6023: for(j2=1; j2<=nlstate;j2++){
6024: cptj2= (j2-1)*nlstate+i2;
6025: if(cptj2 <= cptj)
6026: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6027: }
6028: }
6029: fprintf(ficrescveij,"\n");
6030:
6031: if(estepm < stepm){
6032: printf ("Problem %d lower than %d\n",estepm, stepm);
6033: }
6034: else hstepm=estepm;
6035: /* We compute the life expectancy from trapezoids spaced every estepm months
6036: * This is mainly to measure the difference between two models: for example
6037: * if stepm=24 months pijx are given only every 2 years and by summing them
6038: * we are calculating an estimate of the Life Expectancy assuming a linear
6039: * progression in between and thus overestimating or underestimating according
6040: * to the curvature of the survival function. If, for the same date, we
6041: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6042: * to compare the new estimate of Life expectancy with the same linear
6043: * hypothesis. A more precise result, taking into account a more precise
6044: * curvature will be obtained if estepm is as small as stepm. */
6045:
6046: /* For example we decided to compute the life expectancy with the smallest unit */
6047: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6048: nhstepm is the number of hstepm from age to agelim
6049: nstepm is the number of stepm from age to agelin.
6050: Look at hpijx to understand the reason of that which relies in memory size
6051: and note for a fixed period like estepm months */
6052: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6053: survival function given by stepm (the optimization length). Unfortunately it
6054: means that if the survival funtion is printed only each two years of age and if
6055: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6056: results. So we changed our mind and took the option of the best precision.
6057: */
6058: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6059:
6060: /* If stepm=6 months */
6061: /* nhstepm age range expressed in number of stepm */
6062: agelim=AGESUP;
6063: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6064: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6065: /* if (stepm >= YEARM) hstepm=1;*/
6066: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6067:
6068: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6069: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6070: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6071: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6072: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6073: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6074:
6075: for (age=bage; age<=fage; age ++){
6076: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6077: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6078: /* if (stepm >= YEARM) hstepm=1;*/
6079: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6080:
1.126 brouard 6081: /* If stepm=6 months */
6082: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6083: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6084:
6085: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6086:
1.126 brouard 6087: /* Computing Variances of health expectancies */
6088: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6089: decrease memory allocation */
6090: for(theta=1; theta <=npar; theta++){
6091: for(i=1; i<=npar; i++){
1.222 brouard 6092: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6093: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6094: }
1.235 brouard 6095: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6096: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6097:
1.126 brouard 6098: for(j=1; j<= nlstate; j++){
1.222 brouard 6099: for(i=1; i<=nlstate; i++){
6100: for(h=0; h<=nhstepm-1; h++){
6101: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6102: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6103: }
6104: }
1.126 brouard 6105: }
1.218 brouard 6106:
1.126 brouard 6107: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6108: for(h=0; h<=nhstepm-1; h++){
6109: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6110: }
1.126 brouard 6111: }/* End theta */
6112:
6113:
6114: for(h=0; h<=nhstepm-1; h++)
6115: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6116: for(theta=1; theta <=npar; theta++)
6117: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6118:
1.218 brouard 6119:
1.222 brouard 6120: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6121: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6122: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6123:
1.222 brouard 6124: printf("%d|",(int)age);fflush(stdout);
6125: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6126: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6127: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6128: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6129: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6130: for(ij=1;ij<=nlstate*nlstate;ij++)
6131: for(ji=1;ji<=nlstate*nlstate;ji++)
6132: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6133: }
6134: }
1.320 brouard 6135: /* if((int)age ==50){ */
6136: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6137: /* } */
1.126 brouard 6138: /* Computing expectancies */
1.235 brouard 6139: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6140: for(i=1; i<=nlstate;i++)
6141: for(j=1; j<=nlstate;j++)
1.222 brouard 6142: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6143: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6144:
1.222 brouard 6145: /* 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 6146:
1.222 brouard 6147: }
1.269 brouard 6148:
6149: /* Standard deviation of expectancies ij */
1.126 brouard 6150: fprintf(ficresstdeij,"%3.0f",age );
6151: for(i=1; i<=nlstate;i++){
6152: eip=0.;
6153: vip=0.;
6154: for(j=1; j<=nlstate;j++){
1.222 brouard 6155: eip += eij[i][j][(int)age];
6156: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6157: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6158: 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 6159: }
6160: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6161: }
6162: fprintf(ficresstdeij,"\n");
1.218 brouard 6163:
1.269 brouard 6164: /* Variance of expectancies ij */
1.126 brouard 6165: fprintf(ficrescveij,"%3.0f",age );
6166: for(i=1; i<=nlstate;i++)
6167: for(j=1; j<=nlstate;j++){
1.222 brouard 6168: cptj= (j-1)*nlstate+i;
6169: for(i2=1; i2<=nlstate;i2++)
6170: for(j2=1; j2<=nlstate;j2++){
6171: cptj2= (j2-1)*nlstate+i2;
6172: if(cptj2 <= cptj)
6173: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6174: }
1.126 brouard 6175: }
6176: fprintf(ficrescveij,"\n");
1.218 brouard 6177:
1.126 brouard 6178: }
6179: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6180: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6181: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6182: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6183: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6184: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6185: printf("\n");
6186: fprintf(ficlog,"\n");
1.218 brouard 6187:
1.126 brouard 6188: free_vector(xm,1,npar);
6189: free_vector(xp,1,npar);
6190: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6191: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6192: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6193: }
1.218 brouard 6194:
1.126 brouard 6195: /************ Variance ******************/
1.235 brouard 6196: 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 6197: {
1.279 brouard 6198: /** Variance of health expectancies
6199: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6200: * double **newm;
6201: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6202: */
1.218 brouard 6203:
6204: /* int movingaverage(); */
6205: double **dnewm,**doldm;
6206: double **dnewmp,**doldmp;
6207: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6208: int first=0;
1.218 brouard 6209: int k;
6210: double *xp;
1.279 brouard 6211: double **gp, **gm; /**< for var eij */
6212: double ***gradg, ***trgradg; /**< for var eij */
6213: double **gradgp, **trgradgp; /**< for var p point j */
6214: double *gpp, *gmp; /**< for var p point j */
6215: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6216: double ***p3mat;
6217: double age,agelim, hf;
6218: /* double ***mobaverage; */
6219: int theta;
6220: char digit[4];
6221: char digitp[25];
6222:
6223: char fileresprobmorprev[FILENAMELENGTH];
6224:
6225: if(popbased==1){
6226: if(mobilav!=0)
6227: strcpy(digitp,"-POPULBASED-MOBILAV_");
6228: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6229: }
6230: else
6231: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6232:
1.218 brouard 6233: /* if (mobilav!=0) { */
6234: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6235: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6236: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6237: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6238: /* } */
6239: /* } */
6240:
6241: strcpy(fileresprobmorprev,"PRMORPREV-");
6242: sprintf(digit,"%-d",ij);
6243: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6244: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6245: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6246: strcat(fileresprobmorprev,fileresu);
6247: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6248: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6249: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6250: }
6251: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6252: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6253: pstamp(ficresprobmorprev);
6254: 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 6255: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
6256: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
6257: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
6258: }
6259: for(j=1;j<=cptcoveff;j++)
6260: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
6261: fprintf(ficresprobmorprev,"\n");
6262:
1.218 brouard 6263: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6264: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6265: fprintf(ficresprobmorprev," p.%-d SE",j);
6266: for(i=1; i<=nlstate;i++)
6267: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6268: }
6269: fprintf(ficresprobmorprev,"\n");
6270:
6271: fprintf(ficgp,"\n# Routine varevsij");
6272: fprintf(ficgp,"\nunset title \n");
6273: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6274: 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");
6275: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6276:
1.218 brouard 6277: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6278: pstamp(ficresvij);
6279: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6280: if(popbased==1)
6281: 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);
6282: else
6283: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6284: fprintf(ficresvij,"# Age");
6285: for(i=1; i<=nlstate;i++)
6286: for(j=1; j<=nlstate;j++)
6287: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6288: fprintf(ficresvij,"\n");
6289:
6290: xp=vector(1,npar);
6291: dnewm=matrix(1,nlstate,1,npar);
6292: doldm=matrix(1,nlstate,1,nlstate);
6293: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6294: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6295:
6296: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6297: gpp=vector(nlstate+1,nlstate+ndeath);
6298: gmp=vector(nlstate+1,nlstate+ndeath);
6299: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6300:
1.218 brouard 6301: if(estepm < stepm){
6302: printf ("Problem %d lower than %d\n",estepm, stepm);
6303: }
6304: else hstepm=estepm;
6305: /* For example we decided to compute the life expectancy with the smallest unit */
6306: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6307: nhstepm is the number of hstepm from age to agelim
6308: nstepm is the number of stepm from age to agelim.
6309: Look at function hpijx to understand why because of memory size limitations,
6310: we decided (b) to get a life expectancy respecting the most precise curvature of the
6311: survival function given by stepm (the optimization length). Unfortunately it
6312: means that if the survival funtion is printed every two years of age and if
6313: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6314: results. So we changed our mind and took the option of the best precision.
6315: */
6316: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6317: agelim = AGESUP;
6318: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6319: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6320: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6321: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6322: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6323: gp=matrix(0,nhstepm,1,nlstate);
6324: gm=matrix(0,nhstepm,1,nlstate);
6325:
6326:
6327: for(theta=1; theta <=npar; theta++){
6328: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6329: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6330: }
1.279 brouard 6331: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6332: * returns into prlim .
1.288 brouard 6333: */
1.242 brouard 6334: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6335:
6336: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6337: if (popbased==1) {
6338: if(mobilav ==0){
6339: for(i=1; i<=nlstate;i++)
6340: prlim[i][i]=probs[(int)age][i][ij];
6341: }else{ /* mobilav */
6342: for(i=1; i<=nlstate;i++)
6343: prlim[i][i]=mobaverage[(int)age][i][ij];
6344: }
6345: }
1.295 brouard 6346: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6347: */
6348: 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 6349: /**< 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 6350: * at horizon h in state j including mortality.
6351: */
1.218 brouard 6352: for(j=1; j<= nlstate; j++){
6353: for(h=0; h<=nhstepm; h++){
6354: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6355: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6356: }
6357: }
1.279 brouard 6358: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6359: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6360: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6361: */
6362: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6363: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6364: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6365: }
6366:
6367: /* Again with minus shift */
1.218 brouard 6368:
6369: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6370: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6371:
1.242 brouard 6372: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6373:
6374: if (popbased==1) {
6375: if(mobilav ==0){
6376: for(i=1; i<=nlstate;i++)
6377: prlim[i][i]=probs[(int)age][i][ij];
6378: }else{ /* mobilav */
6379: for(i=1; i<=nlstate;i++)
6380: prlim[i][i]=mobaverage[(int)age][i][ij];
6381: }
6382: }
6383:
1.235 brouard 6384: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6385:
6386: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6387: for(h=0; h<=nhstepm; h++){
6388: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6389: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6390: }
6391: }
6392: /* This for computing probability of death (h=1 means
6393: computed over hstepm matrices product = hstepm*stepm months)
6394: as a weighted average of prlim.
6395: */
6396: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6397: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6398: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6399: }
1.279 brouard 6400: /* end shifting computations */
6401:
6402: /**< Computing gradient matrix at horizon h
6403: */
1.218 brouard 6404: for(j=1; j<= nlstate; j++) /* vareij */
6405: for(h=0; h<=nhstepm; h++){
6406: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6407: }
1.279 brouard 6408: /**< Gradient of overall mortality p.3 (or p.j)
6409: */
6410: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6411: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6412: }
6413:
6414: } /* End theta */
1.279 brouard 6415:
6416: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6417: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6418:
6419: for(h=0; h<=nhstepm; h++) /* veij */
6420: for(j=1; j<=nlstate;j++)
6421: for(theta=1; theta <=npar; theta++)
6422: trgradg[h][j][theta]=gradg[h][theta][j];
6423:
6424: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6425: for(theta=1; theta <=npar; theta++)
6426: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6427: /**< as well as its transposed matrix
6428: */
1.218 brouard 6429:
6430: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6431: for(i=1;i<=nlstate;i++)
6432: for(j=1;j<=nlstate;j++)
6433: vareij[i][j][(int)age] =0.;
1.279 brouard 6434:
6435: /* Computing trgradg by matcov by gradg at age and summing over h
6436: * and k (nhstepm) formula 15 of article
6437: * Lievre-Brouard-Heathcote
6438: */
6439:
1.218 brouard 6440: for(h=0;h<=nhstepm;h++){
6441: for(k=0;k<=nhstepm;k++){
6442: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6443: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6444: for(i=1;i<=nlstate;i++)
6445: for(j=1;j<=nlstate;j++)
6446: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6447: }
6448: }
6449:
1.279 brouard 6450: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6451: * p.j overall mortality formula 49 but computed directly because
6452: * we compute the grad (wix pijx) instead of grad (pijx),even if
6453: * wix is independent of theta.
6454: */
1.218 brouard 6455: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6456: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6457: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6458: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6459: varppt[j][i]=doldmp[j][i];
6460: /* end ppptj */
6461: /* x centered again */
6462:
1.242 brouard 6463: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6464:
6465: if (popbased==1) {
6466: if(mobilav ==0){
6467: for(i=1; i<=nlstate;i++)
6468: prlim[i][i]=probs[(int)age][i][ij];
6469: }else{ /* mobilav */
6470: for(i=1; i<=nlstate;i++)
6471: prlim[i][i]=mobaverage[(int)age][i][ij];
6472: }
6473: }
6474:
6475: /* This for computing probability of death (h=1 means
6476: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6477: as a weighted average of prlim.
6478: */
1.235 brouard 6479: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6480: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6481: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6482: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6483: }
6484: /* end probability of death */
6485:
6486: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6487: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6488: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6489: for(i=1; i<=nlstate;i++){
6490: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6491: }
6492: }
6493: fprintf(ficresprobmorprev,"\n");
6494:
6495: fprintf(ficresvij,"%.0f ",age );
6496: for(i=1; i<=nlstate;i++)
6497: for(j=1; j<=nlstate;j++){
6498: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6499: }
6500: fprintf(ficresvij,"\n");
6501: free_matrix(gp,0,nhstepm,1,nlstate);
6502: free_matrix(gm,0,nhstepm,1,nlstate);
6503: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6504: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6505: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6506: } /* End age */
6507: free_vector(gpp,nlstate+1,nlstate+ndeath);
6508: free_vector(gmp,nlstate+1,nlstate+ndeath);
6509: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6510: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6511: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6512: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6513: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6514: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6515: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6516: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6517: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6518: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6519: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6520: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6521: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6522: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6523: 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);
6524: /* 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 6525: */
1.218 brouard 6526: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6527: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6528:
1.218 brouard 6529: free_vector(xp,1,npar);
6530: free_matrix(doldm,1,nlstate,1,nlstate);
6531: free_matrix(dnewm,1,nlstate,1,npar);
6532: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6533: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6534: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6535: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6536: fclose(ficresprobmorprev);
6537: fflush(ficgp);
6538: fflush(fichtm);
6539: } /* end varevsij */
1.126 brouard 6540:
6541: /************ Variance of prevlim ******************/
1.269 brouard 6542: 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 6543: {
1.205 brouard 6544: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6545: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6546:
1.268 brouard 6547: double **dnewmpar,**doldm;
1.126 brouard 6548: int i, j, nhstepm, hstepm;
6549: double *xp;
6550: double *gp, *gm;
6551: double **gradg, **trgradg;
1.208 brouard 6552: double **mgm, **mgp;
1.126 brouard 6553: double age,agelim;
6554: int theta;
6555:
6556: pstamp(ficresvpl);
1.288 brouard 6557: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6558: fprintf(ficresvpl,"# Age ");
6559: if(nresult >=1)
6560: fprintf(ficresvpl," Result# ");
1.126 brouard 6561: for(i=1; i<=nlstate;i++)
6562: fprintf(ficresvpl," %1d-%1d",i,i);
6563: fprintf(ficresvpl,"\n");
6564:
6565: xp=vector(1,npar);
1.268 brouard 6566: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6567: doldm=matrix(1,nlstate,1,nlstate);
6568:
6569: hstepm=1*YEARM; /* Every year of age */
6570: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6571: agelim = AGESUP;
6572: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6573: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6574: if (stepm >= YEARM) hstepm=1;
6575: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6576: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6577: mgp=matrix(1,npar,1,nlstate);
6578: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6579: gp=vector(1,nlstate);
6580: gm=vector(1,nlstate);
6581:
6582: for(theta=1; theta <=npar; theta++){
6583: for(i=1; i<=npar; i++){ /* Computes gradient */
6584: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6585: }
1.288 brouard 6586: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6587: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6588: /* else */
6589: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6590: for(i=1;i<=nlstate;i++){
1.126 brouard 6591: gp[i] = prlim[i][i];
1.208 brouard 6592: mgp[theta][i] = prlim[i][i];
6593: }
1.126 brouard 6594: for(i=1; i<=npar; i++) /* Computes gradient */
6595: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6596: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6597: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6598: /* else */
6599: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6600: for(i=1;i<=nlstate;i++){
1.126 brouard 6601: gm[i] = prlim[i][i];
1.208 brouard 6602: mgm[theta][i] = prlim[i][i];
6603: }
1.126 brouard 6604: for(i=1;i<=nlstate;i++)
6605: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6606: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6607: } /* End theta */
6608:
6609: trgradg =matrix(1,nlstate,1,npar);
6610:
6611: for(j=1; j<=nlstate;j++)
6612: for(theta=1; theta <=npar; theta++)
6613: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6614: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6615: /* printf("\nmgm mgp %d ",(int)age); */
6616: /* for(j=1; j<=nlstate;j++){ */
6617: /* printf(" %d ",j); */
6618: /* for(theta=1; theta <=npar; theta++) */
6619: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6620: /* printf("\n "); */
6621: /* } */
6622: /* } */
6623: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6624: /* printf("\n gradg %d ",(int)age); */
6625: /* for(j=1; j<=nlstate;j++){ */
6626: /* printf("%d ",j); */
6627: /* for(theta=1; theta <=npar; theta++) */
6628: /* printf("%d %lf ",theta,gradg[theta][j]); */
6629: /* printf("\n "); */
6630: /* } */
6631: /* } */
1.126 brouard 6632:
6633: for(i=1;i<=nlstate;i++)
6634: varpl[i][(int)age] =0.;
1.209 brouard 6635: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6636: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6637: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6638: }else{
1.268 brouard 6639: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6640: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6641: }
1.126 brouard 6642: for(i=1;i<=nlstate;i++)
6643: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6644:
6645: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6646: if(nresult >=1)
6647: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6648: for(i=1; i<=nlstate;i++){
1.126 brouard 6649: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6650: /* for(j=1;j<=nlstate;j++) */
6651: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6652: }
1.126 brouard 6653: fprintf(ficresvpl,"\n");
6654: free_vector(gp,1,nlstate);
6655: free_vector(gm,1,nlstate);
1.208 brouard 6656: free_matrix(mgm,1,npar,1,nlstate);
6657: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6658: free_matrix(gradg,1,npar,1,nlstate);
6659: free_matrix(trgradg,1,nlstate,1,npar);
6660: } /* End age */
6661:
6662: free_vector(xp,1,npar);
6663: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6664: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6665:
6666: }
6667:
6668:
6669: /************ Variance of backprevalence limit ******************/
1.269 brouard 6670: 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 6671: {
6672: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6673: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6674:
6675: double **dnewmpar,**doldm;
6676: int i, j, nhstepm, hstepm;
6677: double *xp;
6678: double *gp, *gm;
6679: double **gradg, **trgradg;
6680: double **mgm, **mgp;
6681: double age,agelim;
6682: int theta;
6683:
6684: pstamp(ficresvbl);
6685: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6686: fprintf(ficresvbl,"# Age ");
6687: if(nresult >=1)
6688: fprintf(ficresvbl," Result# ");
6689: for(i=1; i<=nlstate;i++)
6690: fprintf(ficresvbl," %1d-%1d",i,i);
6691: fprintf(ficresvbl,"\n");
6692:
6693: xp=vector(1,npar);
6694: dnewmpar=matrix(1,nlstate,1,npar);
6695: doldm=matrix(1,nlstate,1,nlstate);
6696:
6697: hstepm=1*YEARM; /* Every year of age */
6698: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6699: agelim = AGEINF;
6700: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6701: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6702: if (stepm >= YEARM) hstepm=1;
6703: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6704: gradg=matrix(1,npar,1,nlstate);
6705: mgp=matrix(1,npar,1,nlstate);
6706: mgm=matrix(1,npar,1,nlstate);
6707: gp=vector(1,nlstate);
6708: gm=vector(1,nlstate);
6709:
6710: for(theta=1; theta <=npar; theta++){
6711: for(i=1; i<=npar; i++){ /* Computes gradient */
6712: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6713: }
6714: if(mobilavproj > 0 )
6715: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6716: else
6717: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6718: for(i=1;i<=nlstate;i++){
6719: gp[i] = bprlim[i][i];
6720: mgp[theta][i] = bprlim[i][i];
6721: }
6722: for(i=1; i<=npar; i++) /* Computes gradient */
6723: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6724: if(mobilavproj > 0 )
6725: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6726: else
6727: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6728: for(i=1;i<=nlstate;i++){
6729: gm[i] = bprlim[i][i];
6730: mgm[theta][i] = bprlim[i][i];
6731: }
6732: for(i=1;i<=nlstate;i++)
6733: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6734: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6735: } /* End theta */
6736:
6737: trgradg =matrix(1,nlstate,1,npar);
6738:
6739: for(j=1; j<=nlstate;j++)
6740: for(theta=1; theta <=npar; theta++)
6741: trgradg[j][theta]=gradg[theta][j];
6742: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6743: /* printf("\nmgm mgp %d ",(int)age); */
6744: /* for(j=1; j<=nlstate;j++){ */
6745: /* printf(" %d ",j); */
6746: /* for(theta=1; theta <=npar; theta++) */
6747: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6748: /* printf("\n "); */
6749: /* } */
6750: /* } */
6751: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6752: /* printf("\n gradg %d ",(int)age); */
6753: /* for(j=1; j<=nlstate;j++){ */
6754: /* printf("%d ",j); */
6755: /* for(theta=1; theta <=npar; theta++) */
6756: /* printf("%d %lf ",theta,gradg[theta][j]); */
6757: /* printf("\n "); */
6758: /* } */
6759: /* } */
6760:
6761: for(i=1;i<=nlstate;i++)
6762: varbpl[i][(int)age] =0.;
6763: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6764: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6765: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6766: }else{
6767: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6768: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6769: }
6770: for(i=1;i<=nlstate;i++)
6771: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6772:
6773: fprintf(ficresvbl,"%.0f ",age );
6774: if(nresult >=1)
6775: fprintf(ficresvbl,"%d ",nres );
6776: for(i=1; i<=nlstate;i++)
6777: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6778: fprintf(ficresvbl,"\n");
6779: free_vector(gp,1,nlstate);
6780: free_vector(gm,1,nlstate);
6781: free_matrix(mgm,1,npar,1,nlstate);
6782: free_matrix(mgp,1,npar,1,nlstate);
6783: free_matrix(gradg,1,npar,1,nlstate);
6784: free_matrix(trgradg,1,nlstate,1,npar);
6785: } /* End age */
6786:
6787: free_vector(xp,1,npar);
6788: free_matrix(doldm,1,nlstate,1,npar);
6789: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6790:
6791: }
6792:
6793: /************ Variance of one-step probabilities ******************/
6794: 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 6795: {
6796: int i, j=0, k1, l1, tj;
6797: int k2, l2, j1, z1;
6798: int k=0, l;
6799: int first=1, first1, first2;
6800: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6801: double **dnewm,**doldm;
6802: double *xp;
6803: double *gp, *gm;
6804: double **gradg, **trgradg;
6805: double **mu;
6806: double age, cov[NCOVMAX+1];
6807: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6808: int theta;
6809: char fileresprob[FILENAMELENGTH];
6810: char fileresprobcov[FILENAMELENGTH];
6811: char fileresprobcor[FILENAMELENGTH];
6812: double ***varpij;
6813:
6814: strcpy(fileresprob,"PROB_");
6815: strcat(fileresprob,fileres);
6816: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6817: printf("Problem with resultfile: %s\n", fileresprob);
6818: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6819: }
6820: strcpy(fileresprobcov,"PROBCOV_");
6821: strcat(fileresprobcov,fileresu);
6822: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6823: printf("Problem with resultfile: %s\n", fileresprobcov);
6824: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6825: }
6826: strcpy(fileresprobcor,"PROBCOR_");
6827: strcat(fileresprobcor,fileresu);
6828: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6829: printf("Problem with resultfile: %s\n", fileresprobcor);
6830: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6831: }
6832: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6833: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6834: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6835: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6836: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6837: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6838: pstamp(ficresprob);
6839: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6840: fprintf(ficresprob,"# Age");
6841: pstamp(ficresprobcov);
6842: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6843: fprintf(ficresprobcov,"# Age");
6844: pstamp(ficresprobcor);
6845: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6846: fprintf(ficresprobcor,"# Age");
1.126 brouard 6847:
6848:
1.222 brouard 6849: for(i=1; i<=nlstate;i++)
6850: for(j=1; j<=(nlstate+ndeath);j++){
6851: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6852: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6853: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6854: }
6855: /* fprintf(ficresprob,"\n");
6856: fprintf(ficresprobcov,"\n");
6857: fprintf(ficresprobcor,"\n");
6858: */
6859: xp=vector(1,npar);
6860: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6861: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6862: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6863: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6864: first=1;
6865: fprintf(ficgp,"\n# Routine varprob");
6866: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6867: fprintf(fichtm,"\n");
6868:
1.288 brouard 6869: 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 6870: 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);
6871: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6872: and drawn. It helps understanding how is the covariance between two incidences.\
6873: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6874: 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 6875: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6876: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6877: standard deviations wide on each axis. <br>\
6878: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6879: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6880: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6881:
1.222 brouard 6882: cov[1]=1;
6883: /* tj=cptcoveff; */
1.225 brouard 6884: tj = (int) pow(2,cptcoveff);
1.222 brouard 6885: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6886: j1=0;
1.224 brouard 6887: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6888: if (cptcovn>0) {
6889: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6890: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6891: fprintf(ficresprob, "**********\n#\n");
6892: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6893: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6894: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6895:
1.222 brouard 6896: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6897: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6898: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6899:
6900:
1.222 brouard 6901: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.319 brouard 6902: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]); */
6903: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtmcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6904: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6905:
1.222 brouard 6906: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6907: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6908: fprintf(ficresprobcor, "**********\n#");
6909: if(invalidvarcomb[j1]){
6910: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6911: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6912: continue;
6913: }
6914: }
6915: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6916: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6917: gp=vector(1,(nlstate)*(nlstate+ndeath));
6918: gm=vector(1,(nlstate)*(nlstate+ndeath));
6919: for (age=bage; age<=fage; age ++){
6920: cov[2]=age;
6921: if(nagesqr==1)
6922: cov[3]= age*age;
6923: for (k=1; k<=cptcovn;k++) {
6924: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6925: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6926: * 1 1 1 1 1
6927: * 2 2 1 1 1
6928: * 3 1 2 1 1
6929: */
6930: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6931: }
1.319 brouard 6932: /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 */
6933: /* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] */
6934: /*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6935: for (k=1; k<=cptcovage;k++)
6936: cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.222 brouard 6937: for (k=1; k<=cptcovprod;k++)
6938: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6939:
6940:
1.222 brouard 6941: for(theta=1; theta <=npar; theta++){
6942: for(i=1; i<=npar; i++)
6943: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6944:
1.222 brouard 6945: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6946:
1.222 brouard 6947: k=0;
6948: for(i=1; i<= (nlstate); i++){
6949: for(j=1; j<=(nlstate+ndeath);j++){
6950: k=k+1;
6951: gp[k]=pmmij[i][j];
6952: }
6953: }
1.220 brouard 6954:
1.222 brouard 6955: for(i=1; i<=npar; i++)
6956: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6957:
1.222 brouard 6958: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6959: k=0;
6960: for(i=1; i<=(nlstate); i++){
6961: for(j=1; j<=(nlstate+ndeath);j++){
6962: k=k+1;
6963: gm[k]=pmmij[i][j];
6964: }
6965: }
1.220 brouard 6966:
1.222 brouard 6967: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6968: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6969: }
1.126 brouard 6970:
1.222 brouard 6971: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6972: for(theta=1; theta <=npar; theta++)
6973: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6974:
1.222 brouard 6975: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6976: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6977:
1.222 brouard 6978: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6979:
1.222 brouard 6980: k=0;
6981: for(i=1; i<=(nlstate); i++){
6982: for(j=1; j<=(nlstate+ndeath);j++){
6983: k=k+1;
6984: mu[k][(int) age]=pmmij[i][j];
6985: }
6986: }
6987: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6988: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6989: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6990:
1.222 brouard 6991: /*printf("\n%d ",(int)age);
6992: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6993: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6994: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6995: }*/
1.220 brouard 6996:
1.222 brouard 6997: fprintf(ficresprob,"\n%d ",(int)age);
6998: fprintf(ficresprobcov,"\n%d ",(int)age);
6999: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7000:
1.222 brouard 7001: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7002: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7003: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7004: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7005: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7006: }
7007: i=0;
7008: for (k=1; k<=(nlstate);k++){
7009: for (l=1; l<=(nlstate+ndeath);l++){
7010: i++;
7011: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7012: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7013: for (j=1; j<=i;j++){
7014: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7015: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7016: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7017: }
7018: }
7019: }/* end of loop for state */
7020: } /* end of loop for age */
7021: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7022: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7023: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7024: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7025:
7026: /* Confidence intervalle of pij */
7027: /*
7028: fprintf(ficgp,"\nunset parametric;unset label");
7029: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7030: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7031: 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);
7032: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7033: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7034: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7035: */
7036:
7037: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7038: first1=1;first2=2;
7039: for (k2=1; k2<=(nlstate);k2++){
7040: for (l2=1; l2<=(nlstate+ndeath);l2++){
7041: if(l2==k2) continue;
7042: j=(k2-1)*(nlstate+ndeath)+l2;
7043: for (k1=1; k1<=(nlstate);k1++){
7044: for (l1=1; l1<=(nlstate+ndeath);l1++){
7045: if(l1==k1) continue;
7046: i=(k1-1)*(nlstate+ndeath)+l1;
7047: if(i<=j) continue;
7048: for (age=bage; age<=fage; age ++){
7049: if ((int)age %5==0){
7050: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7051: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7052: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7053: mu1=mu[i][(int) age]/stepm*YEARM ;
7054: mu2=mu[j][(int) age]/stepm*YEARM;
7055: c12=cv12/sqrt(v1*v2);
7056: /* Computing eigen value of matrix of covariance */
7057: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7058: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7059: if ((lc2 <0) || (lc1 <0) ){
7060: if(first2==1){
7061: first1=0;
7062: 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);
7063: }
7064: 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);
7065: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7066: /* lc2=fabs(lc2); */
7067: }
1.220 brouard 7068:
1.222 brouard 7069: /* Eigen vectors */
1.280 brouard 7070: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7071: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7072: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7073: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7074: }else
7075: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7076: /*v21=sqrt(1.-v11*v11); *//* error */
7077: v21=(lc1-v1)/cv12*v11;
7078: v12=-v21;
7079: v22=v11;
7080: tnalp=v21/v11;
7081: if(first1==1){
7082: first1=0;
7083: 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);
7084: }
7085: 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);
7086: /*printf(fignu*/
7087: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7088: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7089: if(first==1){
7090: first=0;
7091: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7092: fprintf(ficgp,"\nset parametric;unset label");
7093: 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);
7094: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7095: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7096: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7097: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7098: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7099: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7100: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7101: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7102: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7103: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7104: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7105: 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 7106: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7107: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7108: }else{
7109: first=0;
7110: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7111: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7112: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7113: 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 7114: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7115: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7116: }/* if first */
7117: } /* age mod 5 */
7118: } /* end loop age */
7119: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7120: first=1;
7121: } /*l12 */
7122: } /* k12 */
7123: } /*l1 */
7124: }/* k1 */
7125: } /* loop on combination of covariates j1 */
7126: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7127: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7128: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7129: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7130: free_vector(xp,1,npar);
7131: fclose(ficresprob);
7132: fclose(ficresprobcov);
7133: fclose(ficresprobcor);
7134: fflush(ficgp);
7135: fflush(fichtmcov);
7136: }
1.126 brouard 7137:
7138:
7139: /******************* Printing html file ***********/
1.201 brouard 7140: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7141: int lastpass, int stepm, int weightopt, char model[],\
7142: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7143: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7144: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7145: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7146: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7147: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7148: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7149: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7150: </ul>");
1.319 brouard 7151: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7152: /* </ul>", model); */
1.214 brouard 7153: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7154: 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",
7155: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
7156: fprintf(fichtm,"<li> - Observed 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 7157: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7158: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7159: fprintf(fichtm,"\
7160: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7161: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7162: fprintf(fichtm,"\
1.217 brouard 7163: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7164: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7165: fprintf(fichtm,"\
1.288 brouard 7166: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7167: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7168: fprintf(fichtm,"\
1.288 brouard 7169: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7170: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7171: fprintf(fichtm,"\
1.211 brouard 7172: - (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 7173: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7174: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7175: if(prevfcast==1){
7176: fprintf(fichtm,"\
7177: - Prevalence projections by age and states: \
1.201 brouard 7178: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7179: }
1.126 brouard 7180:
7181:
1.225 brouard 7182: m=pow(2,cptcoveff);
1.222 brouard 7183: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7184:
1.317 brouard 7185: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7186:
7187: jj1=0;
7188:
7189: fprintf(fichtm," \n<ul>");
7190: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7191: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7192: if(m != 1 && TKresult[nres]!= k1)
7193: continue;
7194: jj1++;
7195: if (cptcovn > 0) {
7196: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
7197: for (cpt=1; cpt<=cptcoveff;cpt++){
7198: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7199: }
7200: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7201: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7202: }
7203: fprintf(fichtm,"\">");
7204:
7205: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7206: fprintf(fichtm,"************ Results for covariates");
7207: for (cpt=1; cpt<=cptcoveff;cpt++){
7208: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7209: }
7210: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7211: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7212: }
7213: if(invalidvarcomb[k1]){
7214: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7215: continue;
7216: }
7217: fprintf(fichtm,"</a></li>");
7218: } /* cptcovn >0 */
7219: }
1.317 brouard 7220: fprintf(fichtm," \n</ul>");
1.264 brouard 7221:
1.222 brouard 7222: jj1=0;
1.237 brouard 7223:
7224: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7225: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7226: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7227: continue;
1.220 brouard 7228:
1.222 brouard 7229: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7230: jj1++;
7231: if (cptcovn > 0) {
1.264 brouard 7232: fprintf(fichtm,"\n<p><a name=\"rescov");
7233: for (cpt=1; cpt<=cptcoveff;cpt++){
7234: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7235: }
7236: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7237: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7238: }
7239: fprintf(fichtm,"\"</a>");
7240:
1.222 brouard 7241: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7242: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7243: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7244: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7245: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7246: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7247: }
1.237 brouard 7248: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7249: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7250: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7251: }
7252:
1.230 brouard 7253: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.321 brouard 7254: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7255: if(invalidvarcomb[k1]){
7256: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7257: printf("\nCombination (%d) ignored because no cases \n",k1);
7258: continue;
7259: }
7260: }
7261: /* aij, bij */
1.259 brouard 7262: 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 7263: <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 7264: /* Pij */
1.241 brouard 7265: 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> \
7266: <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 7267: /* Quasi-incidences */
7268: 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 7269: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7270: 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 7271: 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> \
7272: <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 7273: /* Survival functions (period) in state j */
7274: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7275: 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> \
1.241 brouard 7276: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7277: }
7278: /* State specific survival functions (period) */
7279: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7280: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7281: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 7282: <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7283: }
1.288 brouard 7284: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7285: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7286: 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> \
7287: <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7288: }
1.296 brouard 7289: if(prevbcast==1){
1.288 brouard 7290: /* Backward prevalence in each health state */
1.222 brouard 7291: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7292: fprintf(fichtm,"<br>\n- Convergence to mixed (stable) back prevalence in state %d. Or probability for a person to be in state %d at a younger age, knowing that she/he was in state (1 to %d) at different older ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 7293: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"PB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres,subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.222 brouard 7294: }
1.217 brouard 7295: }
1.222 brouard 7296: if(prevfcast==1){
1.288 brouard 7297: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7298: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7299: 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);
7300: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7301: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7302: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7303: }
7304: }
1.296 brouard 7305: if(prevbcast==1){
1.268 brouard 7306: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7307: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7308: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7309: 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 \
7310: 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 7311: 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);
7312: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7313: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7314: }
7315: }
1.220 brouard 7316:
1.222 brouard 7317: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7318: 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);
7319: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7320: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7321: }
7322: /* } /\* end i1 *\/ */
7323: }/* End k1 */
7324: fprintf(fichtm,"</ul>");
1.126 brouard 7325:
1.222 brouard 7326: fprintf(fichtm,"\
1.126 brouard 7327: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7328: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7329: - 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 7330: But because parameters are usually highly correlated (a higher incidence of disability \
7331: and a higher incidence of recovery can give very close observed transition) it might \
7332: be very useful to look not only at linear confidence intervals estimated from the \
7333: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7334: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7335: covariance matrix of the one-step probabilities. \
7336: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7337:
1.222 brouard 7338: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7339: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7340: fprintf(fichtm,"\
1.126 brouard 7341: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7342: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7343:
1.222 brouard 7344: fprintf(fichtm,"\
1.126 brouard 7345: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7346: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7347: fprintf(fichtm,"\
1.126 brouard 7348: - 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): \
7349: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7350: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7351: fprintf(fichtm,"\
1.126 brouard 7352: - (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): \
7353: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7354: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7355: fprintf(fichtm,"\
1.288 brouard 7356: - 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 7357: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7358: fprintf(fichtm,"\
1.128 brouard 7359: - 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 7360: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7361: fprintf(fichtm,"\
1.288 brouard 7362: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7363: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7364:
7365: /* if(popforecast==1) fprintf(fichtm,"\n */
7366: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7367: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7368: /* <br>",fileres,fileres,fileres,fileres); */
7369: /* else */
7370: /* fprintf(fichtm,"\n No population forecast: popforecast = %d (instead of 1) or stepm = %d (instead of 1) or model=%s (instead of .)<br><br></li>\n",popforecast, stepm, model); */
1.222 brouard 7371: fflush(fichtm);
1.126 brouard 7372:
1.225 brouard 7373: m=pow(2,cptcoveff);
1.222 brouard 7374: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7375:
1.317 brouard 7376: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7377:
7378: jj1=0;
7379:
7380: fprintf(fichtm," \n<ul>");
7381: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7382: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7383: if(m != 1 && TKresult[nres]!= k1)
7384: continue;
7385: jj1++;
7386: if (cptcovn > 0) {
7387: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
7388: for (cpt=1; cpt<=cptcoveff;cpt++){
7389: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7390: }
7391: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7392: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7393: }
7394: fprintf(fichtm,"\">");
7395:
7396: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7397: fprintf(fichtm,"************ Results for covariates");
7398: for (cpt=1; cpt<=cptcoveff;cpt++){
7399: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7400: }
7401: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7402: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7403: }
7404: if(invalidvarcomb[k1]){
7405: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7406: continue;
7407: }
7408: fprintf(fichtm,"</a></li>");
7409: } /* cptcovn >0 */
7410: }
7411: fprintf(fichtm," \n</ul>");
7412:
1.222 brouard 7413: jj1=0;
1.237 brouard 7414:
1.241 brouard 7415: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7416: for(k1=1; k1<=m;k1++){
1.253 brouard 7417: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7418: continue;
1.222 brouard 7419: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7420: jj1++;
1.126 brouard 7421: if (cptcovn > 0) {
1.317 brouard 7422: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
7423: for (cpt=1; cpt<=cptcoveff;cpt++){
7424: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7425: }
7426: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7427: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7428: }
7429: fprintf(fichtm,"\"</a>");
7430:
1.126 brouard 7431: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.317 brouard 7432: for (cpt=1; cpt<=cptcoveff;cpt++){ /**< cptcoveff number of variables */
1.237 brouard 7433: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
1.317 brouard 7434: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
1.237 brouard 7435: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 7436: }
1.237 brouard 7437: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7438: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7439: }
7440:
1.321 brouard 7441: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 7442:
1.222 brouard 7443: if(invalidvarcomb[k1]){
7444: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7445: continue;
7446: }
1.126 brouard 7447: }
7448: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7449: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7450: 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);
7451: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7452: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7453: }
7454: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7455: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7456: true period expectancies (those weighted with period prevalences are also\
7457: drawn in addition to the population based expectancies computed using\
1.314 brouard 7458: 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);
7459: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7460: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7461: /* } /\* end i1 *\/ */
7462: }/* End k1 */
1.241 brouard 7463: }/* End nres */
1.222 brouard 7464: fprintf(fichtm,"</ul>");
7465: fflush(fichtm);
1.126 brouard 7466: }
7467:
7468: /******************* Gnuplot file **************/
1.296 brouard 7469: 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 7470:
7471: char dirfileres[132],optfileres[132];
1.264 brouard 7472: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7473: int cpt=0,k1=0,i=0,k=0,j=0,jk=0,k2=0,k3=0,k4=0,ij=0, ijp=0, l=0;
1.211 brouard 7474: int lv=0, vlv=0, kl=0;
1.130 brouard 7475: int ng=0;
1.201 brouard 7476: int vpopbased;
1.223 brouard 7477: int ioffset; /* variable offset for columns */
1.270 brouard 7478: int iyearc=1; /* variable column for year of projection */
7479: int iagec=1; /* variable column for age of projection */
1.235 brouard 7480: int nres=0; /* Index of resultline */
1.266 brouard 7481: int istart=1; /* For starting graphs in projections */
1.219 brouard 7482:
1.126 brouard 7483: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7484: /* printf("Problem with file %s",optionfilegnuplot); */
7485: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7486: /* } */
7487:
7488: /*#ifdef windows */
7489: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7490: /*#endif */
1.225 brouard 7491: m=pow(2,cptcoveff);
1.126 brouard 7492:
1.274 brouard 7493: /* diagram of the model */
7494: fprintf(ficgp,"\n#Diagram of the model \n");
7495: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7496: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7497: 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);
7498:
7499: fprintf(ficgp,"\n#Centripete arrows (turning in other direction (1-i) instead of (i-1)) \nset for [i=1:%d] arrow (%d+1)*10+i from cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))-(i!=j?(i-j)/abs(i-j)*delta:0), yoff +sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) rto -0.80*(cos(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d))+(i!=j?(i-j)/abs(i-j)*delta:0) ), -0.80*(sin(pi*((1-(%d/2)*2./%d)/2+(1-i)*2./%d)) + (i!=j?(i-j)/abs(i-j)*delta:0) + yoff ) ls 4\n",nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate,nlstate);
7500: fprintf(ficgp,"\n#show arrow\nunset label\n");
7501: 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);
7502: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7503: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7504: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7505: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7506:
1.202 brouard 7507: /* Contribution to likelihood */
7508: /* Plot the probability implied in the likelihood */
1.223 brouard 7509: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7510: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7511: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7512: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7513: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7514: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7515: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7516: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7517: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7518: 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));
7519: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7520: 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));
7521: for (i=1; i<= nlstate ; i ++) {
7522: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7523: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7524: 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);
7525: for (j=2; j<= nlstate+ndeath ; j ++) {
7526: 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);
7527: }
7528: fprintf(ficgp,";\nset out; unset ylabel;\n");
7529: }
7530: /* 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 */
7531: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7532: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7533: fprintf(ficgp,"\nset out;unset log\n");
7534: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7535:
1.126 brouard 7536: strcpy(dirfileres,optionfilefiname);
7537: strcpy(optfileres,"vpl");
1.223 brouard 7538: /* 1eme*/
1.238 brouard 7539: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7540: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7541: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7542: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7543: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7544: continue;
7545: /* We are interested in selected combination by the resultline */
1.246 brouard 7546: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7547: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7548: strcpy(gplotlabel,"(");
1.238 brouard 7549: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7550: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7551: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7552: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7553: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7554: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7555: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7556: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7557: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7558: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7559: }
7560: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7561: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7562: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7563: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7564: }
7565: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7566: /* printf("\n#\n"); */
1.238 brouard 7567: fprintf(ficgp,"\n#\n");
7568: if(invalidvarcomb[k1]){
1.260 brouard 7569: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7570: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7571: continue;
7572: }
1.235 brouard 7573:
1.241 brouard 7574: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7575: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7576: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
1.321 brouard 7577: fprintf(ficgp,"set title \"Alive state %d %s model=%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 7578: 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);
7579: /* 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); */
7580: /* k1-1 error should be nres-1*/
1.238 brouard 7581: for (i=1; i<= nlstate ; i ++) {
7582: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7583: else fprintf(ficgp," %%*lf (%%*lf)");
7584: }
1.288 brouard 7585: 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 7586: for (i=1; i<= nlstate ; i ++) {
7587: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7588: else fprintf(ficgp," %%*lf (%%*lf)");
7589: }
1.260 brouard 7590: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7591: for (i=1; i<= nlstate ; i ++) {
7592: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7593: else fprintf(ficgp," %%*lf (%%*lf)");
7594: }
1.265 brouard 7595: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7596:
7597: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7598: if(cptcoveff ==0){
1.271 brouard 7599: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7600: }else{
7601: kl=0;
7602: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7603: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7604: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7605: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7606: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7607: vlv= nbcode[Tvaraff[k]][lv];
7608: kl++;
7609: /* 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 *\/ */
7610: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7611: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7612: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7613: if(k==cptcoveff){
7614: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7615: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7616: }else{
7617: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7618: kl++;
7619: }
7620: } /* end covariate */
7621: } /* end if no covariate */
7622:
1.296 brouard 7623: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7624: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7625: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7626: if(cptcoveff ==0){
1.245 brouard 7627: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7628: }else{
7629: kl=0;
7630: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7631: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7632: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7633: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7634: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7635: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7636: kl++;
1.238 brouard 7637: /* 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 *\/ */
7638: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7639: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7640: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7641: if(k==cptcoveff){
1.245 brouard 7642: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7643: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7644: }else{
7645: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7646: kl++;
7647: }
7648: } /* end covariate */
7649: } /* end if no covariate */
1.296 brouard 7650: if(prevbcast == 1){
1.268 brouard 7651: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7652: /* k1-1 error should be nres-1*/
7653: for (i=1; i<= nlstate ; i ++) {
7654: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7655: else fprintf(ficgp," %%*lf (%%*lf)");
7656: }
1.271 brouard 7657: 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 7658: for (i=1; i<= nlstate ; i ++) {
7659: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7660: else fprintf(ficgp," %%*lf (%%*lf)");
7661: }
1.276 brouard 7662: 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 7663: for (i=1; i<= nlstate ; i ++) {
7664: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7665: else fprintf(ficgp," %%*lf (%%*lf)");
7666: }
1.274 brouard 7667: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7668: } /* end if backprojcast */
1.296 brouard 7669: } /* end if prevbcast */
1.276 brouard 7670: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7671: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7672: } /* nres */
1.201 brouard 7673: } /* k1 */
7674: } /* cpt */
1.235 brouard 7675:
7676:
1.126 brouard 7677: /*2 eme*/
1.238 brouard 7678: for (k1=1; k1<= m ; k1 ++){
7679: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7680: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7681: continue;
7682: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7683: strcpy(gplotlabel,"(");
1.238 brouard 7684: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7685: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7686: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7687: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7688: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7689: vlv= nbcode[Tvaraff[k]][lv];
7690: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7691: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7692: }
1.237 brouard 7693: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7694: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7695: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7696: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7697: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7698: }
1.264 brouard 7699: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7700: fprintf(ficgp,"\n#\n");
1.223 brouard 7701: if(invalidvarcomb[k1]){
7702: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7703: continue;
7704: }
1.219 brouard 7705:
1.241 brouard 7706: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7707: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7708: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7709: if(vpopbased==0){
1.238 brouard 7710: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7711: }else
1.238 brouard 7712: fprintf(ficgp,"\nreplot ");
7713: for (i=1; i<= nlstate+1 ; i ++) {
7714: k=2*i;
1.261 brouard 7715: 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 7716: for (j=1; j<= nlstate+1 ; j ++) {
7717: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7718: else fprintf(ficgp," %%*lf (%%*lf)");
7719: }
7720: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7721: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7722: 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 7723: for (j=1; j<= nlstate+1 ; j ++) {
7724: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7725: else fprintf(ficgp," %%*lf (%%*lf)");
7726: }
7727: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7728: 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 7729: for (j=1; j<= nlstate+1 ; j ++) {
7730: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7731: else fprintf(ficgp," %%*lf (%%*lf)");
7732: }
7733: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7734: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7735: } /* state */
7736: } /* vpopbased */
1.264 brouard 7737: 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 7738: } /* end nres */
7739: } /* k1 end 2 eme*/
7740:
7741:
7742: /*3eme*/
7743: for (k1=1; k1<= m ; k1 ++){
7744: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7745: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7746: continue;
7747:
7748: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7749: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7750: strcpy(gplotlabel,"(");
1.238 brouard 7751: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7752: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7753: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7754: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7755: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7756: vlv= nbcode[Tvaraff[k]][lv];
7757: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7758: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7759: }
7760: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7761: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7762: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7763: }
1.264 brouard 7764: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7765: fprintf(ficgp,"\n#\n");
7766: if(invalidvarcomb[k1]){
7767: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7768: continue;
7769: }
7770:
7771: /* k=2+nlstate*(2*cpt-2); */
7772: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7773: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7774: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7775: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7776: 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 7777: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7778: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7779: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7780: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7781: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7782: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7783:
1.238 brouard 7784: */
7785: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7786: 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 7787: /* 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 7788:
1.238 brouard 7789: }
1.261 brouard 7790: 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 7791: }
1.264 brouard 7792: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7793: } /* end nres */
7794: } /* end kl 3eme */
1.126 brouard 7795:
1.223 brouard 7796: /* 4eme */
1.201 brouard 7797: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7798: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7799: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7800: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7801: continue;
1.238 brouard 7802: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7803: strcpy(gplotlabel,"(");
1.238 brouard 7804: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7805: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7806: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7807: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7808: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7809: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7810: vlv= nbcode[Tvaraff[k]][lv];
7811: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7812: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7813: }
7814: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7815: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7816: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7817: }
1.264 brouard 7818: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7819: fprintf(ficgp,"\n#\n");
7820: if(invalidvarcomb[k1]){
7821: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7822: continue;
1.223 brouard 7823: }
1.238 brouard 7824:
1.241 brouard 7825: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7826: 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 7827: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7828: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7829: k=3;
7830: for (i=1; i<= nlstate ; i ++){
7831: if(i==1){
7832: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7833: }else{
7834: fprintf(ficgp,", '' ");
7835: }
7836: l=(nlstate+ndeath)*(i-1)+1;
7837: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7838: for (j=2; j<= nlstate+ndeath ; j ++)
7839: fprintf(ficgp,"+$%d",k+l+j-1);
7840: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7841: } /* nlstate */
1.264 brouard 7842: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7843: } /* end cpt state*/
7844: } /* end nres */
7845: } /* end covariate k1 */
7846:
1.220 brouard 7847: /* 5eme */
1.201 brouard 7848: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7849: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7850: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7851: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7852: continue;
1.238 brouard 7853: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7854: strcpy(gplotlabel,"(");
1.238 brouard 7855: 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);
7856: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7857: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7858: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7859: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7860: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7861: vlv= nbcode[Tvaraff[k]][lv];
7862: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7863: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7864: }
7865: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7866: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7867: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7868: }
1.264 brouard 7869: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7870: fprintf(ficgp,"\n#\n");
7871: if(invalidvarcomb[k1]){
7872: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7873: continue;
7874: }
1.227 brouard 7875:
1.241 brouard 7876: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7877: 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 7878: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7879: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7880: k=3;
7881: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7882: if(j==1)
7883: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7884: else
7885: fprintf(ficgp,", '' ");
7886: l=(nlstate+ndeath)*(cpt-1) +j;
7887: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7888: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7889: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7890: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7891: } /* nlstate */
7892: fprintf(ficgp,", '' ");
7893: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7894: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7895: l=(nlstate+ndeath)*(cpt-1) +j;
7896: if(j < nlstate)
7897: fprintf(ficgp,"$%d +",k+l);
7898: else
7899: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7900: }
1.264 brouard 7901: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7902: } /* end cpt state*/
7903: } /* end covariate */
7904: } /* end nres */
1.227 brouard 7905:
1.220 brouard 7906: /* 6eme */
1.202 brouard 7907: /* CV preval stable (period) for each covariate */
1.237 brouard 7908: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7909: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7910: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7911: continue;
1.255 brouard 7912: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7913: strcpy(gplotlabel,"(");
1.288 brouard 7914: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7915: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7916: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7917: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7918: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7919: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7920: vlv= nbcode[Tvaraff[k]][lv];
7921: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7922: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7923: }
1.237 brouard 7924: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7925: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7926: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7927: }
1.264 brouard 7928: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7929: fprintf(ficgp,"\n#\n");
1.223 brouard 7930: if(invalidvarcomb[k1]){
1.227 brouard 7931: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7932: continue;
1.223 brouard 7933: }
1.227 brouard 7934:
1.241 brouard 7935: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7936: 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 7937: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7938: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7939: k=3; /* Offset */
1.255 brouard 7940: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7941: if(i==1)
7942: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7943: else
7944: fprintf(ficgp,", '' ");
1.255 brouard 7945: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7946: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7947: for (j=2; j<= nlstate ; j ++)
7948: fprintf(ficgp,"+$%d",k+l+j-1);
7949: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7950: } /* nlstate */
1.264 brouard 7951: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7952: } /* end cpt state*/
7953: } /* end covariate */
1.227 brouard 7954:
7955:
1.220 brouard 7956: /* 7eme */
1.296 brouard 7957: if(prevbcast == 1){
1.288 brouard 7958: /* CV backward prevalence for each covariate */
1.237 brouard 7959: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7960: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7961: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7962: continue;
1.268 brouard 7963: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7964: strcpy(gplotlabel,"(");
1.288 brouard 7965: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7966: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7967: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7968: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7969: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7970: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7971: vlv= nbcode[Tvaraff[k]][lv];
7972: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7973: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7974: }
1.237 brouard 7975: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7976: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7977: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7978: }
1.264 brouard 7979: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7980: fprintf(ficgp,"\n#\n");
7981: if(invalidvarcomb[k1]){
7982: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7983: continue;
7984: }
7985:
1.241 brouard 7986: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7987: 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 7988: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7989: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7990: k=3; /* Offset */
1.268 brouard 7991: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7992: if(i==1)
7993: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7994: else
7995: fprintf(ficgp,", '' ");
7996: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7997: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 ! brouard 7998: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
! 7999: /* 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 8000: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8001: /* for (j=2; j<= nlstate ; j ++) */
8002: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8003: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8004: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8005: } /* nlstate */
1.264 brouard 8006: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8007: } /* end cpt state*/
8008: } /* end covariate */
1.296 brouard 8009: } /* End if prevbcast */
1.218 brouard 8010:
1.223 brouard 8011: /* 8eme */
1.218 brouard 8012: if(prevfcast==1){
1.288 brouard 8013: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8014:
1.237 brouard 8015: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8016: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8017: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8018: continue;
1.211 brouard 8019: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8020: strcpy(gplotlabel,"(");
1.288 brouard 8021: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8022: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8023: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8024: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8025: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8026: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8027: vlv= nbcode[Tvaraff[k]][lv];
8028: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8029: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8030: }
1.237 brouard 8031: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8032: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8033: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8034: }
1.264 brouard 8035: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8036: fprintf(ficgp,"\n#\n");
8037: if(invalidvarcomb[k1]){
8038: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8039: continue;
8040: }
8041:
8042: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8043: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8044: 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 8045: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8046: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8047:
8048: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8049: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8050: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8051: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8052: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8053: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8054: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8055: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8056: if(i==istart){
1.227 brouard 8057: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8058: }else{
8059: fprintf(ficgp,",\\\n '' ");
8060: }
8061: if(cptcoveff ==0){ /* No covariate */
8062: ioffset=2; /* Age is in 2 */
8063: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8064: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8065: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8066: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8067: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8068: if(i==nlstate+1){
1.270 brouard 8069: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8070: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8071: fprintf(ficgp,",\\\n '' ");
8072: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8073: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8074: offyear, \
1.268 brouard 8075: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8076: }else
1.227 brouard 8077: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8078: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8079: }else{ /* more than 2 covariates */
1.270 brouard 8080: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8081: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8082: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8083: iyearc=ioffset-1;
8084: iagec=ioffset;
1.227 brouard 8085: fprintf(ficgp," u %d:(",ioffset);
8086: kl=0;
8087: strcpy(gplotcondition,"(");
8088: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8089: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8090: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8091: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8092: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8093: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8094: kl++;
8095: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8096: kl++;
8097: if(k <cptcoveff && cptcoveff>1)
8098: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8099: }
8100: strcpy(gplotcondition+strlen(gplotcondition),")");
8101: /* 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 *\/ */
8102: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8103: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8104: /* '' 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*/
8105: if(i==nlstate+1){
1.270 brouard 8106: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8107: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8108: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8109: fprintf(ficgp," u %d:(",iagec);
8110: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8111: iyearc, iagec, offyear, \
8112: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8113: /* '' 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 8114: }else{
8115: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8116: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8117: }
8118: } /* end if covariate */
8119: } /* nlstate */
1.264 brouard 8120: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8121: } /* end cpt state*/
8122: } /* end covariate */
8123: } /* End if prevfcast */
1.227 brouard 8124:
1.296 brouard 8125: if(prevbcast==1){
1.268 brouard 8126: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8127:
8128: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8129: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8130: if(m != 1 && TKresult[nres]!= k1)
8131: continue;
8132: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8133: strcpy(gplotlabel,"(");
8134: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
8135: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8136: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8137: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8138: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8139: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8140: vlv= nbcode[Tvaraff[k]][lv];
8141: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8142: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8143: }
8144: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8145: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8146: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8147: }
8148: strcpy(gplotlabel+strlen(gplotlabel),")");
8149: fprintf(ficgp,"\n#\n");
8150: if(invalidvarcomb[k1]){
8151: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8152: continue;
8153: }
8154:
8155: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8156: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8157: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8158: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8159: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8160:
8161: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8162: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8163: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8164: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8165: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8166: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8167: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8168: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8169: if(i==istart){
8170: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8171: }else{
8172: fprintf(ficgp,",\\\n '' ");
8173: }
8174: if(cptcoveff ==0){ /* No covariate */
8175: ioffset=2; /* Age is in 2 */
8176: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8177: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8178: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8179: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8180: fprintf(ficgp," u %d:(", ioffset);
8181: if(i==nlstate+1){
1.270 brouard 8182: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8183: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8184: fprintf(ficgp,",\\\n '' ");
8185: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8186: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8187: offbyear, \
8188: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8189: }else
8190: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8191: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8192: }else{ /* more than 2 covariates */
1.270 brouard 8193: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8194: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8195: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8196: iyearc=ioffset-1;
8197: iagec=ioffset;
1.268 brouard 8198: fprintf(ficgp," u %d:(",ioffset);
8199: kl=0;
8200: strcpy(gplotcondition,"(");
8201: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8202: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8203: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8204: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8205: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8206: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8207: kl++;
8208: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8209: kl++;
8210: if(k <cptcoveff && cptcoveff>1)
8211: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8212: }
8213: strcpy(gplotcondition+strlen(gplotcondition),")");
8214: /* 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 *\/ */
8215: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8216: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8217: /* '' 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*/
8218: if(i==nlstate+1){
1.270 brouard 8219: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8220: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8221: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8222: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8223: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8224: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8225: iyearc,iagec,offbyear, \
8226: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8227: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8228: }else{
8229: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8230: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8231: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8232: }
8233: } /* end if covariate */
8234: } /* nlstate */
8235: fprintf(ficgp,"\nset out; unset label;\n");
8236: } /* end cpt state*/
8237: } /* end covariate */
1.296 brouard 8238: } /* End if prevbcast */
1.268 brouard 8239:
1.227 brouard 8240:
1.238 brouard 8241: /* 9eme writing MLE parameters */
8242: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8243: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8244: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8245: for(k=1; k <=(nlstate+ndeath); k++){
8246: if (k != i) {
1.227 brouard 8247: fprintf(ficgp,"# current state %d\n",k);
8248: for(j=1; j <=ncovmodel; j++){
8249: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8250: jk++;
8251: }
8252: fprintf(ficgp,"\n");
1.126 brouard 8253: }
8254: }
1.223 brouard 8255: }
1.187 brouard 8256: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8257:
1.145 brouard 8258: /*goto avoid;*/
1.238 brouard 8259: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8260: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8261: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8262: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8263: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8264: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8265: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8266: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8267: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8268: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8269: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8270: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8271: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8272: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8273: fprintf(ficgp,"#\n");
1.223 brouard 8274: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8275: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8276: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8277: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8278: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8279: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8280: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8281: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8282: continue;
1.264 brouard 8283: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8284: strcpy(gplotlabel,"(");
1.276 brouard 8285: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8286: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8287: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8288: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8289: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8290: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8291: vlv= nbcode[Tvaraff[k]][lv];
8292: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8293: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8294: }
1.237 brouard 8295: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8296: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8297: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8298: }
1.264 brouard 8299: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8300: fprintf(ficgp,"\n#\n");
1.264 brouard 8301: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8302: fprintf(ficgp,"\nset key outside ");
8303: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8304: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8305: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8306: if (ng==1){
8307: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8308: fprintf(ficgp,"\nunset log y");
8309: }else if (ng==2){
8310: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8311: fprintf(ficgp,"\nset log y");
8312: }else if (ng==3){
8313: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8314: fprintf(ficgp,"\nset log y");
8315: }else
8316: fprintf(ficgp,"\nunset title ");
8317: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8318: i=1;
8319: for(k2=1; k2<=nlstate; k2++) {
8320: k3=i;
8321: for(k=1; k<=(nlstate+ndeath); k++) {
8322: if (k != k2){
8323: switch( ng) {
8324: case 1:
8325: if(nagesqr==0)
8326: fprintf(ficgp," p%d+p%d*x",i,i+1);
8327: else /* nagesqr =1 */
8328: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8329: break;
8330: case 2: /* ng=2 */
8331: if(nagesqr==0)
8332: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8333: else /* nagesqr =1 */
8334: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8335: break;
8336: case 3:
8337: if(nagesqr==0)
8338: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8339: else /* nagesqr =1 */
8340: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8341: break;
8342: }
8343: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8344: ijp=1; /* product no age */
8345: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8346: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8347: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8348: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8349: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8350: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8351: if(DummyV[j]==0){
8352: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8353: }else{ /* quantitative */
8354: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8355: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8356: }
8357: ij++;
1.237 brouard 8358: }
1.268 brouard 8359: }
8360: }else if(cptcovprod >0){
8361: if(j==Tprod[ijp]) { /* */
8362: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8363: if(ijp <=cptcovprod) { /* Product */
8364: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8365: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8366: /* 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)]); */
8367: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8368: }else{ /* Vn is dummy and Vm is quanti */
8369: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8370: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8371: }
8372: }else{ /* Vn*Vm Vn is quanti */
8373: if(DummyV[Tvard[ijp][2]]==0){
8374: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8375: }else{ /* Both quanti */
8376: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8377: }
1.237 brouard 8378: }
1.268 brouard 8379: ijp++;
1.237 brouard 8380: }
1.268 brouard 8381: } /* end Tprod */
1.237 brouard 8382: } else{ /* simple covariate */
1.264 brouard 8383: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8384: if(Dummy[j]==0){
8385: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8386: }else{ /* quantitative */
8387: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8388: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8389: }
1.237 brouard 8390: } /* end simple */
8391: } /* end j */
1.223 brouard 8392: }else{
8393: i=i-ncovmodel;
8394: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8395: fprintf(ficgp," (1.");
8396: }
1.227 brouard 8397:
1.223 brouard 8398: if(ng != 1){
8399: fprintf(ficgp,")/(1");
1.227 brouard 8400:
1.264 brouard 8401: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8402: if(nagesqr==0)
1.264 brouard 8403: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8404: else /* nagesqr =1 */
1.264 brouard 8405: 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 8406:
1.223 brouard 8407: ij=1;
8408: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8409: if(cptcovage >0){
8410: if((j-2)==Tage[ij]) { /* Bug valgrind */
8411: if(ij <=cptcovage) { /* Bug valgrind */
8412: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8413: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8414: ij++;
8415: }
8416: }
8417: }else
8418: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/* Valgrind bug nbcode */
1.223 brouard 8419: }
8420: fprintf(ficgp,")");
8421: }
8422: fprintf(ficgp,")");
8423: if(ng ==2)
1.276 brouard 8424: 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 8425: else /* ng= 3 */
1.276 brouard 8426: 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);
1.223 brouard 8427: }else{ /* end ng <> 1 */
8428: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8429: 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 8430: }
8431: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8432: fprintf(ficgp,",");
8433: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8434: fprintf(ficgp,",");
8435: i=i+ncovmodel;
8436: } /* end k */
8437: } /* end k2 */
1.276 brouard 8438: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8439: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8440: } /* end k1 */
1.223 brouard 8441: } /* end ng */
8442: /* avoid: */
8443: fflush(ficgp);
1.126 brouard 8444: } /* end gnuplot */
8445:
8446:
8447: /*************** Moving average **************/
1.219 brouard 8448: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8449: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8450:
1.222 brouard 8451: int i, cpt, cptcod;
8452: int modcovmax =1;
8453: int mobilavrange, mob;
8454: int iage=0;
1.288 brouard 8455: int firstA1=0, firstA2=0;
1.222 brouard 8456:
1.266 brouard 8457: double sum=0., sumr=0.;
1.222 brouard 8458: double age;
1.266 brouard 8459: double *sumnewp, *sumnewm, *sumnewmr;
8460: double *agemingood, *agemaxgood;
8461: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8462:
8463:
1.278 brouard 8464: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8465: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8466:
8467: sumnewp = vector(1,ncovcombmax);
8468: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8469: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8470: agemingood = vector(1,ncovcombmax);
1.266 brouard 8471: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8472: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8473: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8474:
8475: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8476: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8477: sumnewp[cptcod]=0.;
1.266 brouard 8478: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8479: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8480: }
8481: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8482:
1.266 brouard 8483: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8484: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8485: else mobilavrange=mobilav;
8486: for (age=bage; age<=fage; age++)
8487: for (i=1; i<=nlstate;i++)
8488: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8489: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8490: /* We keep the original values on the extreme ages bage, fage and for
8491: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8492: we use a 5 terms etc. until the borders are no more concerned.
8493: */
8494: for (mob=3;mob <=mobilavrange;mob=mob+2){
8495: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8496: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8497: sumnewm[cptcod]=0.;
8498: for (i=1; i<=nlstate;i++){
1.222 brouard 8499: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8500: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8501: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8502: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8503: }
8504: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8505: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8506: } /* end i */
8507: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8508: } /* end cptcod */
1.222 brouard 8509: }/* end age */
8510: }/* end mob */
1.266 brouard 8511: }else{
8512: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8513: return -1;
1.266 brouard 8514: }
8515:
8516: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8517: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8518: if(invalidvarcomb[cptcod]){
8519: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8520: continue;
8521: }
1.219 brouard 8522:
1.266 brouard 8523: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8524: sumnewm[cptcod]=0.;
8525: sumnewmr[cptcod]=0.;
8526: for (i=1; i<=nlstate;i++){
8527: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8528: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8529: }
8530: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8531: agemingoodr[cptcod]=age;
8532: }
8533: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8534: agemingood[cptcod]=age;
8535: }
8536: } /* age */
8537: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8538: sumnewm[cptcod]=0.;
1.266 brouard 8539: sumnewmr[cptcod]=0.;
1.222 brouard 8540: for (i=1; i<=nlstate;i++){
8541: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8542: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8543: }
8544: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8545: agemaxgoodr[cptcod]=age;
1.222 brouard 8546: }
8547: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8548: agemaxgood[cptcod]=age;
8549: }
8550: } /* age */
8551: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8552: /* but they will change */
1.288 brouard 8553: firstA1=0;firstA2=0;
1.266 brouard 8554: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8555: sumnewm[cptcod]=0.;
8556: sumnewmr[cptcod]=0.;
8557: for (i=1; i<=nlstate;i++){
8558: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8559: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8560: }
8561: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8562: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8563: agemaxgoodr[cptcod]=age; /* age min */
8564: for (i=1; i<=nlstate;i++)
8565: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8566: }else{ /* bad we change the value with the values of good ages */
8567: for (i=1; i<=nlstate;i++){
8568: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8569: } /* i */
8570: } /* end bad */
8571: }else{
8572: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8573: agemaxgood[cptcod]=age;
8574: }else{ /* bad we change the value with the values of good ages */
8575: for (i=1; i<=nlstate;i++){
8576: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8577: } /* i */
8578: } /* end bad */
8579: }/* end else */
8580: sum=0.;sumr=0.;
8581: for (i=1; i<=nlstate;i++){
8582: sum+=mobaverage[(int)age][i][cptcod];
8583: sumr+=probs[(int)age][i][cptcod];
8584: }
8585: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8586: if(!firstA1){
8587: firstA1=1;
8588: 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);
8589: }
8590: 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 8591: } /* end bad */
8592: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8593: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8594: if(!firstA2){
8595: firstA2=1;
8596: 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);
8597: }
8598: 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 8599: } /* end bad */
8600: }/* age */
1.266 brouard 8601:
8602: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8603: sumnewm[cptcod]=0.;
1.266 brouard 8604: sumnewmr[cptcod]=0.;
1.222 brouard 8605: for (i=1; i<=nlstate;i++){
8606: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8607: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8608: }
8609: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8610: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8611: agemingoodr[cptcod]=age;
8612: for (i=1; i<=nlstate;i++)
8613: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8614: }else{ /* bad we change the value with the values of good ages */
8615: for (i=1; i<=nlstate;i++){
8616: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8617: } /* i */
8618: } /* end bad */
8619: }else{
8620: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8621: agemingood[cptcod]=age;
8622: }else{ /* bad */
8623: for (i=1; i<=nlstate;i++){
8624: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8625: } /* i */
8626: } /* end bad */
8627: }/* end else */
8628: sum=0.;sumr=0.;
8629: for (i=1; i<=nlstate;i++){
8630: sum+=mobaverage[(int)age][i][cptcod];
8631: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8632: }
1.266 brouard 8633: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8634: 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 8635: } /* end bad */
8636: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8637: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8638: 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 8639: } /* end bad */
8640: }/* age */
1.266 brouard 8641:
1.222 brouard 8642:
8643: for (age=bage; age<=fage; age++){
1.235 brouard 8644: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8645: sumnewp[cptcod]=0.;
8646: sumnewm[cptcod]=0.;
8647: for (i=1; i<=nlstate;i++){
8648: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8649: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8650: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8651: }
8652: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8653: }
8654: /* printf("\n"); */
8655: /* } */
1.266 brouard 8656:
1.222 brouard 8657: /* brutal averaging */
1.266 brouard 8658: /* for (i=1; i<=nlstate;i++){ */
8659: /* for (age=1; age<=bage; age++){ */
8660: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8661: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8662: /* } */
8663: /* for (age=fage; age<=AGESUP; age++){ */
8664: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8665: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8666: /* } */
8667: /* } /\* end i status *\/ */
8668: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8669: /* for (age=1; age<=AGESUP; age++){ */
8670: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8671: /* mobaverage[(int)age][i][cptcod]=0.; */
8672: /* } */
8673: /* } */
1.222 brouard 8674: }/* end cptcod */
1.266 brouard 8675: free_vector(agemaxgoodr,1, ncovcombmax);
8676: free_vector(agemaxgood,1, ncovcombmax);
8677: free_vector(agemingood,1, ncovcombmax);
8678: free_vector(agemingoodr,1, ncovcombmax);
8679: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8680: free_vector(sumnewm,1, ncovcombmax);
8681: free_vector(sumnewp,1, ncovcombmax);
8682: return 0;
8683: }/* End movingaverage */
1.218 brouard 8684:
1.126 brouard 8685:
1.296 brouard 8686:
1.126 brouard 8687: /************** Forecasting ******************/
1.296 brouard 8688: /* 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)*/
8689: 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){
8690: /* dateintemean, mean date of interviews
8691: dateprojd, year, month, day of starting projection
8692: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8693: agemin, agemax range of age
8694: dateprev1 dateprev2 range of dates during which prevalence is computed
8695: */
1.296 brouard 8696: /* double anprojd, mprojd, jprojd; */
8697: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8698: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8699: double agec; /* generic age */
1.296 brouard 8700: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8701: double *popeffectif,*popcount;
8702: double ***p3mat;
1.218 brouard 8703: /* double ***mobaverage; */
1.126 brouard 8704: char fileresf[FILENAMELENGTH];
8705:
8706: agelim=AGESUP;
1.211 brouard 8707: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8708: in each health status at the date of interview (if between dateprev1 and dateprev2).
8709: We still use firstpass and lastpass as another selection.
8710: */
1.214 brouard 8711: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8712: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8713:
1.201 brouard 8714: strcpy(fileresf,"F_");
8715: strcat(fileresf,fileresu);
1.126 brouard 8716: if((ficresf=fopen(fileresf,"w"))==NULL) {
8717: printf("Problem with forecast resultfile: %s\n", fileresf);
8718: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8719: }
1.235 brouard 8720: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8721: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8722:
1.225 brouard 8723: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8724:
8725:
8726: stepsize=(int) (stepm+YEARM-1)/YEARM;
8727: if (stepm<=12) stepsize=1;
8728: if(estepm < stepm){
8729: printf ("Problem %d lower than %d\n",estepm, stepm);
8730: }
1.270 brouard 8731: else{
8732: hstepm=estepm;
8733: }
8734: if(estepm > stepm){ /* Yes every two year */
8735: stepsize=2;
8736: }
1.296 brouard 8737: hstepm=hstepm/stepm;
1.126 brouard 8738:
1.296 brouard 8739:
8740: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8741: /* fractional in yp1 *\/ */
8742: /* aintmean=yp; */
8743: /* yp2=modf((yp1*12),&yp); */
8744: /* mintmean=yp; */
8745: /* yp1=modf((yp2*30.5),&yp); */
8746: /* jintmean=yp; */
8747: /* if(jintmean==0) jintmean=1; */
8748: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8749:
1.296 brouard 8750:
8751: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8752: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8753: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8754: i1=pow(2,cptcoveff);
1.126 brouard 8755: if (cptcovn < 1){i1=1;}
8756:
1.296 brouard 8757: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8758:
8759: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8760:
1.126 brouard 8761: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8762: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8763: for(k=1; k<=i1;k++){
1.253 brouard 8764: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8765: continue;
1.227 brouard 8766: if(invalidvarcomb[k]){
8767: printf("\nCombination (%d) projection ignored because no cases \n",k);
8768: continue;
8769: }
8770: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8771: for(j=1;j<=cptcoveff;j++) {
8772: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8773: }
1.235 brouard 8774: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8775: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8776: }
1.227 brouard 8777: fprintf(ficresf," yearproj age");
8778: for(j=1; j<=nlstate+ndeath;j++){
8779: for(i=1; i<=nlstate;i++)
8780: fprintf(ficresf," p%d%d",i,j);
8781: fprintf(ficresf," wp.%d",j);
8782: }
1.296 brouard 8783: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8784: fprintf(ficresf,"\n");
1.296 brouard 8785: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8786: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8787: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8788: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8789: nhstepm = nhstepm/hstepm;
8790: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8791: oldm=oldms;savm=savms;
1.268 brouard 8792: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8793: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8794: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8795: for (h=0; h<=nhstepm; h++){
8796: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8797: break;
8798: }
8799: }
8800: fprintf(ficresf,"\n");
8801: for(j=1;j<=cptcoveff;j++)
8802: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8803: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8804:
8805: for(j=1; j<=nlstate+ndeath;j++) {
8806: ppij=0.;
8807: for(i=1; i<=nlstate;i++) {
1.278 brouard 8808: if (mobilav>=1)
8809: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8810: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8811: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8812: }
1.268 brouard 8813: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8814: } /* end i */
8815: fprintf(ficresf," %.3f", ppij);
8816: }/* end j */
1.227 brouard 8817: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8818: } /* end agec */
1.266 brouard 8819: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8820: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8821: } /* end yearp */
8822: } /* end k */
1.219 brouard 8823:
1.126 brouard 8824: fclose(ficresf);
1.215 brouard 8825: printf("End of Computing forecasting \n");
8826: fprintf(ficlog,"End of Computing forecasting\n");
8827:
1.126 brouard 8828: }
8829:
1.269 brouard 8830: /************** Back Forecasting ******************/
1.296 brouard 8831: /* 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){ */
8832: 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){
8833: /* back1, year, month, day of starting backprojection
1.267 brouard 8834: agemin, agemax range of age
8835: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8836: anback2 year of end of backprojection (same day and month as back1).
8837: prevacurrent and prev are prevalences.
1.267 brouard 8838: */
8839: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8840: double agec; /* generic age */
1.302 brouard 8841: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 8842: double *popeffectif,*popcount;
8843: double ***p3mat;
8844: /* double ***mobaverage; */
8845: char fileresfb[FILENAMELENGTH];
8846:
1.268 brouard 8847: agelim=AGEINF;
1.267 brouard 8848: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8849: in each health status at the date of interview (if between dateprev1 and dateprev2).
8850: We still use firstpass and lastpass as another selection.
8851: */
8852: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8853: /* firstpass, lastpass, stepm, weightopt, model); */
8854:
8855: /*Do we need to compute prevalence again?*/
8856:
8857: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8858:
8859: strcpy(fileresfb,"FB_");
8860: strcat(fileresfb,fileresu);
8861: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8862: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8863: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8864: }
8865: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8866: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8867:
8868: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8869:
8870:
8871: stepsize=(int) (stepm+YEARM-1)/YEARM;
8872: if (stepm<=12) stepsize=1;
8873: if(estepm < stepm){
8874: printf ("Problem %d lower than %d\n",estepm, stepm);
8875: }
1.270 brouard 8876: else{
8877: hstepm=estepm;
8878: }
8879: if(estepm >= stepm){ /* Yes every two year */
8880: stepsize=2;
8881: }
1.267 brouard 8882:
8883: hstepm=hstepm/stepm;
1.296 brouard 8884: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8885: /* fractional in yp1 *\/ */
8886: /* aintmean=yp; */
8887: /* yp2=modf((yp1*12),&yp); */
8888: /* mintmean=yp; */
8889: /* yp1=modf((yp2*30.5),&yp); */
8890: /* jintmean=yp; */
8891: /* if(jintmean==0) jintmean=1; */
8892: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8893:
8894: i1=pow(2,cptcoveff);
8895: if (cptcovn < 1){i1=1;}
8896:
1.296 brouard 8897: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8898: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8899:
8900: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8901:
8902: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8903: for(k=1; k<=i1;k++){
8904: if(i1 != 1 && TKresult[nres]!= k)
8905: continue;
8906: if(invalidvarcomb[k]){
8907: printf("\nCombination (%d) projection ignored because no cases \n",k);
8908: continue;
8909: }
1.268 brouard 8910: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8911: for(j=1;j<=cptcoveff;j++) {
8912: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8913: }
8914: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8915: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8916: }
8917: fprintf(ficresfb," yearbproj age");
8918: for(j=1; j<=nlstate+ndeath;j++){
8919: for(i=1; i<=nlstate;i++)
1.268 brouard 8920: fprintf(ficresfb," b%d%d",i,j);
8921: fprintf(ficresfb," b.%d",j);
1.267 brouard 8922: }
1.296 brouard 8923: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8924: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8925: fprintf(ficresfb,"\n");
1.296 brouard 8926: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8927: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8928: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8929: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8930: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8931: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8932: nhstepm = nhstepm/hstepm;
8933: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8934: oldm=oldms;savm=savms;
1.268 brouard 8935: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8936: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8937: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8938: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8939: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8940: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8941: for (h=0; h<=nhstepm; h++){
1.268 brouard 8942: if (h*hstepm/YEARM*stepm ==-yearp) {
8943: break;
8944: }
8945: }
8946: fprintf(ficresfb,"\n");
8947: for(j=1;j<=cptcoveff;j++)
8948: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8949: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8950: for(i=1; i<=nlstate+ndeath;i++) {
8951: ppij=0.;ppi=0.;
8952: for(j=1; j<=nlstate;j++) {
8953: /* if (mobilav==1) */
1.269 brouard 8954: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8955: ppi=ppi+prevacurrent[(int)agec][j][k];
8956: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8957: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8958: /* else { */
8959: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8960: /* } */
1.268 brouard 8961: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8962: } /* end j */
8963: if(ppi <0.99){
8964: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8965: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8966: }
8967: fprintf(ficresfb," %.3f", ppij);
8968: }/* end j */
1.267 brouard 8969: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8970: } /* end agec */
8971: } /* end yearp */
8972: } /* end k */
1.217 brouard 8973:
1.267 brouard 8974: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8975:
1.267 brouard 8976: fclose(ficresfb);
8977: printf("End of Computing Back forecasting \n");
8978: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8979:
1.267 brouard 8980: }
1.217 brouard 8981:
1.269 brouard 8982: /* Variance of prevalence limit: varprlim */
8983: 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 8984: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8985:
8986: char fileresvpl[FILENAMELENGTH];
8987: FILE *ficresvpl;
8988: double **oldm, **savm;
8989: double **varpl; /* Variances of prevalence limits by age */
8990: int i1, k, nres, j ;
8991:
8992: strcpy(fileresvpl,"VPL_");
8993: strcat(fileresvpl,fileresu);
8994: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8995: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8996: exit(0);
8997: }
1.288 brouard 8998: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8999: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9000:
9001: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9002: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9003:
9004: i1=pow(2,cptcoveff);
9005: if (cptcovn < 1){i1=1;}
9006:
9007: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9008: for(k=1; k<=i1;k++){
9009: if(i1 != 1 && TKresult[nres]!= k)
9010: continue;
9011: fprintf(ficresvpl,"\n#****** ");
9012: printf("\n#****** ");
9013: fprintf(ficlog,"\n#****** ");
9014: for(j=1;j<=cptcoveff;j++) {
9015: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9016: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9017: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9018: }
9019: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9020: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9021: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9022: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9023: }
9024: fprintf(ficresvpl,"******\n");
9025: printf("******\n");
9026: fprintf(ficlog,"******\n");
9027:
9028: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9029: oldm=oldms;savm=savms;
9030: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9031: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9032: /*}*/
9033: }
9034:
9035: fclose(ficresvpl);
1.288 brouard 9036: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9037: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9038:
9039: }
9040: /* Variance of back prevalence: varbprlim */
9041: 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){
9042: /*------- Variance of back (stable) prevalence------*/
9043:
9044: char fileresvbl[FILENAMELENGTH];
9045: FILE *ficresvbl;
9046:
9047: double **oldm, **savm;
9048: double **varbpl; /* Variances of back prevalence limits by age */
9049: int i1, k, nres, j ;
9050:
9051: strcpy(fileresvbl,"VBL_");
9052: strcat(fileresvbl,fileresu);
9053: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9054: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9055: exit(0);
9056: }
9057: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9058: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9059:
9060:
9061: i1=pow(2,cptcoveff);
9062: if (cptcovn < 1){i1=1;}
9063:
9064: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9065: for(k=1; k<=i1;k++){
9066: if(i1 != 1 && TKresult[nres]!= k)
9067: continue;
9068: fprintf(ficresvbl,"\n#****** ");
9069: printf("\n#****** ");
9070: fprintf(ficlog,"\n#****** ");
9071: for(j=1;j<=cptcoveff;j++) {
9072: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9073: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9074: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9075: }
9076: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9077: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9078: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9079: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9080: }
9081: fprintf(ficresvbl,"******\n");
9082: printf("******\n");
9083: fprintf(ficlog,"******\n");
9084:
9085: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9086: oldm=oldms;savm=savms;
9087:
9088: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9089: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9090: /*}*/
9091: }
9092:
9093: fclose(ficresvbl);
9094: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9095: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9096:
9097: } /* End of varbprlim */
9098:
1.126 brouard 9099: /************** Forecasting *****not tested NB*************/
1.227 brouard 9100: /* 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 9101:
1.227 brouard 9102: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9103: /* int *popage; */
9104: /* double calagedatem, agelim, kk1, kk2; */
9105: /* double *popeffectif,*popcount; */
9106: /* double ***p3mat,***tabpop,***tabpopprev; */
9107: /* /\* double ***mobaverage; *\/ */
9108: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9109:
1.227 brouard 9110: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9111: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9112: /* agelim=AGESUP; */
9113: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9114:
1.227 brouard 9115: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9116:
9117:
1.227 brouard 9118: /* strcpy(filerespop,"POP_"); */
9119: /* strcat(filerespop,fileresu); */
9120: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9121: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9122: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9123: /* } */
9124: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9125: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9126:
1.227 brouard 9127: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9128:
1.227 brouard 9129: /* /\* if (mobilav!=0) { *\/ */
9130: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9131: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9132: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9133: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9134: /* /\* } *\/ */
9135: /* /\* } *\/ */
1.126 brouard 9136:
1.227 brouard 9137: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9138: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9139:
1.227 brouard 9140: /* agelim=AGESUP; */
1.126 brouard 9141:
1.227 brouard 9142: /* hstepm=1; */
9143: /* hstepm=hstepm/stepm; */
1.218 brouard 9144:
1.227 brouard 9145: /* if (popforecast==1) { */
9146: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9147: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9148: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9149: /* } */
9150: /* popage=ivector(0,AGESUP); */
9151: /* popeffectif=vector(0,AGESUP); */
9152: /* popcount=vector(0,AGESUP); */
1.126 brouard 9153:
1.227 brouard 9154: /* i=1; */
9155: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9156:
1.227 brouard 9157: /* imx=i; */
9158: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9159: /* } */
1.218 brouard 9160:
1.227 brouard 9161: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9162: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9163: /* k=k+1; */
9164: /* fprintf(ficrespop,"\n#******"); */
9165: /* for(j=1;j<=cptcoveff;j++) { */
9166: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9167: /* } */
9168: /* fprintf(ficrespop,"******\n"); */
9169: /* fprintf(ficrespop,"# Age"); */
9170: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9171: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9172:
1.227 brouard 9173: /* for (cpt=0; cpt<=0;cpt++) { */
9174: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9175:
1.227 brouard 9176: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9177: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9178: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9179:
1.227 brouard 9180: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9181: /* oldm=oldms;savm=savms; */
9182: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9183:
1.227 brouard 9184: /* for (h=0; h<=nhstepm; h++){ */
9185: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9186: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9187: /* } */
9188: /* for(j=1; j<=nlstate+ndeath;j++) { */
9189: /* kk1=0.;kk2=0; */
9190: /* for(i=1; i<=nlstate;i++) { */
9191: /* if (mobilav==1) */
9192: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9193: /* else { */
9194: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9195: /* } */
9196: /* } */
9197: /* if (h==(int)(calagedatem+12*cpt)){ */
9198: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9199: /* /\*fprintf(ficrespop," %.3f", kk1); */
9200: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9201: /* } */
9202: /* } */
9203: /* for(i=1; i<=nlstate;i++){ */
9204: /* kk1=0.; */
9205: /* for(j=1; j<=nlstate;j++){ */
9206: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9207: /* } */
9208: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9209: /* } */
1.218 brouard 9210:
1.227 brouard 9211: /* if (h==(int)(calagedatem+12*cpt)) */
9212: /* for(j=1; j<=nlstate;j++) */
9213: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9214: /* } */
9215: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9216: /* } */
9217: /* } */
1.218 brouard 9218:
1.227 brouard 9219: /* /\******\/ */
1.218 brouard 9220:
1.227 brouard 9221: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9222: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9223: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9224: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9225: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9226:
1.227 brouard 9227: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9228: /* oldm=oldms;savm=savms; */
9229: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9230: /* for (h=0; h<=nhstepm; h++){ */
9231: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9232: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9233: /* } */
9234: /* for(j=1; j<=nlstate+ndeath;j++) { */
9235: /* kk1=0.;kk2=0; */
9236: /* for(i=1; i<=nlstate;i++) { */
9237: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9238: /* } */
9239: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9240: /* } */
9241: /* } */
9242: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9243: /* } */
9244: /* } */
9245: /* } */
9246: /* } */
1.218 brouard 9247:
1.227 brouard 9248: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9249:
1.227 brouard 9250: /* if (popforecast==1) { */
9251: /* free_ivector(popage,0,AGESUP); */
9252: /* free_vector(popeffectif,0,AGESUP); */
9253: /* free_vector(popcount,0,AGESUP); */
9254: /* } */
9255: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9256: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9257: /* fclose(ficrespop); */
9258: /* } /\* End of popforecast *\/ */
1.218 brouard 9259:
1.126 brouard 9260: int fileappend(FILE *fichier, char *optionfich)
9261: {
9262: if((fichier=fopen(optionfich,"a"))==NULL) {
9263: printf("Problem with file: %s\n", optionfich);
9264: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9265: return (0);
9266: }
9267: fflush(fichier);
9268: return (1);
9269: }
9270:
9271:
9272: /**************** function prwizard **********************/
9273: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9274: {
9275:
9276: /* Wizard to print covariance matrix template */
9277:
1.164 brouard 9278: char ca[32], cb[32];
9279: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9280: int numlinepar;
9281:
9282: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9283: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9284: for(i=1; i <=nlstate; i++){
9285: jj=0;
9286: for(j=1; j <=nlstate+ndeath; j++){
9287: if(j==i) continue;
9288: jj++;
9289: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9290: printf("%1d%1d",i,j);
9291: fprintf(ficparo,"%1d%1d",i,j);
9292: for(k=1; k<=ncovmodel;k++){
9293: /* printf(" %lf",param[i][j][k]); */
9294: /* fprintf(ficparo," %lf",param[i][j][k]); */
9295: printf(" 0.");
9296: fprintf(ficparo," 0.");
9297: }
9298: printf("\n");
9299: fprintf(ficparo,"\n");
9300: }
9301: }
9302: printf("# Scales (for hessian or gradient estimation)\n");
9303: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9304: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9305: for(i=1; i <=nlstate; i++){
9306: jj=0;
9307: for(j=1; j <=nlstate+ndeath; j++){
9308: if(j==i) continue;
9309: jj++;
9310: fprintf(ficparo,"%1d%1d",i,j);
9311: printf("%1d%1d",i,j);
9312: fflush(stdout);
9313: for(k=1; k<=ncovmodel;k++){
9314: /* printf(" %le",delti3[i][j][k]); */
9315: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9316: printf(" 0.");
9317: fprintf(ficparo," 0.");
9318: }
9319: numlinepar++;
9320: printf("\n");
9321: fprintf(ficparo,"\n");
9322: }
9323: }
9324: printf("# Covariance matrix\n");
9325: /* # 121 Var(a12)\n\ */
9326: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9327: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9328: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9329: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9330: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9331: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9332: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9333: fflush(stdout);
9334: fprintf(ficparo,"# Covariance matrix\n");
9335: /* # 121 Var(a12)\n\ */
9336: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9337: /* # ...\n\ */
9338: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9339:
9340: for(itimes=1;itimes<=2;itimes++){
9341: jj=0;
9342: for(i=1; i <=nlstate; i++){
9343: for(j=1; j <=nlstate+ndeath; j++){
9344: if(j==i) continue;
9345: for(k=1; k<=ncovmodel;k++){
9346: jj++;
9347: ca[0]= k+'a'-1;ca[1]='\0';
9348: if(itimes==1){
9349: printf("#%1d%1d%d",i,j,k);
9350: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9351: }else{
9352: printf("%1d%1d%d",i,j,k);
9353: fprintf(ficparo,"%1d%1d%d",i,j,k);
9354: /* printf(" %.5le",matcov[i][j]); */
9355: }
9356: ll=0;
9357: for(li=1;li <=nlstate; li++){
9358: for(lj=1;lj <=nlstate+ndeath; lj++){
9359: if(lj==li) continue;
9360: for(lk=1;lk<=ncovmodel;lk++){
9361: ll++;
9362: if(ll<=jj){
9363: cb[0]= lk +'a'-1;cb[1]='\0';
9364: if(ll<jj){
9365: if(itimes==1){
9366: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9367: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9368: }else{
9369: printf(" 0.");
9370: fprintf(ficparo," 0.");
9371: }
9372: }else{
9373: if(itimes==1){
9374: printf(" Var(%s%1d%1d)",ca,i,j);
9375: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9376: }else{
9377: printf(" 0.");
9378: fprintf(ficparo," 0.");
9379: }
9380: }
9381: }
9382: } /* end lk */
9383: } /* end lj */
9384: } /* end li */
9385: printf("\n");
9386: fprintf(ficparo,"\n");
9387: numlinepar++;
9388: } /* end k*/
9389: } /*end j */
9390: } /* end i */
9391: } /* end itimes */
9392:
9393: } /* end of prwizard */
9394: /******************* Gompertz Likelihood ******************************/
9395: double gompertz(double x[])
9396: {
1.302 brouard 9397: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9398: int i,n=0; /* n is the size of the sample */
9399:
1.220 brouard 9400: for (i=1;i<=imx ; i++) {
1.126 brouard 9401: sump=sump+weight[i];
9402: /* sump=sump+1;*/
9403: num=num+1;
9404: }
1.302 brouard 9405: L=0.0;
9406: /* agegomp=AGEGOMP; */
1.126 brouard 9407: /* for (i=0; i<=imx; i++)
9408: 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]);*/
9409:
1.302 brouard 9410: for (i=1;i<=imx ; i++) {
9411: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9412: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9413: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9414: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9415: * +
9416: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9417: */
9418: if (wav[i] > 1 || agedc[i] < AGESUP) {
9419: if (cens[i] == 1){
9420: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9421: } else if (cens[i] == 0){
1.126 brouard 9422: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9423: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9424: } else
9425: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9426: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9427: L=L+A*weight[i];
1.126 brouard 9428: /* 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 9429: }
9430: }
1.126 brouard 9431:
1.302 brouard 9432: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9433:
9434: return -2*L*num/sump;
9435: }
9436:
1.136 brouard 9437: #ifdef GSL
9438: /******************* Gompertz_f Likelihood ******************************/
9439: double gompertz_f(const gsl_vector *v, void *params)
9440: {
1.302 brouard 9441: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9442: double *x= (double *) v->data;
9443: int i,n=0; /* n is the size of the sample */
9444:
9445: for (i=0;i<=imx-1 ; i++) {
9446: sump=sump+weight[i];
9447: /* sump=sump+1;*/
9448: num=num+1;
9449: }
9450:
9451:
9452: /* for (i=0; i<=imx; i++)
9453: 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]);*/
9454: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9455: for (i=1;i<=imx ; i++)
9456: {
9457: if (cens[i] == 1 && wav[i]>1)
9458: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9459:
9460: if (cens[i] == 0 && wav[i]>1)
9461: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9462: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9463:
9464: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9465: if (wav[i] > 1 ) { /* ??? */
9466: LL=LL+A*weight[i];
9467: /* 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]);*/
9468: }
9469: }
9470:
9471: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9472: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9473:
9474: return -2*LL*num/sump;
9475: }
9476: #endif
9477:
1.126 brouard 9478: /******************* Printing html file ***********/
1.201 brouard 9479: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9480: int lastpass, int stepm, int weightopt, char model[],\
9481: int imx, double p[],double **matcov,double agemortsup){
9482: int i,k;
9483:
9484: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9485: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9486: for (i=1;i<=2;i++)
9487: 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 9488: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9489: fprintf(fichtm,"</ul>");
9490:
9491: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9492:
9493: 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>");
9494:
9495: for (k=agegomp;k<(agemortsup-2);k++)
9496: 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]);
9497:
9498:
9499: fflush(fichtm);
9500: }
9501:
9502: /******************* Gnuplot file **************/
1.201 brouard 9503: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9504:
9505: char dirfileres[132],optfileres[132];
1.164 brouard 9506:
1.126 brouard 9507: int ng;
9508:
9509:
9510: /*#ifdef windows */
9511: fprintf(ficgp,"cd \"%s\" \n",pathc);
9512: /*#endif */
9513:
9514:
9515: strcpy(dirfileres,optionfilefiname);
9516: strcpy(optfileres,"vpl");
1.199 brouard 9517: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9518: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9519: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9520: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9521: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9522:
9523: }
9524:
1.136 brouard 9525: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9526: {
1.126 brouard 9527:
1.136 brouard 9528: /*-------- data file ----------*/
9529: FILE *fic;
9530: char dummy[]=" ";
1.240 brouard 9531: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9532: int lstra;
1.136 brouard 9533: int linei, month, year,iout;
1.302 brouard 9534: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9535: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9536: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9537: char *stratrunc;
1.223 brouard 9538:
1.240 brouard 9539: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9540: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9541:
1.240 brouard 9542: for(v=1; v <=ncovcol;v++){
9543: DummyV[v]=0;
9544: FixedV[v]=0;
9545: }
9546: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9547: DummyV[v]=1;
9548: FixedV[v]=0;
9549: }
9550: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9551: DummyV[v]=0;
9552: FixedV[v]=1;
9553: }
9554: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9555: DummyV[v]=1;
9556: FixedV[v]=1;
9557: }
9558: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9559: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9560: 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]);
9561: }
1.126 brouard 9562:
1.136 brouard 9563: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9564: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9565: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9566: }
1.126 brouard 9567:
1.302 brouard 9568: /* Is it a BOM UTF-8 Windows file? */
9569: /* First data line */
9570: linei=0;
9571: while(fgets(line, MAXLINE, fic)) {
9572: noffset=0;
9573: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9574: {
9575: noffset=noffset+3;
9576: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9577: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9578: fflush(ficlog); return 1;
9579: }
9580: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9581: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9582: {
9583: noffset=noffset+2;
1.304 brouard 9584: 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);
9585: 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 9586: fflush(ficlog); return 1;
9587: }
9588: else if( line[0] == 0 && line[1] == 0)
9589: {
9590: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9591: noffset=noffset+4;
1.304 brouard 9592: 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);
9593: 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 9594: fflush(ficlog); return 1;
9595: }
9596: } else{
9597: ;/*printf(" Not a BOM file\n");*/
9598: }
9599: /* If line starts with a # it is a comment */
9600: if (line[noffset] == '#') {
9601: linei=linei+1;
9602: break;
9603: }else{
9604: break;
9605: }
9606: }
9607: fclose(fic);
9608: if((fic=fopen(datafile,"r"))==NULL) {
9609: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9610: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9611: }
9612: /* Not a Bom file */
9613:
1.136 brouard 9614: i=1;
9615: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9616: linei=linei+1;
9617: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9618: if(line[j] == '\t')
9619: line[j] = ' ';
9620: }
9621: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9622: ;
9623: };
9624: line[j+1]=0; /* Trims blanks at end of line */
9625: if(line[0]=='#'){
9626: fprintf(ficlog,"Comment line\n%s\n",line);
9627: printf("Comment line\n%s\n",line);
9628: continue;
9629: }
9630: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9631: strcpy(line, linetmp);
1.223 brouard 9632:
9633: /* Loops on waves */
9634: for (j=maxwav;j>=1;j--){
9635: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9636: cutv(stra, strb, line, ' ');
9637: if(strb[0]=='.') { /* Missing value */
9638: lval=-1;
9639: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9640: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9641: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9642: 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);
9643: 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);
9644: return 1;
9645: }
9646: }else{
9647: errno=0;
9648: /* what_kind_of_number(strb); */
9649: dval=strtod(strb,&endptr);
9650: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9651: /* if(strb != endptr && *endptr == '\0') */
9652: /* dval=dlval; */
9653: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9654: if( strb[0]=='\0' || (*endptr != '\0')){
9655: 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);
9656: 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);
9657: return 1;
9658: }
9659: cotqvar[j][iv][i]=dval;
9660: cotvar[j][ntv+iv][i]=dval;
9661: }
9662: strcpy(line,stra);
1.223 brouard 9663: }/* end loop ntqv */
1.225 brouard 9664:
1.223 brouard 9665: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9666: cutv(stra, strb, line, ' ');
9667: if(strb[0]=='.') { /* Missing value */
9668: lval=-1;
9669: }else{
9670: errno=0;
9671: lval=strtol(strb,&endptr,10);
9672: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9673: if( strb[0]=='\0' || (*endptr != '\0')){
9674: 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);
9675: 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);
9676: return 1;
9677: }
9678: }
9679: if(lval <-1 || lval >1){
9680: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9681: 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 9682: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9683: For example, for multinomial values like 1, 2 and 3,\n \
9684: build V1=0 V2=0 for the reference value (1),\n \
9685: V1=1 V2=0 for (2) \n \
1.223 brouard 9686: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9687: output of IMaCh is often meaningless.\n \
1.319 brouard 9688: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 9689: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9690: 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 9691: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9692: For example, for multinomial values like 1, 2 and 3,\n \
9693: build V1=0 V2=0 for the reference value (1),\n \
9694: V1=1 V2=0 for (2) \n \
1.223 brouard 9695: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9696: output of IMaCh is often meaningless.\n \
1.319 brouard 9697: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 9698: return 1;
9699: }
9700: cotvar[j][iv][i]=(double)(lval);
9701: strcpy(line,stra);
1.223 brouard 9702: }/* end loop ntv */
1.225 brouard 9703:
1.223 brouard 9704: /* Statuses at wave */
1.137 brouard 9705: cutv(stra, strb, line, ' ');
1.223 brouard 9706: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9707: lval=-1;
1.136 brouard 9708: }else{
1.238 brouard 9709: errno=0;
9710: lval=strtol(strb,&endptr,10);
9711: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9712: if( strb[0]=='\0' || (*endptr != '\0')){
9713: 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);
9714: 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);
9715: return 1;
9716: }
1.136 brouard 9717: }
1.225 brouard 9718:
1.136 brouard 9719: s[j][i]=lval;
1.225 brouard 9720:
1.223 brouard 9721: /* Date of Interview */
1.136 brouard 9722: strcpy(line,stra);
9723: cutv(stra, strb,line,' ');
1.169 brouard 9724: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9725: }
1.169 brouard 9726: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9727: month=99;
9728: year=9999;
1.136 brouard 9729: }else{
1.225 brouard 9730: 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);
9731: 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);
9732: return 1;
1.136 brouard 9733: }
9734: anint[j][i]= (double) year;
1.302 brouard 9735: mint[j][i]= (double)month;
9736: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
9737: /* 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]); */
9738: /* 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]); */
9739: /* } */
1.136 brouard 9740: strcpy(line,stra);
1.223 brouard 9741: } /* End loop on waves */
1.225 brouard 9742:
1.223 brouard 9743: /* Date of death */
1.136 brouard 9744: cutv(stra, strb,line,' ');
1.169 brouard 9745: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9746: }
1.169 brouard 9747: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9748: month=99;
9749: year=9999;
9750: }else{
1.141 brouard 9751: 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 9752: 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);
9753: return 1;
1.136 brouard 9754: }
9755: andc[i]=(double) year;
9756: moisdc[i]=(double) month;
9757: strcpy(line,stra);
9758:
1.223 brouard 9759: /* Date of birth */
1.136 brouard 9760: cutv(stra, strb,line,' ');
1.169 brouard 9761: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9762: }
1.169 brouard 9763: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9764: month=99;
9765: year=9999;
9766: }else{
1.141 brouard 9767: 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);
9768: 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 9769: return 1;
1.136 brouard 9770: }
9771: if (year==9999) {
1.141 brouard 9772: 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);
9773: 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 9774: return 1;
9775:
1.136 brouard 9776: }
9777: annais[i]=(double)(year);
1.302 brouard 9778: moisnais[i]=(double)(month);
9779: for (j=1;j<=maxwav;j++){
9780: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
9781: 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]);
9782: 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]);
9783: }
9784: }
9785:
1.136 brouard 9786: strcpy(line,stra);
1.225 brouard 9787:
1.223 brouard 9788: /* Sample weight */
1.136 brouard 9789: cutv(stra, strb,line,' ');
9790: errno=0;
9791: dval=strtod(strb,&endptr);
9792: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9793: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9794: 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 9795: fflush(ficlog);
9796: return 1;
9797: }
9798: weight[i]=dval;
9799: strcpy(line,stra);
1.225 brouard 9800:
1.223 brouard 9801: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9802: cutv(stra, strb, line, ' ');
9803: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9804: lval=-1;
1.311 brouard 9805: coqvar[iv][i]=NAN;
9806: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 9807: }else{
1.225 brouard 9808: errno=0;
9809: /* what_kind_of_number(strb); */
9810: dval=strtod(strb,&endptr);
9811: /* if(strb != endptr && *endptr == '\0') */
9812: /* dval=dlval; */
9813: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9814: if( strb[0]=='\0' || (*endptr != '\0')){
9815: 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);
9816: 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);
9817: return 1;
9818: }
9819: coqvar[iv][i]=dval;
1.226 brouard 9820: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9821: }
9822: strcpy(line,stra);
9823: }/* end loop nqv */
1.136 brouard 9824:
1.223 brouard 9825: /* Covariate values */
1.136 brouard 9826: for (j=ncovcol;j>=1;j--){
9827: cutv(stra, strb,line,' ');
1.223 brouard 9828: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9829: lval=-1;
1.136 brouard 9830: }else{
1.225 brouard 9831: errno=0;
9832: lval=strtol(strb,&endptr,10);
9833: if( strb[0]=='\0' || (*endptr != '\0')){
9834: 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);
9835: 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);
9836: return 1;
9837: }
1.136 brouard 9838: }
9839: if(lval <-1 || lval >1){
1.225 brouard 9840: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9841: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9842: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9843: For example, for multinomial values like 1, 2 and 3,\n \
9844: build V1=0 V2=0 for the reference value (1),\n \
9845: V1=1 V2=0 for (2) \n \
1.136 brouard 9846: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9847: output of IMaCh is often meaningless.\n \
1.136 brouard 9848: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9849: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9850: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9851: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9852: For example, for multinomial values like 1, 2 and 3,\n \
9853: build V1=0 V2=0 for the reference value (1),\n \
9854: V1=1 V2=0 for (2) \n \
1.136 brouard 9855: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9856: output of IMaCh is often meaningless.\n \
1.136 brouard 9857: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9858: return 1;
1.136 brouard 9859: }
9860: covar[j][i]=(double)(lval);
9861: strcpy(line,stra);
9862: }
9863: lstra=strlen(stra);
1.225 brouard 9864:
1.136 brouard 9865: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9866: stratrunc = &(stra[lstra-9]);
9867: num[i]=atol(stratrunc);
9868: }
9869: else
9870: num[i]=atol(stra);
9871: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9872: 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;}*/
9873:
9874: i=i+1;
9875: } /* End loop reading data */
1.225 brouard 9876:
1.136 brouard 9877: *imax=i-1; /* Number of individuals */
9878: fclose(fic);
1.225 brouard 9879:
1.136 brouard 9880: return (0);
1.164 brouard 9881: /* endread: */
1.225 brouard 9882: printf("Exiting readdata: ");
9883: fclose(fic);
9884: return (1);
1.223 brouard 9885: }
1.126 brouard 9886:
1.234 brouard 9887: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9888: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9889: while (*p2 == ' ')
1.234 brouard 9890: p2++;
9891: /* while ((*p1++ = *p2++) !=0) */
9892: /* ; */
9893: /* do */
9894: /* while (*p2 == ' ') */
9895: /* p2++; */
9896: /* while (*p1++ == *p2++); */
9897: *stri=p2;
1.145 brouard 9898: }
9899:
1.235 brouard 9900: int decoderesult ( char resultline[], int nres)
1.230 brouard 9901: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9902: {
1.235 brouard 9903: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9904: char resultsav[MAXLINE];
1.234 brouard 9905: int resultmodel[MAXLINE];
9906: int modelresult[MAXLINE];
1.230 brouard 9907: char stra[80], strb[80], strc[80], strd[80],stre[80];
9908:
1.234 brouard 9909: removefirstspace(&resultline);
1.230 brouard 9910:
9911: if (strstr(resultline,"v") !=0){
9912: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9913: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9914: return 1;
9915: }
9916: trimbb(resultsav, resultline);
9917: if (strlen(resultsav) >1){
9918: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9919: }
1.253 brouard 9920: if(j == 0){ /* Resultline but no = */
9921: TKresult[nres]=0; /* Combination for the nresult and the model */
9922: return (0);
9923: }
1.234 brouard 9924: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.318 brouard 9925: printf("ERROR: the number of variables in this result line, %d, differs from the number of variables used in the model line, %d.\n",j, cptcovs);
1.310 brouard 9926: fprintf(ficlog,"ERROR: the number of variables in the resultline, %d, differs from the number of variables used in the model line, %d.\n",j, cptcovs);
1.234 brouard 9927: }
9928: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9929: if(nbocc(resultsav,'=') >1){
1.318 brouard 9930: 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" */
9931: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.234 brouard 9932: }else
9933: cutl(strc,strd,resultsav,'=');
1.318 brouard 9934: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 9935:
1.230 brouard 9936: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 9937: 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 9938: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9939: /* cptcovsel++; */
9940: if (nbocc(stra,'=') >0)
9941: strcpy(resultsav,stra); /* and analyzes it */
9942: }
1.235 brouard 9943: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 9944: for(k1=1; k1<= cptcovt ;k1++){ /* Loop on model. model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9945: if(Typevar[k1]==0){ /* Single covariate in model *//*0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 9946: match=0;
1.318 brouard 9947: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9948: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9949: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 9950: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 9951: break;
9952: }
9953: }
9954: if(match == 0){
1.310 brouard 9955: printf("Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9956: fprintf(ficlog,"Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9957: return 1;
1.234 brouard 9958: }
9959: }
9960: }
1.235 brouard 9961: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 9962: for(k2=1; k2 <=j;k2++){ /* Loop on resultline variables: result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9963: match=0;
1.318 brouard 9964: for(k1=1; k1<= cptcovt ;k1++){ /* loop on model: model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.235 brouard 9965: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9966: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.318 brouard 9967: resultmodel[k1]=k2; /* k2th variable of the model corresponds to k1 variable of the model. resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9968: ++match;
9969: }
9970: }
9971: }
9972: if(match == 0){
9973: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
1.310 brouard 9974: fprintf(ficlog,"Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9975: return 1;
1.234 brouard 9976: }else if(match > 1){
9977: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 9978: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9979: return 1;
1.234 brouard 9980: }
9981: }
1.235 brouard 9982:
1.234 brouard 9983: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9984: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9985: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9986: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9987: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9988: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9989: /* 1 0 0 0 */
9990: /* 2 1 0 0 */
9991: /* 3 0 1 0 */
9992: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9993: /* 5 0 0 1 */
9994: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9995: /* 7 0 1 1 */
9996: /* 8 1 1 1 */
1.237 brouard 9997: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9998: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9999: /* V5*age V5 known which value for nres? */
10000: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.318 brouard 10001: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* loop on model line */
1.235 brouard 10002: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 10003: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 10004: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
10005: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 10006: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
10007: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10008: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 10009: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
10010: k4++;;
10011: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
1.318 brouard 10012: k3q= resultmodel[k1]; /* resultmodel[1(V5)] = 25.1=k3q */
10013: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.237 brouard 10014: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10015: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
10016: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 10017: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
10018: k4q++;;
10019: }
10020: }
1.234 brouard 10021:
1.235 brouard 10022: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 10023: return (0);
10024: }
1.235 brouard 10025:
1.230 brouard 10026: int decodemodel( char model[], int lastobs)
10027: /**< This routine decodes the model and returns:
1.224 brouard 10028: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10029: * - nagesqr = 1 if age*age in the model, otherwise 0.
10030: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10031: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10032: * - cptcovage number of covariates with age*products =2
10033: * - cptcovs number of simple covariates
10034: * - 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
10035: * which is a new column after the 9 (ncovcol) variables.
1.319 brouard 10036: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10037: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10038: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10039: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10040: */
1.319 brouard 10041: /* 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 10042: {
1.238 brouard 10043: int i, j, k, ks, v;
1.227 brouard 10044: int j1, k1, k2, k3, k4;
1.136 brouard 10045: char modelsav[80];
1.145 brouard 10046: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10047: char *strpt;
1.136 brouard 10048:
1.145 brouard 10049: /*removespace(model);*/
1.136 brouard 10050: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10051: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10052: if (strstr(model,"AGE") !=0){
1.192 brouard 10053: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10054: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10055: return 1;
10056: }
1.141 brouard 10057: if (strstr(model,"v") !=0){
10058: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
10059: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
10060: return 1;
10061: }
1.187 brouard 10062: strcpy(modelsav,model);
10063: if ((strpt=strstr(model,"age*age")) !=0){
10064: printf(" strpt=%s, model=%s\n",strpt, model);
10065: if(strpt != model){
1.234 brouard 10066: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10067: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10068: corresponding column of parameters.\n",model);
1.234 brouard 10069: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10070: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10071: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10072: return 1;
1.225 brouard 10073: }
1.187 brouard 10074: nagesqr=1;
10075: if (strstr(model,"+age*age") !=0)
1.234 brouard 10076: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10077: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10078: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10079: else
1.234 brouard 10080: substrchaine(modelsav, model, "age*age");
1.187 brouard 10081: }else
10082: nagesqr=0;
10083: if (strlen(modelsav) >1){
10084: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10085: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10086: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10087: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10088: * cst, age and age*age
10089: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10090: /* including age products which are counted in cptcovage.
10091: * but the covariates which are products must be treated
10092: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10093: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10094: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10095:
10096:
1.187 brouard 10097: /* Design
10098: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10099: * < ncovcol=8 >
10100: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10101: * k= 1 2 3 4 5 6 7 8
10102: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10103: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10104: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10105: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10106: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10107: * Tage[++cptcovage]=k
10108: * if products, new covar are created after ncovcol with k1
10109: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10110: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10111: * 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
10112: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10113: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10114: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10115: * < ncovcol=8 >
10116: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10117: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10118: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10119: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10120: * p Tprod[1]@2={ 6, 5}
10121: *p Tvard[1][1]@4= {7, 8, 5, 6}
10122: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10123: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10124: *How to reorganize? Tvars(orted)
1.187 brouard 10125: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10126: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
10127: * {2, 1, 4, 8, 5, 6, 3, 7}
10128: * Struct []
10129: */
1.225 brouard 10130:
1.187 brouard 10131: /* This loop fills the array Tvar from the string 'model'.*/
10132: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
10133: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
10134: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
10135: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
10136: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
10137: /* k=1 Tvar[1]=2 (from V2) */
10138: /* k=5 Tvar[5] */
10139: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 10140: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 10141: /* } */
1.198 brouard 10142: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 10143: /*
10144: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 10145: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
10146: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
10147: }
1.187 brouard 10148: cptcovage=0;
1.319 brouard 10149: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
10150: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
10151: 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" */
10152: if (nbocc(modelsav,'+')==0)
10153: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 10154: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
10155: /*scanf("%d",i);*/
1.319 brouard 10156: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
10157: 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 10158: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
10159: /* covar is not filled and then is empty */
10160: cptcovprod--;
10161: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 10162: 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 10163: Typevar[k]=1; /* 1 for age product */
1.319 brouard 10164: cptcovage++; /* Counts the number of covariates which include age as a product */
10165: 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 10166: /*printf("stre=%s ", stre);*/
10167: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
10168: cptcovprod--;
10169: cutl(stre,strb,strc,'V');
10170: Tvar[k]=atoi(stre);
10171: Typevar[k]=1; /* 1 for age product */
10172: cptcovage++;
10173: Tage[cptcovage]=k;
10174: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
10175: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
10176: cptcovn++;
10177: cptcovprodnoage++;k1++;
10178: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10179: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10180: because this model-covariate is a construction we invent a new column
10181: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.319 brouard 10182: If already ncovcol=4 and model=V2 + V1 +V1*V4 +age*V3 +V3*V2
10183: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
10184: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=4 etc */
1.234 brouard 10185: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10186: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10187: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 10188: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 10189: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
10190: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
10191: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10192: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10193: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10194: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10195: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10196: for (i=1; i<=lastobs;i++){
10197: /* Computes the new covariate which is a product of
10198: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10199: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10200: }
10201: } /* End age is not in the model */
10202: } /* End if model includes a product */
1.319 brouard 10203: else { /* not a product */
1.234 brouard 10204: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10205: /* scanf("%d",i);*/
10206: cutl(strd,strc,strb,'V');
10207: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10208: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10209: Tvar[k]=atoi(strd);
10210: Typevar[k]=0; /* 0 for simple covariates */
10211: }
10212: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10213: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10214: scanf("%d",i);*/
1.187 brouard 10215: } /* end of loop + on total covariates */
10216: } /* end if strlen(modelsave == 0) age*age might exist */
10217: } /* end if strlen(model == 0) */
1.136 brouard 10218:
10219: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10220: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10221:
1.136 brouard 10222: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10223: printf("cptcovprod=%d ", cptcovprod);
10224: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10225: scanf("%d ",i);*/
10226:
10227:
1.230 brouard 10228: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10229: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10230: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10231: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10232: k = 1 2 3 4 5 6 7 8 9
10233: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 10234: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 10235: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10236: Dummy[k] 1 0 0 0 3 1 1 2 3
10237: Tmodelind[combination of covar]=k;
1.225 brouard 10238: */
10239: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10240: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10241: /* 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 10242: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 10243: printf("Model=1+age+%s\n\
1.227 brouard 10244: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10245: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10246: 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 10247: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 10248: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10249: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10250: Dummy[k] 0=dummy (0 1), 1 quantitative (single or product without age), 2 dummy with age product, 3 quant with age product\n",model);
1.285 brouard 10251: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10252: for(k=1, ncovf=0, nsd=0, nsq=0, ncovv=0, ncova=0, ncoveff=0, nqfveff=0, ntveff=0, nqtveff=0;k<=cptcovt; k++){ /* or cptocvt */
10253: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10254: Fixed[k]= 0;
10255: Dummy[k]= 0;
1.225 brouard 10256: ncoveff++;
1.232 brouard 10257: ncovf++;
1.234 brouard 10258: nsd++;
10259: modell[k].maintype= FTYPE;
10260: TvarsD[nsd]=Tvar[k];
10261: TvarsDind[nsd]=k;
10262: TvarF[ncovf]=Tvar[k];
10263: TvarFind[ncovf]=k;
10264: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10265: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10266: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10267: Fixed[k]= 0;
10268: Dummy[k]= 0;
10269: ncoveff++;
10270: ncovf++;
10271: modell[k].maintype= FTYPE;
10272: TvarF[ncovf]=Tvar[k];
10273: TvarFind[ncovf]=k;
1.230 brouard 10274: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10275: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10276: }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 10277: Fixed[k]= 0;
10278: Dummy[k]= 1;
1.230 brouard 10279: nqfveff++;
1.234 brouard 10280: modell[k].maintype= FTYPE;
10281: modell[k].subtype= FQ;
10282: nsq++;
10283: TvarsQ[nsq]=Tvar[k];
10284: TvarsQind[nsq]=k;
1.232 brouard 10285: ncovf++;
1.234 brouard 10286: TvarF[ncovf]=Tvar[k];
10287: TvarFind[ncovf]=k;
1.231 brouard 10288: 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 10289: 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 10290: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10291: Fixed[k]= 1;
10292: Dummy[k]= 0;
1.225 brouard 10293: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10294: modell[k].maintype= VTYPE;
10295: modell[k].subtype= VD;
10296: nsd++;
10297: TvarsD[nsd]=Tvar[k];
10298: TvarsDind[nsd]=k;
10299: ncovv++; /* Only simple time varying variables */
10300: TvarV[ncovv]=Tvar[k];
1.242 brouard 10301: 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 10302: 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 */
10303: 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 10304: 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);
10305: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10306: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10307: Fixed[k]= 1;
10308: Dummy[k]= 1;
10309: nqtveff++;
10310: modell[k].maintype= VTYPE;
10311: modell[k].subtype= VQ;
10312: ncovv++; /* Only simple time varying variables */
10313: nsq++;
1.319 brouard 10314: TvarsQ[nsq]=Tvar[k]; /* k=1 Tvar=5 nsq=1 TvarsQ[1]=5 */
1.234 brouard 10315: TvarsQind[nsq]=k;
10316: TvarV[ncovv]=Tvar[k];
1.242 brouard 10317: 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 10318: 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 */
10319: 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 10320: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10321: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10322: printf("Quasi TmodelQind[%d]=%d,Tvar[TmodelQind[%d]]=V%d, ncovcol=%d, nqv=%d, ntv=%d,Tvar[k]- ncovcol-nqv-ntv=%d\n",nqtveff,k,nqtveff,Tvar[k], ncovcol, nqv, ntv, Tvar[k]- ncovcol-nqv-ntv);
1.228 brouard 10323: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10324: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10325: ncova++;
10326: TvarA[ncova]=Tvar[k];
10327: TvarAind[ncova]=k;
1.231 brouard 10328: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10329: Fixed[k]= 2;
10330: Dummy[k]= 2;
10331: modell[k].maintype= ATYPE;
10332: modell[k].subtype= APFD;
10333: /* ncoveff++; */
1.227 brouard 10334: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10335: Fixed[k]= 2;
10336: Dummy[k]= 3;
10337: modell[k].maintype= ATYPE;
10338: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10339: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10340: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10341: Fixed[k]= 3;
10342: Dummy[k]= 2;
10343: modell[k].maintype= ATYPE;
10344: modell[k].subtype= APVD; /* Product age * varying dummy */
10345: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10346: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10347: Fixed[k]= 3;
10348: Dummy[k]= 3;
10349: modell[k].maintype= ATYPE;
10350: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10351: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10352: }
10353: }else if (Typevar[k] == 2) { /* product without age */
10354: k1=Tposprod[k];
10355: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10356: if(Tvard[k1][2] <=ncovcol){
10357: Fixed[k]= 1;
10358: Dummy[k]= 0;
10359: modell[k].maintype= FTYPE;
10360: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10361: ncovf++; /* Fixed variables without age */
10362: TvarF[ncovf]=Tvar[k];
10363: TvarFind[ncovf]=k;
10364: }else if(Tvard[k1][2] <=ncovcol+nqv){
10365: Fixed[k]= 0; /* or 2 ?*/
10366: Dummy[k]= 1;
10367: modell[k].maintype= FTYPE;
10368: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10369: ncovf++; /* Varying variables without age */
10370: TvarF[ncovf]=Tvar[k];
10371: TvarFind[ncovf]=k;
10372: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10373: Fixed[k]= 1;
10374: Dummy[k]= 0;
10375: modell[k].maintype= VTYPE;
10376: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10377: ncovv++; /* Varying variables without age */
10378: TvarV[ncovv]=Tvar[k];
10379: TvarVind[ncovv]=k;
10380: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10381: Fixed[k]= 1;
10382: Dummy[k]= 1;
10383: modell[k].maintype= VTYPE;
10384: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10385: ncovv++; /* Varying variables without age */
10386: TvarV[ncovv]=Tvar[k];
10387: TvarVind[ncovv]=k;
10388: }
1.227 brouard 10389: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10390: if(Tvard[k1][2] <=ncovcol){
10391: Fixed[k]= 0; /* or 2 ?*/
10392: Dummy[k]= 1;
10393: modell[k].maintype= FTYPE;
10394: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10395: ncovf++; /* Fixed variables without age */
10396: TvarF[ncovf]=Tvar[k];
10397: TvarFind[ncovf]=k;
10398: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10399: Fixed[k]= 1;
10400: Dummy[k]= 1;
10401: modell[k].maintype= VTYPE;
10402: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10403: ncovv++; /* Varying variables without age */
10404: TvarV[ncovv]=Tvar[k];
10405: TvarVind[ncovv]=k;
10406: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10407: Fixed[k]= 1;
10408: Dummy[k]= 1;
10409: modell[k].maintype= VTYPE;
10410: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10411: ncovv++; /* Varying variables without age */
10412: TvarV[ncovv]=Tvar[k];
10413: TvarVind[ncovv]=k;
10414: ncovv++; /* Varying variables without age */
10415: TvarV[ncovv]=Tvar[k];
10416: TvarVind[ncovv]=k;
10417: }
1.227 brouard 10418: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10419: if(Tvard[k1][2] <=ncovcol){
10420: Fixed[k]= 1;
10421: Dummy[k]= 1;
10422: modell[k].maintype= VTYPE;
10423: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10424: ncovv++; /* Varying variables without age */
10425: TvarV[ncovv]=Tvar[k];
10426: TvarVind[ncovv]=k;
10427: }else if(Tvard[k1][2] <=ncovcol+nqv){
10428: Fixed[k]= 1;
10429: Dummy[k]= 1;
10430: modell[k].maintype= VTYPE;
10431: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10432: ncovv++; /* Varying variables without age */
10433: TvarV[ncovv]=Tvar[k];
10434: TvarVind[ncovv]=k;
10435: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10436: Fixed[k]= 1;
10437: Dummy[k]= 0;
10438: modell[k].maintype= VTYPE;
10439: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10440: ncovv++; /* Varying variables without age */
10441: TvarV[ncovv]=Tvar[k];
10442: TvarVind[ncovv]=k;
10443: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10444: Fixed[k]= 1;
10445: Dummy[k]= 1;
10446: modell[k].maintype= VTYPE;
10447: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10448: ncovv++; /* Varying variables without age */
10449: TvarV[ncovv]=Tvar[k];
10450: TvarVind[ncovv]=k;
10451: }
1.227 brouard 10452: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10453: if(Tvard[k1][2] <=ncovcol){
10454: Fixed[k]= 1;
10455: Dummy[k]= 1;
10456: modell[k].maintype= VTYPE;
10457: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10458: ncovv++; /* Varying variables without age */
10459: TvarV[ncovv]=Tvar[k];
10460: TvarVind[ncovv]=k;
10461: }else if(Tvard[k1][2] <=ncovcol+nqv){
10462: Fixed[k]= 1;
10463: Dummy[k]= 1;
10464: modell[k].maintype= VTYPE;
10465: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10466: ncovv++; /* Varying variables without age */
10467: TvarV[ncovv]=Tvar[k];
10468: TvarVind[ncovv]=k;
10469: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10470: Fixed[k]= 1;
10471: Dummy[k]= 1;
10472: modell[k].maintype= VTYPE;
10473: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10474: ncovv++; /* Varying variables without age */
10475: TvarV[ncovv]=Tvar[k];
10476: TvarVind[ncovv]=k;
10477: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10478: Fixed[k]= 1;
10479: Dummy[k]= 1;
10480: modell[k].maintype= VTYPE;
10481: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10482: ncovv++; /* Varying variables without age */
10483: TvarV[ncovv]=Tvar[k];
10484: TvarVind[ncovv]=k;
10485: }
1.227 brouard 10486: }else{
1.240 brouard 10487: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10488: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10489: } /*end k1*/
1.225 brouard 10490: }else{
1.226 brouard 10491: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10492: 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 10493: }
1.227 brouard 10494: printf("Decodemodel, k=%d, Tvar[%d]=V%d,Typevar=%d, Fixed=%d, Dummy=%d\n",k, k,Tvar[k],Typevar[k],Fixed[k],Dummy[k]);
1.231 brouard 10495: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10496: 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]);
10497: }
10498: /* Searching for doublons in the model */
10499: for(k1=1; k1<= cptcovt;k1++){
10500: for(k2=1; k2 <k1;k2++){
1.285 brouard 10501: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10502: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10503: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10504: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10505: printf("Error duplication in the model=%s at positions (+) %d and %d, Tvar[%d]=V%d, Tvar[%d]=V%d, Typevar=%d, Fixed=%d, Dummy=%d\n", model, k1,k2, k1, Tvar[k1], k2, Tvar[k2],Typevar[k1],Fixed[k1],Dummy[k1]);
10506: fprintf(ficlog,"Error duplication in the model=%s at positions (+) %d and %d, Tvar[%d]=V%d, Tvar[%d]=V%d, Typevar=%d, Fixed=%d, Dummy=%d\n", model, k1,k2, k1, Tvar[k1], k2, Tvar[k2],Typevar[k1],Fixed[k1],Dummy[k1]); fflush(ficlog);
1.234 brouard 10507: return(1);
10508: }
10509: }else if (Typevar[k1] ==2){
10510: k3=Tposprod[k1];
10511: k4=Tposprod[k2];
10512: 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])) ){
10513: printf("Error duplication in the model=%s at positions (+) %d and %d, V%d*V%d, Typevar=%d, Fixed=%d, Dummy=%d\n",model, k1,k2, Tvard[k3][1], Tvard[k3][2],Typevar[k1],Fixed[Tvar[k1]],Dummy[Tvar[k1]]);
10514: fprintf(ficlog,"Error duplication in the model=%s at positions (+) %d and %d, V%d*V%d, Typevar=%d, Fixed=%d, Dummy=%d\n",model, k1,k2, Tvard[k3][1], Tvard[k3][2],Typevar[k1],Fixed[Tvar[k1]],Dummy[Tvar[k1]]); fflush(ficlog);
10515: return(1);
10516: }
10517: }
1.227 brouard 10518: }
10519: }
1.225 brouard 10520: }
10521: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10522: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10523: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10524: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10525: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10526: /*endread:*/
1.225 brouard 10527: printf("Exiting decodemodel: ");
10528: return (1);
1.136 brouard 10529: }
10530:
1.169 brouard 10531: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10532: {/* Check ages at death */
1.136 brouard 10533: int i, m;
1.218 brouard 10534: int firstone=0;
10535:
1.136 brouard 10536: for (i=1; i<=imx; i++) {
10537: for(m=2; (m<= maxwav); m++) {
10538: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10539: anint[m][i]=9999;
1.216 brouard 10540: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10541: s[m][i]=-1;
1.136 brouard 10542: }
10543: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10544: *nberr = *nberr + 1;
1.218 brouard 10545: if(firstone == 0){
10546: firstone=1;
1.260 brouard 10547: 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 10548: }
1.262 brouard 10549: 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 10550: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10551: }
10552: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10553: (*nberr)++;
1.259 brouard 10554: 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 10555: 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 10556: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10557: }
10558: }
10559: }
10560:
10561: for (i=1; i<=imx; i++) {
10562: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10563: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10564: 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 10565: if (s[m][i] >= nlstate+1) {
1.169 brouard 10566: if(agedc[i]>0){
10567: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10568: agev[m][i]=agedc[i];
1.214 brouard 10569: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10570: }else {
1.136 brouard 10571: if ((int)andc[i]!=9999){
10572: nbwarn++;
10573: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10574: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10575: agev[m][i]=-1;
10576: }
10577: }
1.169 brouard 10578: } /* agedc > 0 */
1.214 brouard 10579: } /* end if */
1.136 brouard 10580: else if(s[m][i] !=9){ /* Standard case, age in fractional
10581: years but with the precision of a month */
10582: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10583: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10584: agev[m][i]=1;
10585: else if(agev[m][i] < *agemin){
10586: *agemin=agev[m][i];
10587: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10588: }
10589: else if(agev[m][i] >*agemax){
10590: *agemax=agev[m][i];
1.156 brouard 10591: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10592: }
10593: /*agev[m][i]=anint[m][i]-annais[i];*/
10594: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10595: } /* en if 9*/
1.136 brouard 10596: else { /* =9 */
1.214 brouard 10597: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10598: agev[m][i]=1;
10599: s[m][i]=-1;
10600: }
10601: }
1.214 brouard 10602: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10603: agev[m][i]=1;
1.214 brouard 10604: else{
10605: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10606: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10607: agev[m][i]=0;
10608: }
10609: } /* End for lastpass */
10610: }
1.136 brouard 10611:
10612: for (i=1; i<=imx; i++) {
10613: for(m=firstpass; (m<=lastpass); m++){
10614: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10615: (*nberr)++;
1.136 brouard 10616: 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);
10617: 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);
10618: return 1;
10619: }
10620: }
10621: }
10622:
10623: /*for (i=1; i<=imx; i++){
10624: for (m=firstpass; (m<lastpass); m++){
10625: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10626: }
10627:
10628: }*/
10629:
10630:
1.139 brouard 10631: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10632: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10633:
10634: return (0);
1.164 brouard 10635: /* endread:*/
1.136 brouard 10636: printf("Exiting calandcheckages: ");
10637: return (1);
10638: }
10639:
1.172 brouard 10640: #if defined(_MSC_VER)
10641: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10642: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10643: //#include "stdafx.h"
10644: //#include <stdio.h>
10645: //#include <tchar.h>
10646: //#include <windows.h>
10647: //#include <iostream>
10648: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10649:
10650: LPFN_ISWOW64PROCESS fnIsWow64Process;
10651:
10652: BOOL IsWow64()
10653: {
10654: BOOL bIsWow64 = FALSE;
10655:
10656: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10657: // (HANDLE, PBOOL);
10658:
10659: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10660:
10661: HMODULE module = GetModuleHandle(_T("kernel32"));
10662: const char funcName[] = "IsWow64Process";
10663: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10664: GetProcAddress(module, funcName);
10665:
10666: if (NULL != fnIsWow64Process)
10667: {
10668: if (!fnIsWow64Process(GetCurrentProcess(),
10669: &bIsWow64))
10670: //throw std::exception("Unknown error");
10671: printf("Unknown error\n");
10672: }
10673: return bIsWow64 != FALSE;
10674: }
10675: #endif
1.177 brouard 10676:
1.191 brouard 10677: void syscompilerinfo(int logged)
1.292 brouard 10678: {
10679: #include <stdint.h>
10680:
10681: /* #include "syscompilerinfo.h"*/
1.185 brouard 10682: /* command line Intel compiler 32bit windows, XP compatible:*/
10683: /* /GS /W3 /Gy
10684: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10685: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10686: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10687: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10688: */
10689: /* 64 bits */
1.185 brouard 10690: /*
10691: /GS /W3 /Gy
10692: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10693: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10694: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10695: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10696: /* Optimization are useless and O3 is slower than O2 */
10697: /*
10698: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10699: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10700: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10701: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10702: */
1.186 brouard 10703: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10704: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10705: /PDB:"visual studio
10706: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10707: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10708: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10709: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10710: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10711: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10712: uiAccess='false'"
10713: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10714: /NOLOGO /TLBID:1
10715: */
1.292 brouard 10716:
10717:
1.177 brouard 10718: #if defined __INTEL_COMPILER
1.178 brouard 10719: #if defined(__GNUC__)
10720: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10721: #endif
1.177 brouard 10722: #elif defined(__GNUC__)
1.179 brouard 10723: #ifndef __APPLE__
1.174 brouard 10724: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10725: #endif
1.177 brouard 10726: struct utsname sysInfo;
1.178 brouard 10727: int cross = CROSS;
10728: if (cross){
10729: printf("Cross-");
1.191 brouard 10730: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10731: }
1.174 brouard 10732: #endif
10733:
1.191 brouard 10734: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10735: #if defined(__clang__)
1.191 brouard 10736: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10737: #endif
10738: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10739: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10740: #endif
10741: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10742: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10743: #endif
10744: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10745: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10746: #endif
10747: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10748: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10749: #endif
10750: #if defined(_MSC_VER)
1.191 brouard 10751: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10752: #endif
10753: #if defined(__PGI)
1.191 brouard 10754: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10755: #endif
10756: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10757: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10758: #endif
1.191 brouard 10759: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10760:
1.167 brouard 10761: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10762: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10763: // Windows (x64 and x86)
1.191 brouard 10764: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10765: #elif __unix__ // all unices, not all compilers
10766: // Unix
1.191 brouard 10767: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10768: #elif __linux__
10769: // linux
1.191 brouard 10770: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10771: #elif __APPLE__
1.174 brouard 10772: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10773: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10774: #endif
10775:
10776: /* __MINGW32__ */
10777: /* __CYGWIN__ */
10778: /* __MINGW64__ */
10779: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10780: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10781: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10782: /* _WIN64 // Defined for applications for Win64. */
10783: /* _M_X64 // Defined for compilations that target x64 processors. */
10784: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10785:
1.167 brouard 10786: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10787: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10788: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10789: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10790: #else
1.191 brouard 10791: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10792: #endif
10793:
1.169 brouard 10794: #if defined(__GNUC__)
10795: # if defined(__GNUC_PATCHLEVEL__)
10796: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10797: + __GNUC_MINOR__ * 100 \
10798: + __GNUC_PATCHLEVEL__)
10799: # else
10800: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10801: + __GNUC_MINOR__ * 100)
10802: # endif
1.174 brouard 10803: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10804: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10805:
10806: if (uname(&sysInfo) != -1) {
10807: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10808: 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 10809: }
10810: else
10811: perror("uname() error");
1.179 brouard 10812: //#ifndef __INTEL_COMPILER
10813: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10814: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10815: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10816: #endif
1.169 brouard 10817: #endif
1.172 brouard 10818:
1.286 brouard 10819: // void main ()
1.172 brouard 10820: // {
1.169 brouard 10821: #if defined(_MSC_VER)
1.174 brouard 10822: if (IsWow64()){
1.191 brouard 10823: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10824: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10825: }
10826: else{
1.191 brouard 10827: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10828: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10829: }
1.172 brouard 10830: // printf("\nPress Enter to continue...");
10831: // getchar();
10832: // }
10833:
1.169 brouard 10834: #endif
10835:
1.167 brouard 10836:
1.219 brouard 10837: }
1.136 brouard 10838:
1.219 brouard 10839: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10840: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10841: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10842: /* double ftolpl = 1.e-10; */
1.180 brouard 10843: double age, agebase, agelim;
1.203 brouard 10844: double tot;
1.180 brouard 10845:
1.202 brouard 10846: strcpy(filerespl,"PL_");
10847: strcat(filerespl,fileresu);
10848: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10849: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10850: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10851: }
1.288 brouard 10852: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10853: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10854: pstamp(ficrespl);
1.288 brouard 10855: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10856: fprintf(ficrespl,"#Age ");
10857: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10858: fprintf(ficrespl,"\n");
1.180 brouard 10859:
1.219 brouard 10860: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10861:
1.219 brouard 10862: agebase=ageminpar;
10863: agelim=agemaxpar;
1.180 brouard 10864:
1.227 brouard 10865: /* i1=pow(2,ncoveff); */
1.234 brouard 10866: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10867: if (cptcovn < 1){i1=1;}
1.180 brouard 10868:
1.238 brouard 10869: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10870: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10871: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10872: continue;
1.235 brouard 10873:
1.238 brouard 10874: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10875: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10876: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10877: /* k=k+1; */
10878: /* to clean */
10879: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10880: fprintf(ficrespl,"#******");
10881: printf("#******");
10882: fprintf(ficlog,"#******");
10883: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10884: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10885: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10886: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10887: }
10888: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10889: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10890: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10891: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10892: }
10893: fprintf(ficrespl,"******\n");
10894: printf("******\n");
10895: fprintf(ficlog,"******\n");
10896: if(invalidvarcomb[k]){
10897: printf("\nCombination (%d) ignored because no case \n",k);
10898: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10899: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10900: continue;
10901: }
1.219 brouard 10902:
1.238 brouard 10903: fprintf(ficrespl,"#Age ");
10904: for(j=1;j<=cptcoveff;j++) {
10905: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10906: }
10907: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10908: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10909:
1.238 brouard 10910: for (age=agebase; age<=agelim; age++){
10911: /* for (age=agebase; age<=agebase; age++){ */
10912: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10913: fprintf(ficrespl,"%.0f ",age );
10914: for(j=1;j<=cptcoveff;j++)
10915: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10916: tot=0.;
10917: for(i=1; i<=nlstate;i++){
10918: tot += prlim[i][i];
10919: fprintf(ficrespl," %.5f", prlim[i][i]);
10920: }
10921: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10922: } /* Age */
10923: /* was end of cptcod */
10924: } /* cptcov */
10925: } /* nres */
1.219 brouard 10926: return 0;
1.180 brouard 10927: }
10928:
1.218 brouard 10929: 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 10930: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10931:
10932: /* Computes the back prevalence limit for any combination of covariate values
10933: * at any age between ageminpar and agemaxpar
10934: */
1.235 brouard 10935: int i, j, k, i1, nres=0 ;
1.217 brouard 10936: /* double ftolpl = 1.e-10; */
10937: double age, agebase, agelim;
10938: double tot;
1.218 brouard 10939: /* double ***mobaverage; */
10940: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10941:
10942: strcpy(fileresplb,"PLB_");
10943: strcat(fileresplb,fileresu);
10944: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10945: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10946: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10947: }
1.288 brouard 10948: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10949: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10950: pstamp(ficresplb);
1.288 brouard 10951: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10952: fprintf(ficresplb,"#Age ");
10953: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10954: fprintf(ficresplb,"\n");
10955:
1.218 brouard 10956:
10957: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10958:
10959: agebase=ageminpar;
10960: agelim=agemaxpar;
10961:
10962:
1.227 brouard 10963: i1=pow(2,cptcoveff);
1.218 brouard 10964: if (cptcovn < 1){i1=1;}
1.227 brouard 10965:
1.238 brouard 10966: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10967: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10968: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10969: continue;
10970: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10971: fprintf(ficresplb,"#******");
10972: printf("#******");
10973: fprintf(ficlog,"#******");
10974: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10975: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10976: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10977: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10978: }
10979: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10980: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10981: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10982: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10983: }
10984: fprintf(ficresplb,"******\n");
10985: printf("******\n");
10986: fprintf(ficlog,"******\n");
10987: if(invalidvarcomb[k]){
10988: printf("\nCombination (%d) ignored because no cases \n",k);
10989: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10990: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10991: continue;
10992: }
1.218 brouard 10993:
1.238 brouard 10994: fprintf(ficresplb,"#Age ");
10995: for(j=1;j<=cptcoveff;j++) {
10996: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10997: }
10998: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10999: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 11000:
11001:
1.238 brouard 11002: for (age=agebase; age<=agelim; age++){
11003: /* for (age=agebase; age<=agebase; age++){ */
11004: if(mobilavproj > 0){
11005: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11006: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11007: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11008: }else if (mobilavproj == 0){
11009: 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);
11010: 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);
11011: exit(1);
11012: }else{
11013: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11014: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11015: /* printf("TOTOT\n"); */
11016: /* exit(1); */
1.238 brouard 11017: }
11018: fprintf(ficresplb,"%.0f ",age );
11019: for(j=1;j<=cptcoveff;j++)
11020: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11021: tot=0.;
11022: for(i=1; i<=nlstate;i++){
11023: tot += bprlim[i][i];
11024: fprintf(ficresplb," %.5f", bprlim[i][i]);
11025: }
11026: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11027: } /* Age */
11028: /* was end of cptcod */
1.255 brouard 11029: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 11030: } /* end of any combination */
11031: } /* end of nres */
1.218 brouard 11032: /* hBijx(p, bage, fage); */
11033: /* fclose(ficrespijb); */
11034:
11035: return 0;
1.217 brouard 11036: }
1.218 brouard 11037:
1.180 brouard 11038: int hPijx(double *p, int bage, int fage){
11039: /*------------- h Pij x at various ages ------------*/
11040:
11041: int stepsize;
11042: int agelim;
11043: int hstepm;
11044: int nhstepm;
1.235 brouard 11045: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11046:
11047: double agedeb;
11048: double ***p3mat;
11049:
1.201 brouard 11050: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 11051: if((ficrespij=fopen(filerespij,"w"))==NULL) {
11052: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
11053: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
11054: }
11055: printf("Computing pij: result on file '%s' \n", filerespij);
11056: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
11057:
11058: stepsize=(int) (stepm+YEARM-1)/YEARM;
11059: /*if (stepm<=24) stepsize=2;*/
11060:
11061: agelim=AGESUP;
11062: hstepm=stepsize*YEARM; /* Every year of age */
11063: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 11064:
1.180 brouard 11065: /* hstepm=1; aff par mois*/
11066: pstamp(ficrespij);
11067: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 11068: i1= pow(2,cptcoveff);
1.218 brouard 11069: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11070: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11071: /* k=k+1; */
1.235 brouard 11072: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11073: for(k=1; k<=i1;k++){
1.253 brouard 11074: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11075: continue;
1.183 brouard 11076: fprintf(ficrespij,"\n#****** ");
1.227 brouard 11077: for(j=1;j<=cptcoveff;j++)
1.198 brouard 11078: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11079: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11080: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11081: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11082: }
1.183 brouard 11083: fprintf(ficrespij,"******\n");
11084:
11085: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
11086: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
11087: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
11088:
11089: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 11090:
1.183 brouard 11091: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11092: oldm=oldms;savm=savms;
1.235 brouard 11093: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 11094: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
11095: for(i=1; i<=nlstate;i++)
11096: for(j=1; j<=nlstate+ndeath;j++)
11097: fprintf(ficrespij," %1d-%1d",i,j);
11098: fprintf(ficrespij,"\n");
11099: for (h=0; h<=nhstepm; h++){
11100: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11101: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 11102: for(i=1; i<=nlstate;i++)
11103: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 11104: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 11105: fprintf(ficrespij,"\n");
11106: }
1.183 brouard 11107: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11108: fprintf(ficrespij,"\n");
11109: }
1.180 brouard 11110: /*}*/
11111: }
1.218 brouard 11112: return 0;
1.180 brouard 11113: }
1.218 brouard 11114:
11115: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 11116: /*------------- h Bij x at various ages ------------*/
11117:
11118: int stepsize;
1.218 brouard 11119: /* int agelim; */
11120: int ageminl;
1.217 brouard 11121: int hstepm;
11122: int nhstepm;
1.238 brouard 11123: int h, i, i1, j, k, nres;
1.218 brouard 11124:
1.217 brouard 11125: double agedeb;
11126: double ***p3mat;
1.218 brouard 11127:
11128: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
11129: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
11130: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11131: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11132: }
11133: printf("Computing pij back: result on file '%s' \n", filerespijb);
11134: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
11135:
11136: stepsize=(int) (stepm+YEARM-1)/YEARM;
11137: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 11138:
1.218 brouard 11139: /* agelim=AGESUP; */
1.289 brouard 11140: ageminl=AGEINF; /* was 30 */
1.218 brouard 11141: hstepm=stepsize*YEARM; /* Every year of age */
11142: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11143:
11144: /* hstepm=1; aff par mois*/
11145: pstamp(ficrespijb);
1.255 brouard 11146: 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 11147: i1= pow(2,cptcoveff);
1.218 brouard 11148: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11149: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11150: /* k=k+1; */
1.238 brouard 11151: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11152: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11153: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11154: continue;
11155: fprintf(ficrespijb,"\n#****** ");
11156: for(j=1;j<=cptcoveff;j++)
11157: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11158: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11159: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11160: }
11161: fprintf(ficrespijb,"******\n");
1.264 brouard 11162: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 11163: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
11164: continue;
11165: }
11166:
11167: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
11168: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
11169: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 11170: 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 */
11171: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 11172:
11173: /* nhstepm=nhstepm*YEARM; aff par mois*/
11174:
1.266 brouard 11175: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
11176: /* and memory limitations if stepm is small */
11177:
1.238 brouard 11178: /* oldm=oldms;savm=savms; */
11179: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 11180: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 11181: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 11182: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 11183: for(i=1; i<=nlstate;i++)
11184: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 11185: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 11186: fprintf(ficrespijb,"\n");
1.238 brouard 11187: for (h=0; h<=nhstepm; h++){
11188: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11189: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
11190: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
11191: for(i=1; i<=nlstate;i++)
11192: for(j=1; j<=nlstate+ndeath;j++)
11193: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
11194: fprintf(ficrespijb,"\n");
11195: }
11196: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11197: fprintf(ficrespijb,"\n");
11198: } /* end age deb */
11199: } /* end combination */
11200: } /* end nres */
1.218 brouard 11201: return 0;
11202: } /* hBijx */
1.217 brouard 11203:
1.180 brouard 11204:
1.136 brouard 11205: /***********************************************/
11206: /**************** Main Program *****************/
11207: /***********************************************/
11208:
11209: int main(int argc, char *argv[])
11210: {
11211: #ifdef GSL
11212: const gsl_multimin_fminimizer_type *T;
11213: size_t iteri = 0, it;
11214: int rval = GSL_CONTINUE;
11215: int status = GSL_SUCCESS;
11216: double ssval;
11217: #endif
11218: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11219: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11220: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11221: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11222: int jj, ll, li, lj, lk;
1.136 brouard 11223: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11224: int num_filled;
1.136 brouard 11225: int itimes;
11226: int NDIM=2;
11227: int vpopbased=0;
1.235 brouard 11228: int nres=0;
1.258 brouard 11229: int endishere=0;
1.277 brouard 11230: int noffset=0;
1.274 brouard 11231: int ncurrv=0; /* Temporary variable */
11232:
1.164 brouard 11233: char ca[32], cb[32];
1.136 brouard 11234: /* FILE *fichtm; *//* Html File */
11235: /* FILE *ficgp;*/ /*Gnuplot File */
11236: struct stat info;
1.191 brouard 11237: double agedeb=0.;
1.194 brouard 11238:
11239: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11240: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11241:
1.165 brouard 11242: double fret;
1.191 brouard 11243: double dum=0.; /* Dummy variable */
1.136 brouard 11244: double ***p3mat;
1.218 brouard 11245: /* double ***mobaverage; */
1.319 brouard 11246: double wald;
1.164 brouard 11247:
11248: char line[MAXLINE];
1.197 brouard 11249: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11250:
1.234 brouard 11251: char modeltemp[MAXLINE];
1.230 brouard 11252: char resultline[MAXLINE];
11253:
1.136 brouard 11254: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11255: char *tok, *val; /* pathtot */
1.290 brouard 11256: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 11257: int c, h , cpt, c2;
1.191 brouard 11258: int jl=0;
11259: int i1, j1, jk, stepsize=0;
1.194 brouard 11260: int count=0;
11261:
1.164 brouard 11262: int *tab;
1.136 brouard 11263: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11264: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11265: /* double anprojf, mprojf, jprojf; */
11266: /* double jintmean,mintmean,aintmean; */
11267: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11268: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11269: double yrfproj= 10.0; /* Number of years of forward projections */
11270: double yrbproj= 10.0; /* Number of years of backward projections */
11271: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11272: int mobilav=0,popforecast=0;
1.191 brouard 11273: int hstepm=0, nhstepm=0;
1.136 brouard 11274: int agemortsup;
11275: float sumlpop=0.;
11276: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11277: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11278:
1.191 brouard 11279: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11280: double ftolpl=FTOL;
11281: double **prlim;
1.217 brouard 11282: double **bprlim;
1.317 brouard 11283: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
11284: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11285: double ***paramstart; /* Matrix of starting parameter values */
11286: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11287: double **matcov; /* Matrix of covariance */
1.203 brouard 11288: double **hess; /* Hessian matrix */
1.136 brouard 11289: double ***delti3; /* Scale */
11290: double *delti; /* Scale */
11291: double ***eij, ***vareij;
11292: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11293:
1.136 brouard 11294: double *epj, vepp;
1.164 brouard 11295:
1.273 brouard 11296: double dateprev1, dateprev2;
1.296 brouard 11297: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11298: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11299:
1.217 brouard 11300:
1.136 brouard 11301: double **ximort;
1.145 brouard 11302: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11303: int *dcwave;
11304:
1.164 brouard 11305: char z[1]="c";
1.136 brouard 11306:
11307: /*char *strt;*/
11308: char strtend[80];
1.126 brouard 11309:
1.164 brouard 11310:
1.126 brouard 11311: /* setlocale (LC_ALL, ""); */
11312: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11313: /* textdomain (PACKAGE); */
11314: /* setlocale (LC_CTYPE, ""); */
11315: /* setlocale (LC_MESSAGES, ""); */
11316:
11317: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11318: rstart_time = time(NULL);
11319: /* (void) gettimeofday(&start_time,&tzp);*/
11320: start_time = *localtime(&rstart_time);
1.126 brouard 11321: curr_time=start_time;
1.157 brouard 11322: /*tml = *localtime(&start_time.tm_sec);*/
11323: /* strcpy(strstart,asctime(&tml)); */
11324: strcpy(strstart,asctime(&start_time));
1.126 brouard 11325:
11326: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11327: /* tp.tm_sec = tp.tm_sec +86400; */
11328: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11329: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11330: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11331: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11332: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11333: /* strt=asctime(&tmg); */
11334: /* printf("Time(after) =%s",strstart); */
11335: /* (void) time (&time_value);
11336: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11337: * tm = *localtime(&time_value);
11338: * strstart=asctime(&tm);
11339: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11340: */
11341:
11342: nberr=0; /* Number of errors and warnings */
11343: nbwarn=0;
1.184 brouard 11344: #ifdef WIN32
11345: _getcwd(pathcd, size);
11346: #else
1.126 brouard 11347: getcwd(pathcd, size);
1.184 brouard 11348: #endif
1.191 brouard 11349: syscompilerinfo(0);
1.196 brouard 11350: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11351: if(argc <=1){
11352: printf("\nEnter the parameter file name: ");
1.205 brouard 11353: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11354: printf("ERROR Empty parameter file name\n");
11355: goto end;
11356: }
1.126 brouard 11357: i=strlen(pathr);
11358: if(pathr[i-1]=='\n')
11359: pathr[i-1]='\0';
1.156 brouard 11360: i=strlen(pathr);
1.205 brouard 11361: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11362: pathr[i-1]='\0';
1.205 brouard 11363: }
11364: i=strlen(pathr);
11365: if( i==0 ){
11366: printf("ERROR Empty parameter file name\n");
11367: goto end;
11368: }
11369: for (tok = pathr; tok != NULL; ){
1.126 brouard 11370: printf("Pathr |%s|\n",pathr);
11371: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11372: printf("val= |%s| pathr=%s\n",val,pathr);
11373: strcpy (pathtot, val);
11374: if(pathr[0] == '\0') break; /* Dirty */
11375: }
11376: }
1.281 brouard 11377: else if (argc<=2){
11378: strcpy(pathtot,argv[1]);
11379: }
1.126 brouard 11380: else{
11381: strcpy(pathtot,argv[1]);
1.281 brouard 11382: strcpy(z,argv[2]);
11383: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11384: }
11385: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11386: /*cygwin_split_path(pathtot,path,optionfile);
11387: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11388: /* cutv(path,optionfile,pathtot,'\\');*/
11389:
11390: /* Split argv[0], imach program to get pathimach */
11391: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11392: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11393: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11394: /* strcpy(pathimach,argv[0]); */
11395: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11396: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11397: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11398: #ifdef WIN32
11399: _chdir(path); /* Can be a relative path */
11400: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11401: #else
1.126 brouard 11402: chdir(path); /* Can be a relative path */
1.184 brouard 11403: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11404: #endif
11405: printf("Current directory %s!\n",pathcd);
1.126 brouard 11406: strcpy(command,"mkdir ");
11407: strcat(command,optionfilefiname);
11408: if((outcmd=system(command)) != 0){
1.169 brouard 11409: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11410: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11411: /* fclose(ficlog); */
11412: /* exit(1); */
11413: }
11414: /* if((imk=mkdir(optionfilefiname))<0){ */
11415: /* perror("mkdir"); */
11416: /* } */
11417:
11418: /*-------- arguments in the command line --------*/
11419:
1.186 brouard 11420: /* Main Log file */
1.126 brouard 11421: strcat(filelog, optionfilefiname);
11422: strcat(filelog,".log"); /* */
11423: if((ficlog=fopen(filelog,"w"))==NULL) {
11424: printf("Problem with logfile %s\n",filelog);
11425: goto end;
11426: }
11427: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11428: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11429: fprintf(ficlog,"\nEnter the parameter file name: \n");
11430: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11431: path=%s \n\
11432: optionfile=%s\n\
11433: optionfilext=%s\n\
1.156 brouard 11434: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11435:
1.197 brouard 11436: syscompilerinfo(1);
1.167 brouard 11437:
1.126 brouard 11438: printf("Local time (at start):%s",strstart);
11439: fprintf(ficlog,"Local time (at start): %s",strstart);
11440: fflush(ficlog);
11441: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11442: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11443:
11444: /* */
11445: strcpy(fileres,"r");
11446: strcat(fileres, optionfilefiname);
1.201 brouard 11447: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11448: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11449: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11450:
1.186 brouard 11451: /* Main ---------arguments file --------*/
1.126 brouard 11452:
11453: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11454: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11455: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11456: fflush(ficlog);
1.149 brouard 11457: /* goto end; */
11458: exit(70);
1.126 brouard 11459: }
11460:
11461: strcpy(filereso,"o");
1.201 brouard 11462: strcat(filereso,fileresu);
1.126 brouard 11463: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11464: printf("Problem with Output resultfile: %s\n", filereso);
11465: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11466: fflush(ficlog);
11467: goto end;
11468: }
1.278 brouard 11469: /*-------- Rewriting parameter file ----------*/
11470: strcpy(rfileres,"r"); /* "Rparameterfile */
11471: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11472: strcat(rfileres,"."); /* */
11473: strcat(rfileres,optionfilext); /* Other files have txt extension */
11474: if((ficres =fopen(rfileres,"w"))==NULL) {
11475: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11476: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11477: fflush(ficlog);
11478: goto end;
11479: }
11480: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11481:
1.278 brouard 11482:
1.126 brouard 11483: /* Reads comments: lines beginning with '#' */
11484: numlinepar=0;
1.277 brouard 11485: /* Is it a BOM UTF-8 Windows file? */
11486: /* First parameter line */
1.197 brouard 11487: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11488: noffset=0;
11489: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11490: {
11491: noffset=noffset+3;
11492: printf("# File is an UTF8 Bom.\n"); // 0xBF
11493: }
1.302 brouard 11494: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11495: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11496: {
11497: noffset=noffset+2;
11498: printf("# File is an UTF16BE BOM file\n");
11499: }
11500: else if( line[0] == 0 && line[1] == 0)
11501: {
11502: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11503: noffset=noffset+4;
11504: printf("# File is an UTF16BE BOM file\n");
11505: }
11506: } else{
11507: ;/*printf(" Not a BOM file\n");*/
11508: }
11509:
1.197 brouard 11510: /* If line starts with a # it is a comment */
1.277 brouard 11511: if (line[noffset] == '#') {
1.197 brouard 11512: numlinepar++;
11513: fputs(line,stdout);
11514: fputs(line,ficparo);
1.278 brouard 11515: fputs(line,ficres);
1.197 brouard 11516: fputs(line,ficlog);
11517: continue;
11518: }else
11519: break;
11520: }
11521: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11522: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11523: if (num_filled != 5) {
11524: printf("Should be 5 parameters\n");
1.283 brouard 11525: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11526: }
1.126 brouard 11527: numlinepar++;
1.197 brouard 11528: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11529: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11530: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11531: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11532: }
11533: /* Second parameter line */
11534: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11535: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11536: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11537: if (line[0] == '#') {
11538: numlinepar++;
1.283 brouard 11539: printf("%s",line);
11540: fprintf(ficres,"%s",line);
11541: fprintf(ficparo,"%s",line);
11542: fprintf(ficlog,"%s",line);
1.197 brouard 11543: continue;
11544: }else
11545: break;
11546: }
1.223 brouard 11547: 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", \
11548: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11549: if (num_filled != 11) {
11550: 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 11551: printf("but line=%s\n",line);
1.283 brouard 11552: 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");
11553: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11554: }
1.286 brouard 11555: if( lastpass > maxwav){
11556: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11557: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11558: fflush(ficlog);
11559: goto end;
11560: }
11561: 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 11562: 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 11563: 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 11564: 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 11565: }
1.203 brouard 11566: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11567: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11568: /* Third parameter line */
11569: while(fgets(line, MAXLINE, ficpar)) {
11570: /* If line starts with a # it is a comment */
11571: if (line[0] == '#') {
11572: numlinepar++;
1.283 brouard 11573: printf("%s",line);
11574: fprintf(ficres,"%s",line);
11575: fprintf(ficparo,"%s",line);
11576: fprintf(ficlog,"%s",line);
1.197 brouard 11577: continue;
11578: }else
11579: break;
11580: }
1.201 brouard 11581: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11582: if (num_filled != 1){
1.302 brouard 11583: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
11584: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11585: model[0]='\0';
11586: goto end;
11587: }
11588: else{
11589: if (model[0]=='+'){
11590: for(i=1; i<=strlen(model);i++)
11591: modeltemp[i-1]=model[i];
1.201 brouard 11592: strcpy(model,modeltemp);
1.197 brouard 11593: }
11594: }
1.199 brouard 11595: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11596: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11597: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11598: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11599: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11600: }
11601: /* 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); */
11602: /* numlinepar=numlinepar+3; /\* In general *\/ */
11603: /* 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 11604: /* 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); */
11605: /* 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 11606: fflush(ficlog);
1.190 brouard 11607: /* if(model[0]=='#'|| model[0]== '\0'){ */
11608: if(model[0]=='#'){
1.279 brouard 11609: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11610: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11611: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11612: if(mle != -1){
1.279 brouard 11613: 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 11614: exit(1);
11615: }
11616: }
1.126 brouard 11617: while((c=getc(ficpar))=='#' && c!= EOF){
11618: ungetc(c,ficpar);
11619: fgets(line, MAXLINE, ficpar);
11620: numlinepar++;
1.195 brouard 11621: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11622: z[0]=line[1];
11623: }
11624: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11625: fputs(line, stdout);
11626: //puts(line);
1.126 brouard 11627: fputs(line,ficparo);
11628: fputs(line,ficlog);
11629: }
11630: ungetc(c,ficpar);
11631:
11632:
1.290 brouard 11633: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11634: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11635: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11636: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11637: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11638: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11639: v1+v2*age+v2*v3 makes cptcovn = 3
11640: */
11641: if (strlen(model)>1)
1.187 brouard 11642: 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 11643: else
1.187 brouard 11644: ncovmodel=2; /* Constant and age */
1.133 brouard 11645: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11646: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11647: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11648: 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);
11649: 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);
11650: fflush(stdout);
11651: fclose (ficlog);
11652: goto end;
11653: }
1.126 brouard 11654: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11655: delti=delti3[1][1];
11656: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11657: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11658: /* We could also provide initial parameters values giving by simple logistic regression
11659: * only one way, that is without matrix product. We will have nlstate maximizations */
11660: /* for(i=1;i<nlstate;i++){ */
11661: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11662: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11663: /* } */
1.126 brouard 11664: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11665: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11666: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11667: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11668: fclose (ficparo);
11669: fclose (ficlog);
11670: goto end;
11671: exit(0);
1.220 brouard 11672: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11673: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11674: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11675: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11676: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11677: matcov=matrix(1,npar,1,npar);
1.203 brouard 11678: hess=matrix(1,npar,1,npar);
1.220 brouard 11679: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11680: /* Read guessed parameters */
1.126 brouard 11681: /* Reads comments: lines beginning with '#' */
11682: while((c=getc(ficpar))=='#' && c!= EOF){
11683: ungetc(c,ficpar);
11684: fgets(line, MAXLINE, ficpar);
11685: numlinepar++;
1.141 brouard 11686: fputs(line,stdout);
1.126 brouard 11687: fputs(line,ficparo);
11688: fputs(line,ficlog);
11689: }
11690: ungetc(c,ficpar);
11691:
11692: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11693: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11694: for(i=1; i <=nlstate; i++){
1.234 brouard 11695: j=0;
1.126 brouard 11696: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11697: if(jj==i) continue;
11698: j++;
1.292 brouard 11699: while((c=getc(ficpar))=='#' && c!= EOF){
11700: ungetc(c,ficpar);
11701: fgets(line, MAXLINE, ficpar);
11702: numlinepar++;
11703: fputs(line,stdout);
11704: fputs(line,ficparo);
11705: fputs(line,ficlog);
11706: }
11707: ungetc(c,ficpar);
1.234 brouard 11708: fscanf(ficpar,"%1d%1d",&i1,&j1);
11709: if ((i1 != i) || (j1 != jj)){
11710: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11711: It might be a problem of design; if ncovcol and the model are correct\n \
11712: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11713: exit(1);
11714: }
11715: fprintf(ficparo,"%1d%1d",i1,j1);
11716: if(mle==1)
11717: printf("%1d%1d",i,jj);
11718: fprintf(ficlog,"%1d%1d",i,jj);
11719: for(k=1; k<=ncovmodel;k++){
11720: fscanf(ficpar," %lf",¶m[i][j][k]);
11721: if(mle==1){
11722: printf(" %lf",param[i][j][k]);
11723: fprintf(ficlog," %lf",param[i][j][k]);
11724: }
11725: else
11726: fprintf(ficlog," %lf",param[i][j][k]);
11727: fprintf(ficparo," %lf",param[i][j][k]);
11728: }
11729: fscanf(ficpar,"\n");
11730: numlinepar++;
11731: if(mle==1)
11732: printf("\n");
11733: fprintf(ficlog,"\n");
11734: fprintf(ficparo,"\n");
1.126 brouard 11735: }
11736: }
11737: fflush(ficlog);
1.234 brouard 11738:
1.251 brouard 11739: /* Reads parameters values */
1.126 brouard 11740: p=param[1][1];
1.251 brouard 11741: pstart=paramstart[1][1];
1.126 brouard 11742:
11743: /* Reads comments: lines beginning with '#' */
11744: while((c=getc(ficpar))=='#' && c!= EOF){
11745: ungetc(c,ficpar);
11746: fgets(line, MAXLINE, ficpar);
11747: numlinepar++;
1.141 brouard 11748: fputs(line,stdout);
1.126 brouard 11749: fputs(line,ficparo);
11750: fputs(line,ficlog);
11751: }
11752: ungetc(c,ficpar);
11753:
11754: for(i=1; i <=nlstate; i++){
11755: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11756: fscanf(ficpar,"%1d%1d",&i1,&j1);
11757: if ( (i1-i) * (j1-j) != 0){
11758: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11759: exit(1);
11760: }
11761: printf("%1d%1d",i,j);
11762: fprintf(ficparo,"%1d%1d",i1,j1);
11763: fprintf(ficlog,"%1d%1d",i1,j1);
11764: for(k=1; k<=ncovmodel;k++){
11765: fscanf(ficpar,"%le",&delti3[i][j][k]);
11766: printf(" %le",delti3[i][j][k]);
11767: fprintf(ficparo," %le",delti3[i][j][k]);
11768: fprintf(ficlog," %le",delti3[i][j][k]);
11769: }
11770: fscanf(ficpar,"\n");
11771: numlinepar++;
11772: printf("\n");
11773: fprintf(ficparo,"\n");
11774: fprintf(ficlog,"\n");
1.126 brouard 11775: }
11776: }
11777: fflush(ficlog);
1.234 brouard 11778:
1.145 brouard 11779: /* Reads covariance matrix */
1.126 brouard 11780: delti=delti3[1][1];
1.220 brouard 11781:
11782:
1.126 brouard 11783: /* 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 11784:
1.126 brouard 11785: /* Reads comments: lines beginning with '#' */
11786: while((c=getc(ficpar))=='#' && c!= EOF){
11787: ungetc(c,ficpar);
11788: fgets(line, MAXLINE, ficpar);
11789: numlinepar++;
1.141 brouard 11790: fputs(line,stdout);
1.126 brouard 11791: fputs(line,ficparo);
11792: fputs(line,ficlog);
11793: }
11794: ungetc(c,ficpar);
1.220 brouard 11795:
1.126 brouard 11796: matcov=matrix(1,npar,1,npar);
1.203 brouard 11797: hess=matrix(1,npar,1,npar);
1.131 brouard 11798: for(i=1; i <=npar; i++)
11799: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11800:
1.194 brouard 11801: /* Scans npar lines */
1.126 brouard 11802: for(i=1; i <=npar; i++){
1.226 brouard 11803: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11804: if(count != 3){
1.226 brouard 11805: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11806: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11807: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11808: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11809: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11810: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11811: exit(1);
1.220 brouard 11812: }else{
1.226 brouard 11813: if(mle==1)
11814: printf("%1d%1d%d",i1,j1,jk);
11815: }
11816: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11817: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11818: for(j=1; j <=i; j++){
1.226 brouard 11819: fscanf(ficpar," %le",&matcov[i][j]);
11820: if(mle==1){
11821: printf(" %.5le",matcov[i][j]);
11822: }
11823: fprintf(ficlog," %.5le",matcov[i][j]);
11824: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11825: }
11826: fscanf(ficpar,"\n");
11827: numlinepar++;
11828: if(mle==1)
1.220 brouard 11829: printf("\n");
1.126 brouard 11830: fprintf(ficlog,"\n");
11831: fprintf(ficparo,"\n");
11832: }
1.194 brouard 11833: /* End of read covariance matrix npar lines */
1.126 brouard 11834: for(i=1; i <=npar; i++)
11835: for(j=i+1;j<=npar;j++)
1.226 brouard 11836: matcov[i][j]=matcov[j][i];
1.126 brouard 11837:
11838: if(mle==1)
11839: printf("\n");
11840: fprintf(ficlog,"\n");
11841:
11842: fflush(ficlog);
11843:
11844: } /* End of mle != -3 */
1.218 brouard 11845:
1.186 brouard 11846: /* Main data
11847: */
1.290 brouard 11848: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11849: /* num=lvector(1,n); */
11850: /* moisnais=vector(1,n); */
11851: /* annais=vector(1,n); */
11852: /* moisdc=vector(1,n); */
11853: /* andc=vector(1,n); */
11854: /* weight=vector(1,n); */
11855: /* agedc=vector(1,n); */
11856: /* cod=ivector(1,n); */
11857: /* for(i=1;i<=n;i++){ */
11858: num=lvector(firstobs,lastobs);
11859: moisnais=vector(firstobs,lastobs);
11860: annais=vector(firstobs,lastobs);
11861: moisdc=vector(firstobs,lastobs);
11862: andc=vector(firstobs,lastobs);
11863: weight=vector(firstobs,lastobs);
11864: agedc=vector(firstobs,lastobs);
11865: cod=ivector(firstobs,lastobs);
11866: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11867: num[i]=0;
11868: moisnais[i]=0;
11869: annais[i]=0;
11870: moisdc[i]=0;
11871: andc[i]=0;
11872: agedc[i]=0;
11873: cod[i]=0;
11874: weight[i]=1.0; /* Equal weights, 1 by default */
11875: }
1.290 brouard 11876: mint=matrix(1,maxwav,firstobs,lastobs);
11877: anint=matrix(1,maxwav,firstobs,lastobs);
11878: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11879: tab=ivector(1,NCOVMAX);
1.144 brouard 11880: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11881: 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 11882:
1.136 brouard 11883: /* Reads data from file datafile */
11884: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11885: goto end;
11886:
11887: /* Calculation of the number of parameters from char model */
1.234 brouard 11888: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11889: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11890: k=3 V4 Tvar[k=3]= 4 (from V4)
11891: k=2 V1 Tvar[k=2]= 1 (from V1)
11892: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11893: */
11894:
11895: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11896: TvarsDind=ivector(1,NCOVMAX); /* */
11897: TvarsD=ivector(1,NCOVMAX); /* */
11898: TvarsQind=ivector(1,NCOVMAX); /* */
11899: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11900: TvarF=ivector(1,NCOVMAX); /* */
11901: TvarFind=ivector(1,NCOVMAX); /* */
11902: TvarV=ivector(1,NCOVMAX); /* */
11903: TvarVind=ivector(1,NCOVMAX); /* */
11904: TvarA=ivector(1,NCOVMAX); /* */
11905: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11906: TvarFD=ivector(1,NCOVMAX); /* */
11907: TvarFDind=ivector(1,NCOVMAX); /* */
11908: TvarFQ=ivector(1,NCOVMAX); /* */
11909: TvarFQind=ivector(1,NCOVMAX); /* */
11910: TvarVD=ivector(1,NCOVMAX); /* */
11911: TvarVDind=ivector(1,NCOVMAX); /* */
11912: TvarVQ=ivector(1,NCOVMAX); /* */
11913: TvarVQind=ivector(1,NCOVMAX); /* */
11914:
1.230 brouard 11915: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11916: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11917: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11918: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11919: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11920: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11921: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11922: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11923: */
11924: /* For model-covariate k tells which data-covariate to use but
11925: because this model-covariate is a construction we invent a new column
11926: ncovcol + k1
11927: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11928: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11929: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11930: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11931: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11932: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11933: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11934: */
1.145 brouard 11935: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11936: 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 11937: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11938: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11939: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11940: 4 covariates (3 plus signs)
11941: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11942: */
1.230 brouard 11943: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11944: * individual dummy, fixed or varying:
11945: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11946: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11947: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11948: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11949: * Tmodelind[1]@9={9,0,3,2,}*/
11950: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11951: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11952: * individual quantitative, fixed or varying:
11953: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11954: * 3, 1, 0, 0, 0, 0, 0, 0},
11955: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11956: /* Main decodemodel */
11957:
1.187 brouard 11958:
1.223 brouard 11959: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11960: goto end;
11961:
1.137 brouard 11962: if((double)(lastobs-imx)/(double)imx > 1.10){
11963: nbwarn++;
11964: 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);
11965: 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);
11966: }
1.136 brouard 11967: /* if(mle==1){*/
1.137 brouard 11968: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11969: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11970: }
11971:
11972: /*-calculation of age at interview from date of interview and age at death -*/
11973: agev=matrix(1,maxwav,1,imx);
11974:
11975: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11976: goto end;
11977:
1.126 brouard 11978:
1.136 brouard 11979: agegomp=(int)agemin;
1.290 brouard 11980: free_vector(moisnais,firstobs,lastobs);
11981: free_vector(annais,firstobs,lastobs);
1.126 brouard 11982: /* free_matrix(mint,1,maxwav,1,n);
11983: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11984: /* free_vector(moisdc,1,n); */
11985: /* free_vector(andc,1,n); */
1.145 brouard 11986: /* */
11987:
1.126 brouard 11988: wav=ivector(1,imx);
1.214 brouard 11989: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11990: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11991: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11992: 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.*/
11993: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11994: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11995:
11996: /* Concatenates waves */
1.214 brouard 11997: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11998: Death is a valid wave (if date is known).
11999: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
12000: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
12001: and mw[mi+1][i]. dh depends on stepm.
12002: */
12003:
1.126 brouard 12004: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 12005: /* Concatenates waves */
1.145 brouard 12006:
1.290 brouard 12007: free_vector(moisdc,firstobs,lastobs);
12008: free_vector(andc,firstobs,lastobs);
1.215 brouard 12009:
1.126 brouard 12010: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12011: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12012: ncodemax[1]=1;
1.145 brouard 12013: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12014: cptcoveff=0;
1.220 brouard 12015: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
12016: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 12017: }
12018:
12019: ncovcombmax=pow(2,cptcoveff);
12020: invalidvarcomb=ivector(1, ncovcombmax);
12021: for(i=1;i<ncovcombmax;i++)
12022: invalidvarcomb[i]=0;
12023:
1.211 brouard 12024: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12025: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12026: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12027:
1.200 brouard 12028: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12029: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12030: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12031: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12032: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12033: * (currently 0 or 1) in the data.
12034: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12035: * corresponding modality (h,j).
12036: */
12037:
1.145 brouard 12038: h=0;
12039: /*if (cptcovn > 0) */
1.126 brouard 12040: m=pow(2,cptcoveff);
12041:
1.144 brouard 12042: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12043: * For k=4 covariates, h goes from 1 to m=2**k
12044: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12045: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 12046: * h\k 1 2 3 4
1.143 brouard 12047: *______________________________
12048: * 1 i=1 1 i=1 1 i=1 1 i=1 1
12049: * 2 2 1 1 1
12050: * 3 i=2 1 2 1 1
12051: * 4 2 2 1 1
12052: * 5 i=3 1 i=2 1 2 1
12053: * 6 2 1 2 1
12054: * 7 i=4 1 2 2 1
12055: * 8 2 2 2 1
1.197 brouard 12056: * 9 i=5 1 i=3 1 i=2 1 2
12057: * 10 2 1 1 2
12058: * 11 i=6 1 2 1 2
12059: * 12 2 2 1 2
12060: * 13 i=7 1 i=4 1 2 2
12061: * 14 2 1 2 2
12062: * 15 i=8 1 2 2 2
12063: * 16 2 2 2 2
1.143 brouard 12064: */
1.212 brouard 12065: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 12066: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
12067: * and the value of each covariate?
12068: * V1=1, V2=1, V3=2, V4=1 ?
12069: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
12070: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
12071: * In order to get the real value in the data, we use nbcode
12072: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
12073: * We are keeping this crazy system in order to be able (in the future?)
12074: * to have more than 2 values (0 or 1) for a covariate.
12075: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
12076: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
12077: * bbbbbbbb
12078: * 76543210
12079: * h-1 00000101 (6-1=5)
1.219 brouard 12080: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 12081: * &
12082: * 1 00000001 (1)
1.219 brouard 12083: * 00000000 = 1 & ((h-1) >> (k-1))
12084: * +1= 00000001 =1
1.211 brouard 12085: *
12086: * h=14, k=3 => h'=h-1=13, k'=k-1=2
12087: * h' 1101 =2^3+2^2+0x2^1+2^0
12088: * >>k' 11
12089: * & 00000001
12090: * = 00000001
12091: * +1 = 00000010=2 = codtabm(14,3)
12092: * Reverse h=6 and m=16?
12093: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
12094: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
12095: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
12096: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
12097: * V3=decodtabm(14,3,2**4)=2
12098: * h'=13 1101 =2^3+2^2+0x2^1+2^0
12099: *(h-1) >> (j-1) 0011 =13 >> 2
12100: * &1 000000001
12101: * = 000000001
12102: * +1= 000000010 =2
12103: * 2211
12104: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
12105: * V3=2
1.220 brouard 12106: * codtabm and decodtabm are identical
1.211 brouard 12107: */
12108:
1.145 brouard 12109:
12110: free_ivector(Ndum,-1,NCOVMAX);
12111:
12112:
1.126 brouard 12113:
1.186 brouard 12114: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 12115: strcpy(optionfilegnuplot,optionfilefiname);
12116: if(mle==-3)
1.201 brouard 12117: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 12118: strcat(optionfilegnuplot,".gp");
12119:
12120: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
12121: printf("Problem with file %s",optionfilegnuplot);
12122: }
12123: else{
1.204 brouard 12124: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 12125: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 12126: //fprintf(ficgp,"set missing 'NaNq'\n");
12127: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 12128: }
12129: /* fclose(ficgp);*/
1.186 brouard 12130:
12131:
12132: /* Initialisation of --------- index.htm --------*/
1.126 brouard 12133:
12134: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
12135: if(mle==-3)
1.201 brouard 12136: strcat(optionfilehtm,"-MORT_");
1.126 brouard 12137: strcat(optionfilehtm,".htm");
12138: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 12139: printf("Problem with %s \n",optionfilehtm);
12140: exit(0);
1.126 brouard 12141: }
12142:
12143: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
12144: strcat(optionfilehtmcov,"-cov.htm");
12145: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
12146: printf("Problem with %s \n",optionfilehtmcov), exit(0);
12147: }
12148: else{
12149: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
12150: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12151: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 12152: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
12153: }
12154:
1.324 ! brouard 12155: fprintf(fichtm,"<html><head>\n<head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<title>IMaCh %s</title></head>\n <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n<font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>-EUROREVES-Institut de longévité-2013-2016-Japan Society for the Promotion of Sciences 日本学術振興会 (<a href=https://www.jsps.go.jp/english/e-grants/>Grant-in-Aid for Scientific Research 25293121</a>) - <a href=https://software.intel.com/en-us>Intel Software 2015-2018</a></font><br> \
1.204 brouard 12156: <hr size=\"2\" color=\"#EC5E5E\"> \n\
12157: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 12158: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12159: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 12160: \n\
12161: <hr size=\"2\" color=\"#EC5E5E\">\
12162: <ul><li><h4>Parameter files</h4>\n\
12163: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
12164: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
12165: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
12166: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
12167: - Date and time at start: %s</ul>\n",\
12168: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
12169: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
12170: fileres,fileres,\
12171: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
12172: fflush(fichtm);
12173:
12174: strcpy(pathr,path);
12175: strcat(pathr,optionfilefiname);
1.184 brouard 12176: #ifdef WIN32
12177: _chdir(optionfilefiname); /* Move to directory named optionfile */
12178: #else
1.126 brouard 12179: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12180: #endif
12181:
1.126 brouard 12182:
1.220 brouard 12183: /* Calculates basic frequencies. Computes observed prevalence at single age
12184: and for any valid combination of covariates
1.126 brouard 12185: and prints on file fileres'p'. */
1.251 brouard 12186: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12187: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12188:
12189: fprintf(fichtm,"\n");
1.286 brouard 12190: 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 12191: ftol, stepm);
12192: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12193: ncurrv=1;
12194: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12195: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12196: ncurrv=i;
12197: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12198: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12199: ncurrv=i;
12200: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12201: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12202: ncurrv=i;
12203: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12204: 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", \
12205: nlstate, ndeath, maxwav, mle, weightopt);
12206:
12207: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12208: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12209:
12210:
1.317 brouard 12211: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12212: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12213: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12214: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12215: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12216: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12217: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12218: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12219: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12220:
1.126 brouard 12221: /* For Powell, parameters are in a vector p[] starting at p[1]
12222: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12223: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12224:
12225: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12226: /* For mortality only */
1.126 brouard 12227: if (mle==-3){
1.136 brouard 12228: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12229: for(i=1;i<=NDIM;i++)
12230: for(j=1;j<=NDIM;j++)
12231: ximort[i][j]=0.;
1.186 brouard 12232: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12233: cens=ivector(firstobs,lastobs);
12234: ageexmed=vector(firstobs,lastobs);
12235: agecens=vector(firstobs,lastobs);
12236: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12237:
1.126 brouard 12238: for (i=1; i<=imx; i++){
12239: dcwave[i]=-1;
12240: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12241: if (s[m][i]>nlstate) {
12242: dcwave[i]=m;
12243: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12244: break;
12245: }
1.126 brouard 12246: }
1.226 brouard 12247:
1.126 brouard 12248: for (i=1; i<=imx; i++) {
12249: if (wav[i]>0){
1.226 brouard 12250: ageexmed[i]=agev[mw[1][i]][i];
12251: j=wav[i];
12252: agecens[i]=1.;
12253:
12254: if (ageexmed[i]> 1 && wav[i] > 0){
12255: agecens[i]=agev[mw[j][i]][i];
12256: cens[i]= 1;
12257: }else if (ageexmed[i]< 1)
12258: cens[i]= -1;
12259: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12260: cens[i]=0 ;
1.126 brouard 12261: }
12262: else cens[i]=-1;
12263: }
12264:
12265: for (i=1;i<=NDIM;i++) {
12266: for (j=1;j<=NDIM;j++)
1.226 brouard 12267: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12268: }
12269:
1.302 brouard 12270: p[1]=0.0268; p[NDIM]=0.083;
12271: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12272:
12273:
1.136 brouard 12274: #ifdef GSL
12275: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12276: #else
1.126 brouard 12277: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12278: #endif
1.201 brouard 12279: strcpy(filerespow,"POW-MORT_");
12280: strcat(filerespow,fileresu);
1.126 brouard 12281: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12282: printf("Problem with resultfile: %s\n", filerespow);
12283: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12284: }
1.136 brouard 12285: #ifdef GSL
12286: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12287: #else
1.126 brouard 12288: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12289: #endif
1.126 brouard 12290: /* for (i=1;i<=nlstate;i++)
12291: for(j=1;j<=nlstate+ndeath;j++)
12292: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12293: */
12294: fprintf(ficrespow,"\n");
1.136 brouard 12295: #ifdef GSL
12296: /* gsl starts here */
12297: T = gsl_multimin_fminimizer_nmsimplex;
12298: gsl_multimin_fminimizer *sfm = NULL;
12299: gsl_vector *ss, *x;
12300: gsl_multimin_function minex_func;
12301:
12302: /* Initial vertex size vector */
12303: ss = gsl_vector_alloc (NDIM);
12304:
12305: if (ss == NULL){
12306: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12307: }
12308: /* Set all step sizes to 1 */
12309: gsl_vector_set_all (ss, 0.001);
12310:
12311: /* Starting point */
1.126 brouard 12312:
1.136 brouard 12313: x = gsl_vector_alloc (NDIM);
12314:
12315: if (x == NULL){
12316: gsl_vector_free(ss);
12317: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12318: }
12319:
12320: /* Initialize method and iterate */
12321: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12322: /* gsl_vector_set(x, 0, 0.0268); */
12323: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12324: gsl_vector_set(x, 0, p[1]);
12325: gsl_vector_set(x, 1, p[2]);
12326:
12327: minex_func.f = &gompertz_f;
12328: minex_func.n = NDIM;
12329: minex_func.params = (void *)&p; /* ??? */
12330:
12331: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12332: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12333:
12334: printf("Iterations beginning .....\n\n");
12335: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12336:
12337: iteri=0;
12338: while (rval == GSL_CONTINUE){
12339: iteri++;
12340: status = gsl_multimin_fminimizer_iterate(sfm);
12341:
12342: if (status) printf("error: %s\n", gsl_strerror (status));
12343: fflush(0);
12344:
12345: if (status)
12346: break;
12347:
12348: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12349: ssval = gsl_multimin_fminimizer_size (sfm);
12350:
12351: if (rval == GSL_SUCCESS)
12352: printf ("converged to a local maximum at\n");
12353:
12354: printf("%5d ", iteri);
12355: for (it = 0; it < NDIM; it++){
12356: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12357: }
12358: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12359: }
12360:
12361: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12362:
12363: gsl_vector_free(x); /* initial values */
12364: gsl_vector_free(ss); /* inital step size */
12365: for (it=0; it<NDIM; it++){
12366: p[it+1]=gsl_vector_get(sfm->x,it);
12367: fprintf(ficrespow," %.12lf", p[it]);
12368: }
12369: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12370: #endif
12371: #ifdef POWELL
12372: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12373: #endif
1.126 brouard 12374: fclose(ficrespow);
12375:
1.203 brouard 12376: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12377:
12378: for(i=1; i <=NDIM; i++)
12379: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12380: matcov[i][j]=matcov[j][i];
1.126 brouard 12381:
12382: printf("\nCovariance matrix\n ");
1.203 brouard 12383: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12384: for(i=1; i <=NDIM; i++) {
12385: for(j=1;j<=NDIM;j++){
1.220 brouard 12386: printf("%f ",matcov[i][j]);
12387: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12388: }
1.203 brouard 12389: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12390: }
12391:
12392: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12393: for (i=1;i<=NDIM;i++) {
1.126 brouard 12394: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12395: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12396: }
1.302 brouard 12397: lsurv=vector(agegomp,AGESUP);
12398: lpop=vector(agegomp,AGESUP);
12399: tpop=vector(agegomp,AGESUP);
1.126 brouard 12400: lsurv[agegomp]=100000;
12401:
12402: for (k=agegomp;k<=AGESUP;k++) {
12403: agemortsup=k;
12404: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12405: }
12406:
12407: for (k=agegomp;k<agemortsup;k++)
12408: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12409:
12410: for (k=agegomp;k<agemortsup;k++){
12411: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12412: sumlpop=sumlpop+lpop[k];
12413: }
12414:
12415: tpop[agegomp]=sumlpop;
12416: for (k=agegomp;k<(agemortsup-3);k++){
12417: /* tpop[k+1]=2;*/
12418: tpop[k+1]=tpop[k]-lpop[k];
12419: }
12420:
12421:
12422: printf("\nAge lx qx dx Lx Tx e(x)\n");
12423: for (k=agegomp;k<(agemortsup-2);k++)
12424: 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]);
12425:
12426:
12427: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12428: ageminpar=50;
12429: agemaxpar=100;
1.194 brouard 12430: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12431: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12432: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12433: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12434: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12435: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12436: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12437: }else{
12438: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12439: 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 12440: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12441: }
1.201 brouard 12442: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12443: stepm, weightopt,\
12444: model,imx,p,matcov,agemortsup);
12445:
1.302 brouard 12446: free_vector(lsurv,agegomp,AGESUP);
12447: free_vector(lpop,agegomp,AGESUP);
12448: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12449: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12450: free_ivector(dcwave,firstobs,lastobs);
12451: free_vector(agecens,firstobs,lastobs);
12452: free_vector(ageexmed,firstobs,lastobs);
12453: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12454: #ifdef GSL
1.136 brouard 12455: #endif
1.186 brouard 12456: } /* Endof if mle==-3 mortality only */
1.205 brouard 12457: /* Standard */
12458: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12459: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12460: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12461: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12462: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12463: for (k=1; k<=npar;k++)
12464: printf(" %d %8.5f",k,p[k]);
12465: printf("\n");
1.205 brouard 12466: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12467: /* mlikeli uses func not funcone */
1.247 brouard 12468: /* for(i=1;i<nlstate;i++){ */
12469: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12470: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12471: /* } */
1.205 brouard 12472: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12473: }
12474: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12475: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12476: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12477: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12478: }
12479: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12480: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12481: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12482: for (k=1; k<=npar;k++)
12483: printf(" %d %8.5f",k,p[k]);
12484: printf("\n");
12485:
12486: /*--------- results files --------------*/
1.283 brouard 12487: /* 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 12488:
12489:
12490: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12491: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 12492: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12493:
12494: printf("#model= 1 + age ");
12495: fprintf(ficres,"#model= 1 + age ");
12496: fprintf(ficlog,"#model= 1 + age ");
12497: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
12498: </ul>", model);
12499:
12500: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
12501: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
12502: if(nagesqr==1){
12503: printf(" + age*age ");
12504: fprintf(ficres," + age*age ");
12505: fprintf(ficlog," + age*age ");
12506: fprintf(fichtm, "<th>+ age*age</th>");
12507: }
12508: for(j=1;j <=ncovmodel-2;j++){
12509: if(Typevar[j]==0) {
12510: printf(" + V%d ",Tvar[j]);
12511: fprintf(ficres," + V%d ",Tvar[j]);
12512: fprintf(ficlog," + V%d ",Tvar[j]);
12513: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12514: }else if(Typevar[j]==1) {
12515: printf(" + V%d*age ",Tvar[j]);
12516: fprintf(ficres," + V%d*age ",Tvar[j]);
12517: fprintf(ficlog," + V%d*age ",Tvar[j]);
12518: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12519: }else if(Typevar[j]==2) {
12520: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12521: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12522: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12523: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12524: }
12525: }
12526: printf("\n");
12527: fprintf(ficres,"\n");
12528: fprintf(ficlog,"\n");
12529: fprintf(fichtm, "</tr>");
12530: fprintf(fichtm, "\n");
12531:
12532:
1.126 brouard 12533: for(i=1,jk=1; i <=nlstate; i++){
12534: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12535: if (k != i) {
1.319 brouard 12536: fprintf(fichtm, "<tr>");
1.225 brouard 12537: printf("%d%d ",i,k);
12538: fprintf(ficlog,"%d%d ",i,k);
12539: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 12540: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12541: for(j=1; j <=ncovmodel; j++){
12542: printf("%12.7f ",p[jk]);
12543: fprintf(ficlog,"%12.7f ",p[jk]);
12544: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 12545: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 12546: jk++;
12547: }
12548: printf("\n");
12549: fprintf(ficlog,"\n");
12550: fprintf(ficres,"\n");
1.319 brouard 12551: fprintf(fichtm, "</tr>\n");
1.225 brouard 12552: }
1.126 brouard 12553: }
12554: }
1.319 brouard 12555: /* fprintf(fichtm,"</tr>\n"); */
12556: fprintf(fichtm,"</table>\n");
12557: fprintf(fichtm, "\n");
12558:
1.203 brouard 12559: if(mle != 0){
12560: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12561: ftolhess=ftol; /* Usually correct */
1.203 brouard 12562: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12563: 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");
12564: 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 12565: 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 12566: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
12567: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
12568: if(nagesqr==1){
12569: printf(" + age*age ");
12570: fprintf(ficres," + age*age ");
12571: fprintf(ficlog," + age*age ");
12572: fprintf(fichtm, "<th>+ age*age</th>");
12573: }
12574: for(j=1;j <=ncovmodel-2;j++){
12575: if(Typevar[j]==0) {
12576: printf(" + V%d ",Tvar[j]);
12577: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12578: }else if(Typevar[j]==1) {
12579: printf(" + V%d*age ",Tvar[j]);
12580: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12581: }else if(Typevar[j]==2) {
12582: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12583: }
12584: }
12585: fprintf(fichtm, "</tr>\n");
12586:
1.203 brouard 12587: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12588: for(k=1; k <=(nlstate+ndeath); k++){
12589: if (k != i) {
1.319 brouard 12590: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 12591: printf("%d%d ",i,k);
12592: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 12593: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12594: for(j=1; j <=ncovmodel; j++){
1.319 brouard 12595: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 ! brouard 12596: 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]));
! 12597: 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 12598: if(fabs(wald) > 1.96){
1.321 brouard 12599: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 12600: }else{
12601: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
12602: }
1.324 ! brouard 12603: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 12604: 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 12605: jk++;
12606: }
12607: printf("\n");
12608: fprintf(ficlog,"\n");
1.319 brouard 12609: fprintf(fichtm, "</tr>\n");
1.225 brouard 12610: }
12611: }
1.193 brouard 12612: }
1.203 brouard 12613: } /* end of hesscov and Wald tests */
1.319 brouard 12614: fprintf(fichtm,"</table>\n");
1.225 brouard 12615:
1.203 brouard 12616: /* */
1.126 brouard 12617: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12618: printf("# Scales (for hessian or gradient estimation)\n");
12619: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12620: for(i=1,jk=1; i <=nlstate; i++){
12621: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12622: if (j!=i) {
12623: fprintf(ficres,"%1d%1d",i,j);
12624: printf("%1d%1d",i,j);
12625: fprintf(ficlog,"%1d%1d",i,j);
12626: for(k=1; k<=ncovmodel;k++){
12627: printf(" %.5e",delti[jk]);
12628: fprintf(ficlog," %.5e",delti[jk]);
12629: fprintf(ficres," %.5e",delti[jk]);
12630: jk++;
12631: }
12632: printf("\n");
12633: fprintf(ficlog,"\n");
12634: fprintf(ficres,"\n");
12635: }
1.126 brouard 12636: }
12637: }
12638:
12639: 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 12640: if(mle >= 1) /* To big for the screen */
1.126 brouard 12641: 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");
12642: 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");
12643: /* # 121 Var(a12)\n\ */
12644: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12645: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12646: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12647: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12648: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12649: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12650: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12651:
12652:
12653: /* Just to have a covariance matrix which will be more understandable
12654: even is we still don't want to manage dictionary of variables
12655: */
12656: for(itimes=1;itimes<=2;itimes++){
12657: jj=0;
12658: for(i=1; i <=nlstate; i++){
1.225 brouard 12659: for(j=1; j <=nlstate+ndeath; j++){
12660: if(j==i) continue;
12661: for(k=1; k<=ncovmodel;k++){
12662: jj++;
12663: ca[0]= k+'a'-1;ca[1]='\0';
12664: if(itimes==1){
12665: if(mle>=1)
12666: printf("#%1d%1d%d",i,j,k);
12667: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12668: fprintf(ficres,"#%1d%1d%d",i,j,k);
12669: }else{
12670: if(mle>=1)
12671: printf("%1d%1d%d",i,j,k);
12672: fprintf(ficlog,"%1d%1d%d",i,j,k);
12673: fprintf(ficres,"%1d%1d%d",i,j,k);
12674: }
12675: ll=0;
12676: for(li=1;li <=nlstate; li++){
12677: for(lj=1;lj <=nlstate+ndeath; lj++){
12678: if(lj==li) continue;
12679: for(lk=1;lk<=ncovmodel;lk++){
12680: ll++;
12681: if(ll<=jj){
12682: cb[0]= lk +'a'-1;cb[1]='\0';
12683: if(ll<jj){
12684: if(itimes==1){
12685: if(mle>=1)
12686: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12687: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12688: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12689: }else{
12690: if(mle>=1)
12691: printf(" %.5e",matcov[jj][ll]);
12692: fprintf(ficlog," %.5e",matcov[jj][ll]);
12693: fprintf(ficres," %.5e",matcov[jj][ll]);
12694: }
12695: }else{
12696: if(itimes==1){
12697: if(mle>=1)
12698: printf(" Var(%s%1d%1d)",ca,i,j);
12699: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12700: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12701: }else{
12702: if(mle>=1)
12703: printf(" %.7e",matcov[jj][ll]);
12704: fprintf(ficlog," %.7e",matcov[jj][ll]);
12705: fprintf(ficres," %.7e",matcov[jj][ll]);
12706: }
12707: }
12708: }
12709: } /* end lk */
12710: } /* end lj */
12711: } /* end li */
12712: if(mle>=1)
12713: printf("\n");
12714: fprintf(ficlog,"\n");
12715: fprintf(ficres,"\n");
12716: numlinepar++;
12717: } /* end k*/
12718: } /*end j */
1.126 brouard 12719: } /* end i */
12720: } /* end itimes */
12721:
12722: fflush(ficlog);
12723: fflush(ficres);
1.225 brouard 12724: while(fgets(line, MAXLINE, ficpar)) {
12725: /* If line starts with a # it is a comment */
12726: if (line[0] == '#') {
12727: numlinepar++;
12728: fputs(line,stdout);
12729: fputs(line,ficparo);
12730: fputs(line,ficlog);
1.299 brouard 12731: fputs(line,ficres);
1.225 brouard 12732: continue;
12733: }else
12734: break;
12735: }
12736:
1.209 brouard 12737: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12738: /* ungetc(c,ficpar); */
12739: /* fgets(line, MAXLINE, ficpar); */
12740: /* fputs(line,stdout); */
12741: /* fputs(line,ficparo); */
12742: /* } */
12743: /* ungetc(c,ficpar); */
1.126 brouard 12744:
12745: estepm=0;
1.209 brouard 12746: 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 12747:
12748: if (num_filled != 6) {
12749: 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);
12750: 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);
12751: goto end;
12752: }
12753: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12754: }
12755: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12756: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12757:
1.209 brouard 12758: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12759: if (estepm==0 || estepm < stepm) estepm=stepm;
12760: if (fage <= 2) {
12761: bage = ageminpar;
12762: fage = agemaxpar;
12763: }
12764:
12765: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12766: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12767: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12768:
1.186 brouard 12769: /* Other stuffs, more or less useful */
1.254 brouard 12770: while(fgets(line, MAXLINE, ficpar)) {
12771: /* If line starts with a # it is a comment */
12772: if (line[0] == '#') {
12773: numlinepar++;
12774: fputs(line,stdout);
12775: fputs(line,ficparo);
12776: fputs(line,ficlog);
1.299 brouard 12777: fputs(line,ficres);
1.254 brouard 12778: continue;
12779: }else
12780: break;
12781: }
12782:
12783: 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){
12784:
12785: if (num_filled != 7) {
12786: 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);
12787: 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);
12788: goto end;
12789: }
12790: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12791: 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);
12792: 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);
12793: 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 12794: }
1.254 brouard 12795:
12796: while(fgets(line, MAXLINE, ficpar)) {
12797: /* If line starts with a # it is a comment */
12798: if (line[0] == '#') {
12799: numlinepar++;
12800: fputs(line,stdout);
12801: fputs(line,ficparo);
12802: fputs(line,ficlog);
1.299 brouard 12803: fputs(line,ficres);
1.254 brouard 12804: continue;
12805: }else
12806: break;
1.126 brouard 12807: }
12808:
12809:
12810: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12811: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12812:
1.254 brouard 12813: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12814: if (num_filled != 1) {
12815: 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);
12816: 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);
12817: goto end;
12818: }
12819: printf("pop_based=%d\n",popbased);
12820: fprintf(ficlog,"pop_based=%d\n",popbased);
12821: fprintf(ficparo,"pop_based=%d\n",popbased);
12822: fprintf(ficres,"pop_based=%d\n",popbased);
12823: }
12824:
1.258 brouard 12825: /* Results */
1.307 brouard 12826: endishere=0;
1.258 brouard 12827: nresult=0;
1.308 brouard 12828: parameterline=0;
1.258 brouard 12829: do{
12830: if(!fgets(line, MAXLINE, ficpar)){
12831: endishere=1;
1.308 brouard 12832: parameterline=15;
1.258 brouard 12833: }else if (line[0] == '#') {
12834: /* If line starts with a # it is a comment */
1.254 brouard 12835: numlinepar++;
12836: fputs(line,stdout);
12837: fputs(line,ficparo);
12838: fputs(line,ficlog);
1.299 brouard 12839: fputs(line,ficres);
1.254 brouard 12840: continue;
1.258 brouard 12841: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12842: parameterline=11;
1.296 brouard 12843: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12844: parameterline=12;
1.307 brouard 12845: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 12846: parameterline=13;
1.307 brouard 12847: }
1.258 brouard 12848: else{
12849: parameterline=14;
1.254 brouard 12850: }
1.308 brouard 12851: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 12852: case 11:
1.296 brouard 12853: 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)){
12854: 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 12855: 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);
12856: 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);
12857: 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);
12858: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12859: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12860: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12861: prvforecast = 1;
12862: }
12863: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 12864: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12865: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12866: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12867: prvforecast = 2;
12868: }
12869: else {
12870: 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);
12871: 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);
12872: goto end;
1.258 brouard 12873: }
1.254 brouard 12874: break;
1.258 brouard 12875: case 12:
1.296 brouard 12876: 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)){
12877: 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);
12878: 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);
12879: 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);
12880: 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);
12881: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12882: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12883: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12884: prvbackcast = 1;
12885: }
12886: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 12887: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12888: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12889: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12890: prvbackcast = 2;
12891: }
12892: else {
12893: 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);
12894: 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);
12895: goto end;
1.258 brouard 12896: }
1.230 brouard 12897: break;
1.258 brouard 12898: case 13:
1.307 brouard 12899: num_filled=sscanf(line,"result:%[^\n]\n",resultline);
12900: nresult++; /* Sum of resultlines */
12901: printf("Result %d: result:%s\n",nresult, resultline);
1.318 brouard 12902: if(nresult > MAXRESULTLINESPONE-1){
12903: 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);
12904: 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 12905: goto end;
12906: }
1.310 brouard 12907: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 12908: fprintf(ficparo,"result: %s\n",resultline);
12909: fprintf(ficres,"result: %s\n",resultline);
12910: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 12911: } else
12912: goto end;
1.307 brouard 12913: break;
12914: case 14:
12915: printf("Error: Unknown command '%s'\n",line);
12916: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 12917: if(line[0] == ' ' || line[0] == '\n'){
12918: printf("It should not be an empty line '%s'\n",line);
12919: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
12920: }
1.307 brouard 12921: if(ncovmodel >=2 && nresult==0 ){
12922: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
12923: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12924: }
1.307 brouard 12925: /* goto end; */
12926: break;
1.308 brouard 12927: case 15:
12928: printf("End of resultlines.\n");
12929: fprintf(ficlog,"End of resultlines.\n");
12930: break;
12931: default: /* parameterline =0 */
1.307 brouard 12932: nresult=1;
12933: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 12934: } /* End switch parameterline */
12935: }while(endishere==0); /* End do */
1.126 brouard 12936:
1.230 brouard 12937: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12938: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12939:
12940: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12941: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12942: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12943: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12944: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12945: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12946: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12947: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12948: }else{
1.270 brouard 12949: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12950: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12951: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12952: if(prvforecast==1){
12953: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12954: jprojd=jproj1;
12955: mprojd=mproj1;
12956: anprojd=anproj1;
12957: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12958: jprojf=jproj2;
12959: mprojf=mproj2;
12960: anprojf=anproj2;
12961: } else if(prvforecast == 2){
12962: dateprojd=dateintmean;
12963: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12964: dateprojf=dateintmean+yrfproj;
12965: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12966: }
12967: if(prvbackcast==1){
12968: datebackd=(jback1+12*mback1+365*anback1)/365;
12969: jbackd=jback1;
12970: mbackd=mback1;
12971: anbackd=anback1;
12972: datebackf=(jback2+12*mback2+365*anback2)/365;
12973: jbackf=jback2;
12974: mbackf=mback2;
12975: anbackf=anback2;
12976: } else if(prvbackcast == 2){
12977: datebackd=dateintmean;
12978: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12979: datebackf=dateintmean-yrbproj;
12980: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12981: }
12982:
12983: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12984: }
12985: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12986: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12987: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12988:
1.225 brouard 12989: /*------------ free_vector -------------*/
12990: /* chdir(path); */
1.220 brouard 12991:
1.215 brouard 12992: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12993: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12994: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12995: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12996: free_lvector(num,firstobs,lastobs);
12997: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12998: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12999: /*free_matrix(covar,1,NCOVMAX,1,n);*/
13000: fclose(ficparo);
13001: fclose(ficres);
1.220 brouard 13002:
13003:
1.186 brouard 13004: /* Other results (useful)*/
1.220 brouard 13005:
13006:
1.126 brouard 13007: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13008: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13009: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 13010: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13011: fclose(ficrespl);
13012:
13013: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13014: /*#include "hpijx.h"*/
13015: hPijx(p, bage, fage);
1.145 brouard 13016: fclose(ficrespij);
1.227 brouard 13017:
1.220 brouard 13018: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 13019: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 13020: k=1;
1.126 brouard 13021: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13022:
1.269 brouard 13023: /* Prevalence for each covariate combination in probs[age][status][cov] */
13024: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13025: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 13026: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 13027: for(k=1;k<=ncovcombmax;k++)
13028: probs[i][j][k]=0.;
1.269 brouard 13029: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
13030: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 13031: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 13032: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13033: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 13034: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 13035: for(k=1;k<=ncovcombmax;k++)
13036: mobaverages[i][j][k]=0.;
1.219 brouard 13037: mobaverage=mobaverages;
13038: if (mobilav!=0) {
1.235 brouard 13039: printf("Movingaveraging observed prevalence\n");
1.258 brouard 13040: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 13041: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
13042: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
13043: printf(" Error in movingaverage mobilav=%d\n",mobilav);
13044: }
1.269 brouard 13045: } else if (mobilavproj !=0) {
1.235 brouard 13046: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 13047: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 13048: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
13049: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
13050: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
13051: }
1.269 brouard 13052: }else{
13053: printf("Internal error moving average\n");
13054: fflush(stdout);
13055: exit(1);
1.219 brouard 13056: }
13057: }/* end if moving average */
1.227 brouard 13058:
1.126 brouard 13059: /*---------- Forecasting ------------------*/
1.296 brouard 13060: if(prevfcast==1){
13061: /* /\* if(stepm ==1){*\/ */
13062: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13063: /*This done previously after freqsummary.*/
13064: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
13065: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
13066:
13067: /* } else if (prvforecast==2){ */
13068: /* /\* if(stepm ==1){*\/ */
13069: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13070: /* } */
13071: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
13072: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 13073: }
1.269 brouard 13074:
1.296 brouard 13075: /* Prevbcasting */
13076: if(prevbcast==1){
1.219 brouard 13077: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13078: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13079: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13080:
13081: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
13082:
13083: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 13084:
1.219 brouard 13085: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
13086: fclose(ficresplb);
13087:
1.222 brouard 13088: hBijx(p, bage, fage, mobaverage);
13089: fclose(ficrespijb);
1.219 brouard 13090:
1.296 brouard 13091: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
13092: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
13093: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
13094: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
13095: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
13096: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
13097:
13098:
1.269 brouard 13099: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13100:
13101:
1.269 brouard 13102: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 13103: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13104: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13105: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 13106: } /* end Prevbcasting */
1.268 brouard 13107:
1.186 brouard 13108:
13109: /* ------ Other prevalence ratios------------ */
1.126 brouard 13110:
1.215 brouard 13111: free_ivector(wav,1,imx);
13112: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
13113: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
13114: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 13115:
13116:
1.127 brouard 13117: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 13118:
1.201 brouard 13119: strcpy(filerese,"E_");
13120: strcat(filerese,fileresu);
1.126 brouard 13121: if((ficreseij=fopen(filerese,"w"))==NULL) {
13122: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13123: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13124: }
1.208 brouard 13125: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
13126: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 13127:
13128: pstamp(ficreseij);
1.219 brouard 13129:
1.235 brouard 13130: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13131: if (cptcovn < 1){i1=1;}
13132:
13133: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13134: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13135: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13136: continue;
1.219 brouard 13137: fprintf(ficreseij,"\n#****** ");
1.235 brouard 13138: printf("\n#****** ");
1.225 brouard 13139: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 13140: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 13141: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13142: }
13143: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13144: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13145: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 13146: }
13147: fprintf(ficreseij,"******\n");
1.235 brouard 13148: printf("******\n");
1.219 brouard 13149:
13150: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13151: oldm=oldms;savm=savms;
1.235 brouard 13152: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 13153:
1.219 brouard 13154: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 13155: }
13156: fclose(ficreseij);
1.208 brouard 13157: printf("done evsij\n");fflush(stdout);
13158: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 13159:
1.218 brouard 13160:
1.227 brouard 13161: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 13162:
1.201 brouard 13163: strcpy(filerest,"T_");
13164: strcat(filerest,fileresu);
1.127 brouard 13165: if((ficrest=fopen(filerest,"w"))==NULL) {
13166: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
13167: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
13168: }
1.208 brouard 13169: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
13170: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 13171: strcpy(fileresstde,"STDE_");
13172: strcat(fileresstde,fileresu);
1.126 brouard 13173: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 13174: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
13175: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 13176: }
1.227 brouard 13177: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
13178: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 13179:
1.201 brouard 13180: strcpy(filerescve,"CVE_");
13181: strcat(filerescve,fileresu);
1.126 brouard 13182: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 13183: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
13184: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 13185: }
1.227 brouard 13186: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
13187: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 13188:
1.201 brouard 13189: strcpy(fileresv,"V_");
13190: strcat(fileresv,fileresu);
1.126 brouard 13191: if((ficresvij=fopen(fileresv,"w"))==NULL) {
13192: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
13193: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
13194: }
1.227 brouard 13195: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
13196: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 13197:
1.235 brouard 13198: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13199: if (cptcovn < 1){i1=1;}
13200:
13201: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13202: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13203: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13204: continue;
1.321 brouard 13205: printf("\n# model %s \n#****** Result for:", model);
13206: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
13207: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.227 brouard 13208: for(j=1;j<=cptcoveff;j++){
13209: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13210: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13211: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13212: }
1.235 brouard 13213: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13214: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13215: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13216: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13217: }
1.208 brouard 13218: fprintf(ficrest,"******\n");
1.227 brouard 13219: fprintf(ficlog,"******\n");
13220: printf("******\n");
1.208 brouard 13221:
13222: fprintf(ficresstdeij,"\n#****** ");
13223: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 13224: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 13225: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13226: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 13227: }
1.235 brouard 13228: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13229: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13230: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13231: }
1.208 brouard 13232: fprintf(ficresstdeij,"******\n");
13233: fprintf(ficrescveij,"******\n");
13234:
13235: fprintf(ficresvij,"\n#****** ");
1.238 brouard 13236: /* pstamp(ficresvij); */
1.225 brouard 13237: for(j=1;j<=cptcoveff;j++)
1.227 brouard 13238: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 13239: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13240: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13241: }
1.208 brouard 13242: fprintf(ficresvij,"******\n");
13243:
13244: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13245: oldm=oldms;savm=savms;
1.235 brouard 13246: printf(" cvevsij ");
13247: fprintf(ficlog, " cvevsij ");
13248: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 13249: printf(" end cvevsij \n ");
13250: fprintf(ficlog, " end cvevsij \n ");
13251:
13252: /*
13253: */
13254: /* goto endfree; */
13255:
13256: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13257: pstamp(ficrest);
13258:
1.269 brouard 13259: epj=vector(1,nlstate+1);
1.208 brouard 13260: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 13261: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
13262: cptcod= 0; /* To be deleted */
13263: printf("varevsij vpopbased=%d \n",vpopbased);
13264: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 13265: 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 13266: 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 ");
13267: if(vpopbased==1)
13268: 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);
13269: else
1.288 brouard 13270: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13271: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
13272: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
13273: fprintf(ficrest,"\n");
13274: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 13275: printf("Computing age specific forward period (stable) prevalences in each health state \n");
13276: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13277: for(age=bage; age <=fage ;age++){
1.235 brouard 13278: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 13279: if (vpopbased==1) {
13280: if(mobilav ==0){
13281: for(i=1; i<=nlstate;i++)
13282: prlim[i][i]=probs[(int)age][i][k];
13283: }else{ /* mobilav */
13284: for(i=1; i<=nlstate;i++)
13285: prlim[i][i]=mobaverage[(int)age][i][k];
13286: }
13287: }
1.219 brouard 13288:
1.227 brouard 13289: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
13290: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
13291: /* printf(" age %4.0f ",age); */
13292: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
13293: for(i=1, epj[j]=0.;i <=nlstate;i++) {
13294: epj[j] += prlim[i][i]*eij[i][j][(int)age];
13295: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
13296: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
13297: }
13298: epj[nlstate+1] +=epj[j];
13299: }
13300: /* printf(" age %4.0f \n",age); */
1.219 brouard 13301:
1.227 brouard 13302: for(i=1, vepp=0.;i <=nlstate;i++)
13303: for(j=1;j <=nlstate;j++)
13304: vepp += vareij[i][j][(int)age];
13305: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
13306: for(j=1;j <=nlstate;j++){
13307: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
13308: }
13309: fprintf(ficrest,"\n");
13310: }
1.208 brouard 13311: } /* End vpopbased */
1.269 brouard 13312: free_vector(epj,1,nlstate+1);
1.208 brouard 13313: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
13314: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 13315: printf("done selection\n");fflush(stdout);
13316: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 13317:
1.235 brouard 13318: } /* End k selection */
1.227 brouard 13319:
13320: printf("done State-specific expectancies\n");fflush(stdout);
13321: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
13322:
1.288 brouard 13323: /* variance-covariance of forward period prevalence*/
1.269 brouard 13324: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13325:
1.227 brouard 13326:
1.290 brouard 13327: free_vector(weight,firstobs,lastobs);
1.227 brouard 13328: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 13329: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
13330: free_matrix(anint,1,maxwav,firstobs,lastobs);
13331: free_matrix(mint,1,maxwav,firstobs,lastobs);
13332: free_ivector(cod,firstobs,lastobs);
1.227 brouard 13333: free_ivector(tab,1,NCOVMAX);
13334: fclose(ficresstdeij);
13335: fclose(ficrescveij);
13336: fclose(ficresvij);
13337: fclose(ficrest);
13338: fclose(ficpar);
13339:
13340:
1.126 brouard 13341: /*---------- End : free ----------------*/
1.219 brouard 13342: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 13343: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
13344: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 13345: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
13346: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 13347: } /* mle==-3 arrives here for freeing */
1.227 brouard 13348: /* endfree:*/
13349: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
13350: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
13351: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 13352: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
13353: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
13354: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
13355: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 13356: free_matrix(matcov,1,npar,1,npar);
13357: free_matrix(hess,1,npar,1,npar);
13358: /*free_vector(delti,1,npar);*/
13359: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13360: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 13361: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 13362: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13363:
13364: free_ivector(ncodemax,1,NCOVMAX);
13365: free_ivector(ncodemaxwundef,1,NCOVMAX);
13366: free_ivector(Dummy,-1,NCOVMAX);
13367: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 13368: free_ivector(DummyV,1,NCOVMAX);
13369: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 13370: free_ivector(Typevar,-1,NCOVMAX);
13371: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 13372: free_ivector(TvarsQ,1,NCOVMAX);
13373: free_ivector(TvarsQind,1,NCOVMAX);
13374: free_ivector(TvarsD,1,NCOVMAX);
13375: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 13376: free_ivector(TvarFD,1,NCOVMAX);
13377: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 13378: free_ivector(TvarF,1,NCOVMAX);
13379: free_ivector(TvarFind,1,NCOVMAX);
13380: free_ivector(TvarV,1,NCOVMAX);
13381: free_ivector(TvarVind,1,NCOVMAX);
13382: free_ivector(TvarA,1,NCOVMAX);
13383: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 13384: free_ivector(TvarFQ,1,NCOVMAX);
13385: free_ivector(TvarFQind,1,NCOVMAX);
13386: free_ivector(TvarVD,1,NCOVMAX);
13387: free_ivector(TvarVDind,1,NCOVMAX);
13388: free_ivector(TvarVQ,1,NCOVMAX);
13389: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 13390: free_ivector(Tvarsel,1,NCOVMAX);
13391: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 13392: free_ivector(Tposprod,1,NCOVMAX);
13393: free_ivector(Tprod,1,NCOVMAX);
13394: free_ivector(Tvaraff,1,NCOVMAX);
13395: free_ivector(invalidvarcomb,1,ncovcombmax);
13396: free_ivector(Tage,1,NCOVMAX);
13397: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 13398: free_ivector(TmodelInvind,1,NCOVMAX);
13399: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 13400:
13401: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
13402: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 13403: fflush(fichtm);
13404: fflush(ficgp);
13405:
1.227 brouard 13406:
1.126 brouard 13407: if((nberr >0) || (nbwarn>0)){
1.216 brouard 13408: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
13409: 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 13410: }else{
13411: printf("End of Imach\n");
13412: fprintf(ficlog,"End of Imach\n");
13413: }
13414: printf("See log file on %s\n",filelog);
13415: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 13416: /*(void) gettimeofday(&end_time,&tzp);*/
13417: rend_time = time(NULL);
13418: end_time = *localtime(&rend_time);
13419: /* tml = *localtime(&end_time.tm_sec); */
13420: strcpy(strtend,asctime(&end_time));
1.126 brouard 13421: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
13422: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 13423: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13424:
1.157 brouard 13425: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13426: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13427: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13428: /* printf("Total time was %d uSec.\n", total_usecs);*/
13429: /* if(fileappend(fichtm,optionfilehtm)){ */
13430: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13431: fclose(fichtm);
13432: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13433: fclose(fichtmcov);
13434: fclose(ficgp);
13435: fclose(ficlog);
13436: /*------ End -----------*/
1.227 brouard 13437:
1.281 brouard 13438:
13439: /* Executes gnuplot */
1.227 brouard 13440:
13441: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13442: #ifdef WIN32
1.227 brouard 13443: if (_chdir(pathcd) != 0)
13444: printf("Can't move to directory %s!\n",path);
13445: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13446: #else
1.227 brouard 13447: if(chdir(pathcd) != 0)
13448: printf("Can't move to directory %s!\n", path);
13449: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13450: #endif
1.126 brouard 13451: printf("Current directory %s!\n",pathcd);
13452: /*strcat(plotcmd,CHARSEPARATOR);*/
13453: sprintf(plotcmd,"gnuplot");
1.157 brouard 13454: #ifdef _WIN32
1.126 brouard 13455: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13456: #endif
13457: if(!stat(plotcmd,&info)){
1.158 brouard 13458: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13459: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13460: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13461: }else
13462: strcpy(pplotcmd,plotcmd);
1.157 brouard 13463: #ifdef __unix
1.126 brouard 13464: strcpy(plotcmd,GNUPLOTPROGRAM);
13465: if(!stat(plotcmd,&info)){
1.158 brouard 13466: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13467: }else
13468: strcpy(pplotcmd,plotcmd);
13469: #endif
13470: }else
13471: strcpy(pplotcmd,plotcmd);
13472:
13473: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13474: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13475: strcpy(pplotcmd,plotcmd);
1.227 brouard 13476:
1.126 brouard 13477: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13478: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13479: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13480: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13481: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13482: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13483: strcpy(plotcmd,pplotcmd);
13484: }
1.126 brouard 13485: }
1.158 brouard 13486: printf(" Successful, please wait...");
1.126 brouard 13487: while (z[0] != 'q') {
13488: /* chdir(path); */
1.154 brouard 13489: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13490: scanf("%s",z);
13491: /* if (z[0] == 'c') system("./imach"); */
13492: if (z[0] == 'e') {
1.158 brouard 13493: #ifdef __APPLE__
1.152 brouard 13494: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13495: #elif __linux
13496: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13497: #else
1.152 brouard 13498: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13499: #endif
13500: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13501: system(pplotcmd);
1.126 brouard 13502: }
13503: else if (z[0] == 'g') system(plotcmd);
13504: else if (z[0] == 'q') exit(0);
13505: }
1.227 brouard 13506: end:
1.126 brouard 13507: while (z[0] != 'q') {
1.195 brouard 13508: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13509: scanf("%s",z);
13510: }
1.283 brouard 13511: printf("End\n");
1.282 brouard 13512: exit(0);
1.126 brouard 13513: }
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