Annotation of imach/src/imach.c, revision 1.330
1.330 ! brouard 1: /* $Id: imach.c,v 1.329 2022/08/03 17:29:54 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.330 ! brouard 4: Revision 1.329 2022/08/03 17:29:54 brouard
! 5: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
! 6:
1.329 brouard 7: Revision 1.328 2022/07/27 17:40:48 brouard
8: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
9:
1.328 brouard 10: Revision 1.327 2022/07/27 14:47:35 brouard
11: Summary: Still a problem for one-step probabilities in case of quantitative variables
12:
1.327 brouard 13: Revision 1.326 2022/07/26 17:33:55 brouard
14: Summary: some test with nres=1
15:
1.326 brouard 16: Revision 1.325 2022/07/25 14:27:23 brouard
17: Summary: r30
18:
19: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
20: coredumped, revealed by Feiuno, thank you.
21:
1.325 brouard 22: Revision 1.324 2022/07/23 17:44:26 brouard
23: *** empty log message ***
24:
1.324 brouard 25: Revision 1.323 2022/07/22 12:30:08 brouard
26: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
27:
1.323 brouard 28: Revision 1.322 2022/07/22 12:27:48 brouard
29: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
30:
1.322 brouard 31: Revision 1.321 2022/07/22 12:04:24 brouard
32: Summary: r28
33:
34: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
35:
1.321 brouard 36: Revision 1.320 2022/06/02 05:10:11 brouard
37: *** empty log message ***
38:
1.320 brouard 39: Revision 1.319 2022/06/02 04:45:11 brouard
40: * imach.c (Module): Adding the Wald tests from the log to the main
41: htm for better display of the maximum likelihood estimators.
42:
1.319 brouard 43: Revision 1.318 2022/05/24 08:10:59 brouard
44: * imach.c (Module): Some attempts to find a bug of wrong estimates
45: of confidencce intervals with product in the equation modelC
46:
1.318 brouard 47: Revision 1.317 2022/05/15 15:06:23 brouard
48: * imach.c (Module): Some minor improvements
49:
1.317 brouard 50: Revision 1.316 2022/05/11 15:11:31 brouard
51: Summary: r27
52:
1.316 brouard 53: Revision 1.315 2022/05/11 15:06:32 brouard
54: *** empty log message ***
55:
1.315 brouard 56: Revision 1.314 2022/04/13 17:43:09 brouard
57: * imach.c (Module): Adding link to text data files
58:
1.314 brouard 59: Revision 1.313 2022/04/11 15:57:42 brouard
60: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
61:
1.313 brouard 62: Revision 1.312 2022/04/05 21:24:39 brouard
63: *** empty log message ***
64:
1.312 brouard 65: Revision 1.311 2022/04/05 21:03:51 brouard
66: Summary: Fixed quantitative covariates
67:
68: Fixed covariates (dummy or quantitative)
69: with missing values have never been allowed but are ERRORS and
70: program quits. Standard deviations of fixed covariates were
71: wrongly computed. Mean and standard deviations of time varying
72: covariates are still not computed.
73:
1.311 brouard 74: Revision 1.310 2022/03/17 08:45:53 brouard
75: Summary: 99r25
76:
77: Improving detection of errors: result lines should be compatible with
78: the model.
79:
1.310 brouard 80: Revision 1.309 2021/05/20 12:39:14 brouard
81: Summary: Version 0.99r24
82:
1.309 brouard 83: Revision 1.308 2021/03/31 13:11:57 brouard
84: Summary: Version 0.99r23
85:
86:
87: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
88:
1.308 brouard 89: Revision 1.307 2021/03/08 18:11:32 brouard
90: Summary: 0.99r22 fixed bug on result:
91:
1.307 brouard 92: Revision 1.306 2021/02/20 15:44:02 brouard
93: Summary: Version 0.99r21
94:
95: * imach.c (Module): Fix bug on quitting after result lines!
96: (Module): Version 0.99r21
97:
1.306 brouard 98: Revision 1.305 2021/02/20 15:28:30 brouard
99: * imach.c (Module): Fix bug on quitting after result lines!
100:
1.305 brouard 101: Revision 1.304 2021/02/12 11:34:20 brouard
102: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
103:
1.304 brouard 104: Revision 1.303 2021/02/11 19:50:15 brouard
105: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
106:
1.303 brouard 107: Revision 1.302 2020/02/22 21:00:05 brouard
108: * (Module): imach.c Update mle=-3 (for computing Life expectancy
109: and life table from the data without any state)
110:
1.302 brouard 111: Revision 1.301 2019/06/04 13:51:20 brouard
112: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
113:
1.301 brouard 114: Revision 1.300 2019/05/22 19:09:45 brouard
115: Summary: version 0.99r19 of May 2019
116:
1.300 brouard 117: Revision 1.299 2019/05/22 18:37:08 brouard
118: Summary: Cleaned 0.99r19
119:
1.299 brouard 120: Revision 1.298 2019/05/22 18:19:56 brouard
121: *** empty log message ***
122:
1.298 brouard 123: Revision 1.297 2019/05/22 17:56:10 brouard
124: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
125:
1.297 brouard 126: Revision 1.296 2019/05/20 13:03:18 brouard
127: Summary: Projection syntax simplified
128:
129:
130: We can now start projections, forward or backward, from the mean date
131: of inteviews up to or down to a number of years of projection:
132: prevforecast=1 yearsfproj=15.3 mobil_average=0
133: or
134: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
135: or
136: prevbackcast=1 yearsbproj=12.3 mobil_average=1
137: or
138: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
139:
1.296 brouard 140: Revision 1.295 2019/05/18 09:52:50 brouard
141: Summary: doxygen tex bug
142:
1.295 brouard 143: Revision 1.294 2019/05/16 14:54:33 brouard
144: Summary: There was some wrong lines added
145:
1.294 brouard 146: Revision 1.293 2019/05/09 15:17:34 brouard
147: *** empty log message ***
148:
1.293 brouard 149: Revision 1.292 2019/05/09 14:17:20 brouard
150: Summary: Some updates
151:
1.292 brouard 152: Revision 1.291 2019/05/09 13:44:18 brouard
153: Summary: Before ncovmax
154:
1.291 brouard 155: Revision 1.290 2019/05/09 13:39:37 brouard
156: Summary: 0.99r18 unlimited number of individuals
157:
158: 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.
159:
1.290 brouard 160: Revision 1.289 2018/12/13 09:16:26 brouard
161: Summary: Bug for young ages (<-30) will be in r17
162:
1.289 brouard 163: Revision 1.288 2018/05/02 20:58:27 brouard
164: Summary: Some bugs fixed
165:
1.288 brouard 166: Revision 1.287 2018/05/01 17:57:25 brouard
167: Summary: Bug fixed by providing frequencies only for non missing covariates
168:
1.287 brouard 169: Revision 1.286 2018/04/27 14:27:04 brouard
170: Summary: some minor bugs
171:
1.286 brouard 172: Revision 1.285 2018/04/21 21:02:16 brouard
173: Summary: Some bugs fixed, valgrind tested
174:
1.285 brouard 175: Revision 1.284 2018/04/20 05:22:13 brouard
176: Summary: Computing mean and stdeviation of fixed quantitative variables
177:
1.284 brouard 178: Revision 1.283 2018/04/19 14:49:16 brouard
179: Summary: Some minor bugs fixed
180:
1.283 brouard 181: Revision 1.282 2018/02/27 22:50:02 brouard
182: *** empty log message ***
183:
1.282 brouard 184: Revision 1.281 2018/02/27 19:25:23 brouard
185: Summary: Adding second argument for quitting
186:
1.281 brouard 187: Revision 1.280 2018/02/21 07:58:13 brouard
188: Summary: 0.99r15
189:
190: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
191:
1.280 brouard 192: Revision 1.279 2017/07/20 13:35:01 brouard
193: Summary: temporary working
194:
1.279 brouard 195: Revision 1.278 2017/07/19 14:09:02 brouard
196: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
197:
1.278 brouard 198: Revision 1.277 2017/07/17 08:53:49 brouard
199: Summary: BOM files can be read now
200:
1.277 brouard 201: Revision 1.276 2017/06/30 15:48:31 brouard
202: Summary: Graphs improvements
203:
1.276 brouard 204: Revision 1.275 2017/06/30 13:39:33 brouard
205: Summary: Saito's color
206:
1.275 brouard 207: Revision 1.274 2017/06/29 09:47:08 brouard
208: Summary: Version 0.99r14
209:
1.274 brouard 210: Revision 1.273 2017/06/27 11:06:02 brouard
211: Summary: More documentation on projections
212:
1.273 brouard 213: Revision 1.272 2017/06/27 10:22:40 brouard
214: Summary: Color of backprojection changed from 6 to 5(yellow)
215:
1.272 brouard 216: Revision 1.271 2017/06/27 10:17:50 brouard
217: Summary: Some bug with rint
218:
1.271 brouard 219: Revision 1.270 2017/05/24 05:45:29 brouard
220: *** empty log message ***
221:
1.270 brouard 222: Revision 1.269 2017/05/23 08:39:25 brouard
223: Summary: Code into subroutine, cleanings
224:
1.269 brouard 225: Revision 1.268 2017/05/18 20:09:32 brouard
226: Summary: backprojection and confidence intervals of backprevalence
227:
1.268 brouard 228: Revision 1.267 2017/05/13 10:25:05 brouard
229: Summary: temporary save for backprojection
230:
1.267 brouard 231: Revision 1.266 2017/05/13 07:26:12 brouard
232: Summary: Version 0.99r13 (improvements and bugs fixed)
233:
1.266 brouard 234: Revision 1.265 2017/04/26 16:22:11 brouard
235: Summary: imach 0.99r13 Some bugs fixed
236:
1.265 brouard 237: Revision 1.264 2017/04/26 06:01:29 brouard
238: Summary: Labels in graphs
239:
1.264 brouard 240: Revision 1.263 2017/04/24 15:23:15 brouard
241: Summary: to save
242:
1.263 brouard 243: Revision 1.262 2017/04/18 16:48:12 brouard
244: *** empty log message ***
245:
1.262 brouard 246: Revision 1.261 2017/04/05 10:14:09 brouard
247: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
248:
1.261 brouard 249: Revision 1.260 2017/04/04 17:46:59 brouard
250: Summary: Gnuplot indexations fixed (humm)
251:
1.260 brouard 252: Revision 1.259 2017/04/04 13:01:16 brouard
253: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
254:
1.259 brouard 255: Revision 1.258 2017/04/03 10:17:47 brouard
256: Summary: Version 0.99r12
257:
258: Some cleanings, conformed with updated documentation.
259:
1.258 brouard 260: Revision 1.257 2017/03/29 16:53:30 brouard
261: Summary: Temp
262:
1.257 brouard 263: Revision 1.256 2017/03/27 05:50:23 brouard
264: Summary: Temporary
265:
1.256 brouard 266: Revision 1.255 2017/03/08 16:02:28 brouard
267: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
268:
1.255 brouard 269: Revision 1.254 2017/03/08 07:13:00 brouard
270: Summary: Fixing data parameter line
271:
1.254 brouard 272: Revision 1.253 2016/12/15 11:59:41 brouard
273: Summary: 0.99 in progress
274:
1.253 brouard 275: Revision 1.252 2016/09/15 21:15:37 brouard
276: *** empty log message ***
277:
1.252 brouard 278: Revision 1.251 2016/09/15 15:01:13 brouard
279: Summary: not working
280:
1.251 brouard 281: Revision 1.250 2016/09/08 16:07:27 brouard
282: Summary: continue
283:
1.250 brouard 284: Revision 1.249 2016/09/07 17:14:18 brouard
285: Summary: Starting values from frequencies
286:
1.249 brouard 287: Revision 1.248 2016/09/07 14:10:18 brouard
288: *** empty log message ***
289:
1.248 brouard 290: Revision 1.247 2016/09/02 11:11:21 brouard
291: *** empty log message ***
292:
1.247 brouard 293: Revision 1.246 2016/09/02 08:49:22 brouard
294: *** empty log message ***
295:
1.246 brouard 296: Revision 1.245 2016/09/02 07:25:01 brouard
297: *** empty log message ***
298:
1.245 brouard 299: Revision 1.244 2016/09/02 07:17:34 brouard
300: *** empty log message ***
301:
1.244 brouard 302: Revision 1.243 2016/09/02 06:45:35 brouard
303: *** empty log message ***
304:
1.243 brouard 305: Revision 1.242 2016/08/30 15:01:20 brouard
306: Summary: Fixing a lots
307:
1.242 brouard 308: Revision 1.241 2016/08/29 17:17:25 brouard
309: Summary: gnuplot problem in Back projection to fix
310:
1.241 brouard 311: Revision 1.240 2016/08/29 07:53:18 brouard
312: Summary: Better
313:
1.240 brouard 314: Revision 1.239 2016/08/26 15:51:03 brouard
315: Summary: Improvement in Powell output in order to copy and paste
316:
317: Author:
318:
1.239 brouard 319: Revision 1.238 2016/08/26 14:23:35 brouard
320: Summary: Starting tests of 0.99
321:
1.238 brouard 322: Revision 1.237 2016/08/26 09:20:19 brouard
323: Summary: to valgrind
324:
1.237 brouard 325: Revision 1.236 2016/08/25 10:50:18 brouard
326: *** empty log message ***
327:
1.236 brouard 328: Revision 1.235 2016/08/25 06:59:23 brouard
329: *** empty log message ***
330:
1.235 brouard 331: Revision 1.234 2016/08/23 16:51:20 brouard
332: *** empty log message ***
333:
1.234 brouard 334: Revision 1.233 2016/08/23 07:40:50 brouard
335: Summary: not working
336:
1.233 brouard 337: Revision 1.232 2016/08/22 14:20:21 brouard
338: Summary: not working
339:
1.232 brouard 340: Revision 1.231 2016/08/22 07:17:15 brouard
341: Summary: not working
342:
1.231 brouard 343: Revision 1.230 2016/08/22 06:55:53 brouard
344: Summary: Not working
345:
1.230 brouard 346: Revision 1.229 2016/07/23 09:45:53 brouard
347: Summary: Completing for func too
348:
1.229 brouard 349: Revision 1.228 2016/07/22 17:45:30 brouard
350: Summary: Fixing some arrays, still debugging
351:
1.227 brouard 352: Revision 1.226 2016/07/12 18:42:34 brouard
353: Summary: temp
354:
1.226 brouard 355: Revision 1.225 2016/07/12 08:40:03 brouard
356: Summary: saving but not running
357:
1.225 brouard 358: Revision 1.224 2016/07/01 13:16:01 brouard
359: Summary: Fixes
360:
1.224 brouard 361: Revision 1.223 2016/02/19 09:23:35 brouard
362: Summary: temporary
363:
1.223 brouard 364: Revision 1.222 2016/02/17 08:14:50 brouard
365: Summary: Probably last 0.98 stable version 0.98r6
366:
1.222 brouard 367: Revision 1.221 2016/02/15 23:35:36 brouard
368: Summary: minor bug
369:
1.220 brouard 370: Revision 1.219 2016/02/15 00:48:12 brouard
371: *** empty log message ***
372:
1.219 brouard 373: Revision 1.218 2016/02/12 11:29:23 brouard
374: Summary: 0.99 Back projections
375:
1.218 brouard 376: Revision 1.217 2015/12/23 17:18:31 brouard
377: Summary: Experimental backcast
378:
1.217 brouard 379: Revision 1.216 2015/12/18 17:32:11 brouard
380: Summary: 0.98r4 Warning and status=-2
381:
382: Version 0.98r4 is now:
383: - displaying an error when status is -1, date of interview unknown and date of death known;
384: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
385: Older changes concerning s=-2, dating from 2005 have been supersed.
386:
1.216 brouard 387: Revision 1.215 2015/12/16 08:52:24 brouard
388: Summary: 0.98r4 working
389:
1.215 brouard 390: Revision 1.214 2015/12/16 06:57:54 brouard
391: Summary: temporary not working
392:
1.214 brouard 393: Revision 1.213 2015/12/11 18:22:17 brouard
394: Summary: 0.98r4
395:
1.213 brouard 396: Revision 1.212 2015/11/21 12:47:24 brouard
397: Summary: minor typo
398:
1.212 brouard 399: Revision 1.211 2015/11/21 12:41:11 brouard
400: Summary: 0.98r3 with some graph of projected cross-sectional
401:
402: Author: Nicolas Brouard
403:
1.211 brouard 404: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 405: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 406: Summary: Adding ftolpl parameter
407: Author: N Brouard
408:
409: We had difficulties to get smoothed confidence intervals. It was due
410: to the period prevalence which wasn't computed accurately. The inner
411: parameter ftolpl is now an outer parameter of the .imach parameter
412: file after estepm. If ftolpl is small 1.e-4 and estepm too,
413: computation are long.
414:
1.209 brouard 415: Revision 1.208 2015/11/17 14:31:57 brouard
416: Summary: temporary
417:
1.208 brouard 418: Revision 1.207 2015/10/27 17:36:57 brouard
419: *** empty log message ***
420:
1.207 brouard 421: Revision 1.206 2015/10/24 07:14:11 brouard
422: *** empty log message ***
423:
1.206 brouard 424: Revision 1.205 2015/10/23 15:50:53 brouard
425: Summary: 0.98r3 some clarification for graphs on likelihood contributions
426:
1.205 brouard 427: Revision 1.204 2015/10/01 16:20:26 brouard
428: Summary: Some new graphs of contribution to likelihood
429:
1.204 brouard 430: Revision 1.203 2015/09/30 17:45:14 brouard
431: Summary: looking at better estimation of the hessian
432:
433: Also a better criteria for convergence to the period prevalence And
434: therefore adding the number of years needed to converge. (The
435: prevalence in any alive state shold sum to one
436:
1.203 brouard 437: Revision 1.202 2015/09/22 19:45:16 brouard
438: Summary: Adding some overall graph on contribution to likelihood. Might change
439:
1.202 brouard 440: Revision 1.201 2015/09/15 17:34:58 brouard
441: Summary: 0.98r0
442:
443: - Some new graphs like suvival functions
444: - Some bugs fixed like model=1+age+V2.
445:
1.201 brouard 446: Revision 1.200 2015/09/09 16:53:55 brouard
447: Summary: Big bug thanks to Flavia
448:
449: Even model=1+age+V2. did not work anymore
450:
1.200 brouard 451: Revision 1.199 2015/09/07 14:09:23 brouard
452: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
453:
1.199 brouard 454: Revision 1.198 2015/09/03 07:14:39 brouard
455: Summary: 0.98q5 Flavia
456:
1.198 brouard 457: Revision 1.197 2015/09/01 18:24:39 brouard
458: *** empty log message ***
459:
1.197 brouard 460: Revision 1.196 2015/08/18 23:17:52 brouard
461: Summary: 0.98q5
462:
1.196 brouard 463: Revision 1.195 2015/08/18 16:28:39 brouard
464: Summary: Adding a hack for testing purpose
465:
466: After reading the title, ftol and model lines, if the comment line has
467: a q, starting with #q, the answer at the end of the run is quit. It
468: permits to run test files in batch with ctest. The former workaround was
469: $ echo q | imach foo.imach
470:
1.195 brouard 471: Revision 1.194 2015/08/18 13:32:00 brouard
472: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
473:
1.194 brouard 474: Revision 1.193 2015/08/04 07:17:42 brouard
475: Summary: 0.98q4
476:
1.193 brouard 477: Revision 1.192 2015/07/16 16:49:02 brouard
478: Summary: Fixing some outputs
479:
1.192 brouard 480: Revision 1.191 2015/07/14 10:00:33 brouard
481: Summary: Some fixes
482:
1.191 brouard 483: Revision 1.190 2015/05/05 08:51:13 brouard
484: Summary: Adding digits in output parameters (7 digits instead of 6)
485:
486: Fix 1+age+.
487:
1.190 brouard 488: Revision 1.189 2015/04/30 14:45:16 brouard
489: Summary: 0.98q2
490:
1.189 brouard 491: Revision 1.188 2015/04/30 08:27:53 brouard
492: *** empty log message ***
493:
1.188 brouard 494: Revision 1.187 2015/04/29 09:11:15 brouard
495: *** empty log message ***
496:
1.187 brouard 497: Revision 1.186 2015/04/23 12:01:52 brouard
498: Summary: V1*age is working now, version 0.98q1
499:
500: Some codes had been disabled in order to simplify and Vn*age was
501: working in the optimization phase, ie, giving correct MLE parameters,
502: but, as usual, outputs were not correct and program core dumped.
503:
1.186 brouard 504: Revision 1.185 2015/03/11 13:26:42 brouard
505: Summary: Inclusion of compile and links command line for Intel Compiler
506:
1.185 brouard 507: Revision 1.184 2015/03/11 11:52:39 brouard
508: Summary: Back from Windows 8. Intel Compiler
509:
1.184 brouard 510: Revision 1.183 2015/03/10 20:34:32 brouard
511: Summary: 0.98q0, trying with directest, mnbrak fixed
512:
513: We use directest instead of original Powell test; probably no
514: incidence on the results, but better justifications;
515: We fixed Numerical Recipes mnbrak routine which was wrong and gave
516: wrong results.
517:
1.183 brouard 518: Revision 1.182 2015/02/12 08:19:57 brouard
519: Summary: Trying to keep directest which seems simpler and more general
520: Author: Nicolas Brouard
521:
1.182 brouard 522: Revision 1.181 2015/02/11 23:22:24 brouard
523: Summary: Comments on Powell added
524:
525: Author:
526:
1.181 brouard 527: Revision 1.180 2015/02/11 17:33:45 brouard
528: Summary: Finishing move from main to function (hpijx and prevalence_limit)
529:
1.180 brouard 530: Revision 1.179 2015/01/04 09:57:06 brouard
531: Summary: back to OS/X
532:
1.179 brouard 533: Revision 1.178 2015/01/04 09:35:48 brouard
534: *** empty log message ***
535:
1.178 brouard 536: Revision 1.177 2015/01/03 18:40:56 brouard
537: Summary: Still testing ilc32 on OSX
538:
1.177 brouard 539: Revision 1.176 2015/01/03 16:45:04 brouard
540: *** empty log message ***
541:
1.176 brouard 542: Revision 1.175 2015/01/03 16:33:42 brouard
543: *** empty log message ***
544:
1.175 brouard 545: Revision 1.174 2015/01/03 16:15:49 brouard
546: Summary: Still in cross-compilation
547:
1.174 brouard 548: Revision 1.173 2015/01/03 12:06:26 brouard
549: Summary: trying to detect cross-compilation
550:
1.173 brouard 551: Revision 1.172 2014/12/27 12:07:47 brouard
552: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
553:
1.172 brouard 554: Revision 1.171 2014/12/23 13:26:59 brouard
555: Summary: Back from Visual C
556:
557: Still problem with utsname.h on Windows
558:
1.171 brouard 559: Revision 1.170 2014/12/23 11:17:12 brouard
560: Summary: Cleaning some \%% back to %%
561:
562: The escape was mandatory for a specific compiler (which one?), but too many warnings.
563:
1.170 brouard 564: Revision 1.169 2014/12/22 23:08:31 brouard
565: Summary: 0.98p
566:
567: Outputs some informations on compiler used, OS etc. Testing on different platforms.
568:
1.169 brouard 569: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 570: Summary: update
1.169 brouard 571:
1.168 brouard 572: Revision 1.167 2014/12/22 13:50:56 brouard
573: Summary: Testing uname and compiler version and if compiled 32 or 64
574:
575: Testing on Linux 64
576:
1.167 brouard 577: Revision 1.166 2014/12/22 11:40:47 brouard
578: *** empty log message ***
579:
1.166 brouard 580: Revision 1.165 2014/12/16 11:20:36 brouard
581: Summary: After compiling on Visual C
582:
583: * imach.c (Module): Merging 1.61 to 1.162
584:
1.165 brouard 585: Revision 1.164 2014/12/16 10:52:11 brouard
586: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
587:
588: * imach.c (Module): Merging 1.61 to 1.162
589:
1.164 brouard 590: Revision 1.163 2014/12/16 10:30:11 brouard
591: * imach.c (Module): Merging 1.61 to 1.162
592:
1.163 brouard 593: Revision 1.162 2014/09/25 11:43:39 brouard
594: Summary: temporary backup 0.99!
595:
1.162 brouard 596: Revision 1.1 2014/09/16 11:06:58 brouard
597: Summary: With some code (wrong) for nlopt
598:
599: Author:
600:
601: Revision 1.161 2014/09/15 20:41:41 brouard
602: Summary: Problem with macro SQR on Intel compiler
603:
1.161 brouard 604: Revision 1.160 2014/09/02 09:24:05 brouard
605: *** empty log message ***
606:
1.160 brouard 607: Revision 1.159 2014/09/01 10:34:10 brouard
608: Summary: WIN32
609: Author: Brouard
610:
1.159 brouard 611: Revision 1.158 2014/08/27 17:11:51 brouard
612: *** empty log message ***
613:
1.158 brouard 614: Revision 1.157 2014/08/27 16:26:55 brouard
615: Summary: Preparing windows Visual studio version
616: Author: Brouard
617:
618: In order to compile on Visual studio, time.h is now correct and time_t
619: and tm struct should be used. difftime should be used but sometimes I
620: just make the differences in raw time format (time(&now).
621: Trying to suppress #ifdef LINUX
622: Add xdg-open for __linux in order to open default browser.
623:
1.157 brouard 624: Revision 1.156 2014/08/25 20:10:10 brouard
625: *** empty log message ***
626:
1.156 brouard 627: Revision 1.155 2014/08/25 18:32:34 brouard
628: Summary: New compile, minor changes
629: Author: Brouard
630:
1.155 brouard 631: Revision 1.154 2014/06/20 17:32:08 brouard
632: Summary: Outputs now all graphs of convergence to period prevalence
633:
1.154 brouard 634: Revision 1.153 2014/06/20 16:45:46 brouard
635: Summary: If 3 live state, convergence to period prevalence on same graph
636: Author: Brouard
637:
1.153 brouard 638: Revision 1.152 2014/06/18 17:54:09 brouard
639: Summary: open browser, use gnuplot on same dir than imach if not found in the path
640:
1.152 brouard 641: Revision 1.151 2014/06/18 16:43:30 brouard
642: *** empty log message ***
643:
1.151 brouard 644: Revision 1.150 2014/06/18 16:42:35 brouard
645: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
646: Author: brouard
647:
1.150 brouard 648: Revision 1.149 2014/06/18 15:51:14 brouard
649: Summary: Some fixes in parameter files errors
650: Author: Nicolas Brouard
651:
1.149 brouard 652: Revision 1.148 2014/06/17 17:38:48 brouard
653: Summary: Nothing new
654: Author: Brouard
655:
656: Just a new packaging for OS/X version 0.98nS
657:
1.148 brouard 658: Revision 1.147 2014/06/16 10:33:11 brouard
659: *** empty log message ***
660:
1.147 brouard 661: Revision 1.146 2014/06/16 10:20:28 brouard
662: Summary: Merge
663: Author: Brouard
664:
665: Merge, before building revised version.
666:
1.146 brouard 667: Revision 1.145 2014/06/10 21:23:15 brouard
668: Summary: Debugging with valgrind
669: Author: Nicolas Brouard
670:
671: Lot of changes in order to output the results with some covariates
672: After the Edimburgh REVES conference 2014, it seems mandatory to
673: improve the code.
674: No more memory valgrind error but a lot has to be done in order to
675: continue the work of splitting the code into subroutines.
676: Also, decodemodel has been improved. Tricode is still not
677: optimal. nbcode should be improved. Documentation has been added in
678: the source code.
679:
1.144 brouard 680: Revision 1.143 2014/01/26 09:45:38 brouard
681: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
682:
683: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
684: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
685:
1.143 brouard 686: Revision 1.142 2014/01/26 03:57:36 brouard
687: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
688:
689: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
690:
1.142 brouard 691: Revision 1.141 2014/01/26 02:42:01 brouard
692: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
693:
1.141 brouard 694: Revision 1.140 2011/09/02 10:37:54 brouard
695: Summary: times.h is ok with mingw32 now.
696:
1.140 brouard 697: Revision 1.139 2010/06/14 07:50:17 brouard
698: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
699: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
700:
1.139 brouard 701: Revision 1.138 2010/04/30 18:19:40 brouard
702: *** empty log message ***
703:
1.138 brouard 704: Revision 1.137 2010/04/29 18:11:38 brouard
705: (Module): Checking covariates for more complex models
706: than V1+V2. A lot of change to be done. Unstable.
707:
1.137 brouard 708: Revision 1.136 2010/04/26 20:30:53 brouard
709: (Module): merging some libgsl code. Fixing computation
710: of likelione (using inter/intrapolation if mle = 0) in order to
711: get same likelihood as if mle=1.
712: Some cleaning of code and comments added.
713:
1.136 brouard 714: Revision 1.135 2009/10/29 15:33:14 brouard
715: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
716:
1.135 brouard 717: Revision 1.134 2009/10/29 13:18:53 brouard
718: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
719:
1.134 brouard 720: Revision 1.133 2009/07/06 10:21:25 brouard
721: just nforces
722:
1.133 brouard 723: Revision 1.132 2009/07/06 08:22:05 brouard
724: Many tings
725:
1.132 brouard 726: Revision 1.131 2009/06/20 16:22:47 brouard
727: Some dimensions resccaled
728:
1.131 brouard 729: Revision 1.130 2009/05/26 06:44:34 brouard
730: (Module): Max Covariate is now set to 20 instead of 8. A
731: lot of cleaning with variables initialized to 0. Trying to make
732: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
733:
1.130 brouard 734: Revision 1.129 2007/08/31 13:49:27 lievre
735: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
736:
1.129 lievre 737: Revision 1.128 2006/06/30 13:02:05 brouard
738: (Module): Clarifications on computing e.j
739:
1.128 brouard 740: Revision 1.127 2006/04/28 18:11:50 brouard
741: (Module): Yes the sum of survivors was wrong since
742: imach-114 because nhstepm was no more computed in the age
743: loop. Now we define nhstepma in the age loop.
744: (Module): In order to speed up (in case of numerous covariates) we
745: compute health expectancies (without variances) in a first step
746: and then all the health expectancies with variances or standard
747: deviation (needs data from the Hessian matrices) which slows the
748: computation.
749: In the future we should be able to stop the program is only health
750: expectancies and graph are needed without standard deviations.
751:
1.127 brouard 752: Revision 1.126 2006/04/28 17:23:28 brouard
753: (Module): Yes the sum of survivors was wrong since
754: imach-114 because nhstepm was no more computed in the age
755: loop. Now we define nhstepma in the age loop.
756: Version 0.98h
757:
1.126 brouard 758: Revision 1.125 2006/04/04 15:20:31 lievre
759: Errors in calculation of health expectancies. Age was not initialized.
760: Forecasting file added.
761:
762: Revision 1.124 2006/03/22 17:13:53 lievre
763: Parameters are printed with %lf instead of %f (more numbers after the comma).
764: The log-likelihood is printed in the log file
765:
766: Revision 1.123 2006/03/20 10:52:43 brouard
767: * imach.c (Module): <title> changed, corresponds to .htm file
768: name. <head> headers where missing.
769:
770: * imach.c (Module): Weights can have a decimal point as for
771: English (a comma might work with a correct LC_NUMERIC environment,
772: otherwise the weight is truncated).
773: Modification of warning when the covariates values are not 0 or
774: 1.
775: Version 0.98g
776:
777: Revision 1.122 2006/03/20 09:45:41 brouard
778: (Module): Weights can have a decimal point as for
779: English (a comma might work with a correct LC_NUMERIC environment,
780: otherwise the weight is truncated).
781: Modification of warning when the covariates values are not 0 or
782: 1.
783: Version 0.98g
784:
785: Revision 1.121 2006/03/16 17:45:01 lievre
786: * imach.c (Module): Comments concerning covariates added
787:
788: * imach.c (Module): refinements in the computation of lli if
789: status=-2 in order to have more reliable computation if stepm is
790: not 1 month. Version 0.98f
791:
792: Revision 1.120 2006/03/16 15:10:38 lievre
793: (Module): refinements in the computation of lli if
794: status=-2 in order to have more reliable computation if stepm is
795: not 1 month. Version 0.98f
796:
797: Revision 1.119 2006/03/15 17:42:26 brouard
798: (Module): Bug if status = -2, the loglikelihood was
799: computed as likelihood omitting the logarithm. Version O.98e
800:
801: Revision 1.118 2006/03/14 18:20:07 brouard
802: (Module): varevsij Comments added explaining the second
803: table of variances if popbased=1 .
804: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
805: (Module): Function pstamp added
806: (Module): Version 0.98d
807:
808: Revision 1.117 2006/03/14 17:16:22 brouard
809: (Module): varevsij Comments added explaining the second
810: table of variances if popbased=1 .
811: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
812: (Module): Function pstamp added
813: (Module): Version 0.98d
814:
815: Revision 1.116 2006/03/06 10:29:27 brouard
816: (Module): Variance-covariance wrong links and
817: varian-covariance of ej. is needed (Saito).
818:
819: Revision 1.115 2006/02/27 12:17:45 brouard
820: (Module): One freematrix added in mlikeli! 0.98c
821:
822: Revision 1.114 2006/02/26 12:57:58 brouard
823: (Module): Some improvements in processing parameter
824: filename with strsep.
825:
826: Revision 1.113 2006/02/24 14:20:24 brouard
827: (Module): Memory leaks checks with valgrind and:
828: datafile was not closed, some imatrix were not freed and on matrix
829: allocation too.
830:
831: Revision 1.112 2006/01/30 09:55:26 brouard
832: (Module): Back to gnuplot.exe instead of wgnuplot.exe
833:
834: Revision 1.111 2006/01/25 20:38:18 brouard
835: (Module): Lots of cleaning and bugs added (Gompertz)
836: (Module): Comments can be added in data file. Missing date values
837: can be a simple dot '.'.
838:
839: Revision 1.110 2006/01/25 00:51:50 brouard
840: (Module): Lots of cleaning and bugs added (Gompertz)
841:
842: Revision 1.109 2006/01/24 19:37:15 brouard
843: (Module): Comments (lines starting with a #) are allowed in data.
844:
845: Revision 1.108 2006/01/19 18:05:42 lievre
846: Gnuplot problem appeared...
847: To be fixed
848:
849: Revision 1.107 2006/01/19 16:20:37 brouard
850: Test existence of gnuplot in imach path
851:
852: Revision 1.106 2006/01/19 13:24:36 brouard
853: Some cleaning and links added in html output
854:
855: Revision 1.105 2006/01/05 20:23:19 lievre
856: *** empty log message ***
857:
858: Revision 1.104 2005/09/30 16:11:43 lievre
859: (Module): sump fixed, loop imx fixed, and simplifications.
860: (Module): If the status is missing at the last wave but we know
861: that the person is alive, then we can code his/her status as -2
862: (instead of missing=-1 in earlier versions) and his/her
863: contributions to the likelihood is 1 - Prob of dying from last
864: health status (= 1-p13= p11+p12 in the easiest case of somebody in
865: the healthy state at last known wave). Version is 0.98
866:
867: Revision 1.103 2005/09/30 15:54:49 lievre
868: (Module): sump fixed, loop imx fixed, and simplifications.
869:
870: Revision 1.102 2004/09/15 17:31:30 brouard
871: Add the possibility to read data file including tab characters.
872:
873: Revision 1.101 2004/09/15 10:38:38 brouard
874: Fix on curr_time
875:
876: Revision 1.100 2004/07/12 18:29:06 brouard
877: Add version for Mac OS X. Just define UNIX in Makefile
878:
879: Revision 1.99 2004/06/05 08:57:40 brouard
880: *** empty log message ***
881:
882: Revision 1.98 2004/05/16 15:05:56 brouard
883: New version 0.97 . First attempt to estimate force of mortality
884: directly from the data i.e. without the need of knowing the health
885: state at each age, but using a Gompertz model: log u =a + b*age .
886: This is the basic analysis of mortality and should be done before any
887: other analysis, in order to test if the mortality estimated from the
888: cross-longitudinal survey is different from the mortality estimated
889: from other sources like vital statistic data.
890:
891: The same imach parameter file can be used but the option for mle should be -3.
892:
1.324 brouard 893: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 894: former routines in order to include the new code within the former code.
895:
896: The output is very simple: only an estimate of the intercept and of
897: the slope with 95% confident intervals.
898:
899: Current limitations:
900: A) Even if you enter covariates, i.e. with the
901: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
902: B) There is no computation of Life Expectancy nor Life Table.
903:
904: Revision 1.97 2004/02/20 13:25:42 lievre
905: Version 0.96d. Population forecasting command line is (temporarily)
906: suppressed.
907:
908: Revision 1.96 2003/07/15 15:38:55 brouard
909: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
910: rewritten within the same printf. Workaround: many printfs.
911:
912: Revision 1.95 2003/07/08 07:54:34 brouard
913: * imach.c (Repository):
914: (Repository): Using imachwizard code to output a more meaningful covariance
915: matrix (cov(a12,c31) instead of numbers.
916:
917: Revision 1.94 2003/06/27 13:00:02 brouard
918: Just cleaning
919:
920: Revision 1.93 2003/06/25 16:33:55 brouard
921: (Module): On windows (cygwin) function asctime_r doesn't
922: exist so I changed back to asctime which exists.
923: (Module): Version 0.96b
924:
925: Revision 1.92 2003/06/25 16:30:45 brouard
926: (Module): On windows (cygwin) function asctime_r doesn't
927: exist so I changed back to asctime which exists.
928:
929: Revision 1.91 2003/06/25 15:30:29 brouard
930: * imach.c (Repository): Duplicated warning errors corrected.
931: (Repository): Elapsed time after each iteration is now output. It
932: helps to forecast when convergence will be reached. Elapsed time
933: is stamped in powell. We created a new html file for the graphs
934: concerning matrix of covariance. It has extension -cov.htm.
935:
936: Revision 1.90 2003/06/24 12:34:15 brouard
937: (Module): Some bugs corrected for windows. Also, when
938: mle=-1 a template is output in file "or"mypar.txt with the design
939: of the covariance matrix to be input.
940:
941: Revision 1.89 2003/06/24 12:30:52 brouard
942: (Module): Some bugs corrected for windows. Also, when
943: mle=-1 a template is output in file "or"mypar.txt with the design
944: of the covariance matrix to be input.
945:
946: Revision 1.88 2003/06/23 17:54:56 brouard
947: * 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.
948:
949: Revision 1.87 2003/06/18 12:26:01 brouard
950: Version 0.96
951:
952: Revision 1.86 2003/06/17 20:04:08 brouard
953: (Module): Change position of html and gnuplot routines and added
954: routine fileappend.
955:
956: Revision 1.85 2003/06/17 13:12:43 brouard
957: * imach.c (Repository): Check when date of death was earlier that
958: current date of interview. It may happen when the death was just
959: prior to the death. In this case, dh was negative and likelihood
960: was wrong (infinity). We still send an "Error" but patch by
961: assuming that the date of death was just one stepm after the
962: interview.
963: (Repository): Because some people have very long ID (first column)
964: we changed int to long in num[] and we added a new lvector for
965: memory allocation. But we also truncated to 8 characters (left
966: truncation)
967: (Repository): No more line truncation errors.
968:
969: Revision 1.84 2003/06/13 21:44:43 brouard
970: * imach.c (Repository): Replace "freqsummary" at a correct
971: place. It differs from routine "prevalence" which may be called
972: many times. Probs is memory consuming and must be used with
973: parcimony.
974: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
975:
976: Revision 1.83 2003/06/10 13:39:11 lievre
977: *** empty log message ***
978:
979: Revision 1.82 2003/06/05 15:57:20 brouard
980: Add log in imach.c and fullversion number is now printed.
981:
982: */
983: /*
984: Interpolated Markov Chain
985:
986: Short summary of the programme:
987:
1.227 brouard 988: This program computes Healthy Life Expectancies or State-specific
989: (if states aren't health statuses) Expectancies from
990: cross-longitudinal data. Cross-longitudinal data consist in:
991:
992: -1- a first survey ("cross") where individuals from different ages
993: are interviewed on their health status or degree of disability (in
994: the case of a health survey which is our main interest)
995:
996: -2- at least a second wave of interviews ("longitudinal") which
997: measure each change (if any) in individual health status. Health
998: expectancies are computed from the time spent in each health state
999: according to a model. More health states you consider, more time is
1000: necessary to reach the Maximum Likelihood of the parameters involved
1001: in the model. The simplest model is the multinomial logistic model
1002: where pij is the probability to be observed in state j at the second
1003: wave conditional to be observed in state i at the first
1004: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1005: etc , where 'age' is age and 'sex' is a covariate. If you want to
1006: have a more complex model than "constant and age", you should modify
1007: the program where the markup *Covariates have to be included here
1008: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1009: convergence.
1010:
1011: The advantage of this computer programme, compared to a simple
1012: multinomial logistic model, is clear when the delay between waves is not
1013: identical for each individual. Also, if a individual missed an
1014: intermediate interview, the information is lost, but taken into
1015: account using an interpolation or extrapolation.
1016:
1017: hPijx is the probability to be observed in state i at age x+h
1018: conditional to the observed state i at age x. The delay 'h' can be
1019: split into an exact number (nh*stepm) of unobserved intermediate
1020: states. This elementary transition (by month, quarter,
1021: semester or year) is modelled as a multinomial logistic. The hPx
1022: matrix is simply the matrix product of nh*stepm elementary matrices
1023: and the contribution of each individual to the likelihood is simply
1024: hPijx.
1025:
1026: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1027: of the life expectancies. It also computes the period (stable) prevalence.
1028:
1029: Back prevalence and projections:
1.227 brouard 1030:
1031: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1032: double agemaxpar, double ftolpl, int *ncvyearp, double
1033: dateprev1,double dateprev2, int firstpass, int lastpass, int
1034: mobilavproj)
1035:
1036: Computes the back prevalence limit for any combination of
1037: covariate values k at any age between ageminpar and agemaxpar and
1038: returns it in **bprlim. In the loops,
1039:
1040: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1041: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1042:
1043: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1044: Computes for any combination of covariates k and any age between bage and fage
1045: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1046: oldm=oldms;savm=savms;
1.227 brouard 1047:
1.267 brouard 1048: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1049: Computes the transition matrix starting at age 'age' over
1050: 'nhstepm*hstepm*stepm' months (i.e. until
1051: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1052: nhstepm*hstepm matrices.
1053:
1054: Returns p3mat[i][j][h] after calling
1055: p3mat[i][j][h]=matprod2(newm,
1056: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1057: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1058: oldm);
1.226 brouard 1059:
1060: Important routines
1061:
1062: - func (or funcone), computes logit (pij) distinguishing
1063: o fixed variables (single or product dummies or quantitative);
1064: o varying variables by:
1065: (1) wave (single, product dummies, quantitative),
1066: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1067: % fixed dummy (treated) or quantitative (not done because time-consuming);
1068: % varying dummy (not done) or quantitative (not done);
1069: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1070: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1071: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1072: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1073: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1074:
1.226 brouard 1075:
1076:
1.324 brouard 1077: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1078: Institut national d'études démographiques, Paris.
1.126 brouard 1079: This software have been partly granted by Euro-REVES, a concerted action
1080: from the European Union.
1081: It is copyrighted identically to a GNU software product, ie programme and
1082: software can be distributed freely for non commercial use. Latest version
1083: can be accessed at http://euroreves.ined.fr/imach .
1084:
1085: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1086: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1087:
1088: **********************************************************************/
1089: /*
1090: main
1091: read parameterfile
1092: read datafile
1093: concatwav
1094: freqsummary
1095: if (mle >= 1)
1096: mlikeli
1097: print results files
1098: if mle==1
1099: computes hessian
1100: read end of parameter file: agemin, agemax, bage, fage, estepm
1101: begin-prev-date,...
1102: open gnuplot file
1103: open html file
1.145 brouard 1104: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1105: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1106: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1107: freexexit2 possible for memory heap.
1108:
1109: h Pij x | pij_nom ficrestpij
1110: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1111: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1112: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1113:
1114: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1115: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1116: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1117: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1118: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1119:
1.126 brouard 1120: forecasting if prevfcast==1 prevforecast call prevalence()
1121: health expectancies
1122: Variance-covariance of DFLE
1123: prevalence()
1124: movingaverage()
1125: varevsij()
1126: if popbased==1 varevsij(,popbased)
1127: total life expectancies
1128: Variance of period (stable) prevalence
1129: end
1130: */
1131:
1.187 brouard 1132: /* #define DEBUG */
1133: /* #define DEBUGBRENT */
1.203 brouard 1134: /* #define DEBUGLINMIN */
1135: /* #define DEBUGHESS */
1136: #define DEBUGHESSIJ
1.224 brouard 1137: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1138: #define POWELL /* Instead of NLOPT */
1.224 brouard 1139: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1140: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1141: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1142: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1143:
1144: #include <math.h>
1145: #include <stdio.h>
1146: #include <stdlib.h>
1147: #include <string.h>
1.226 brouard 1148: #include <ctype.h>
1.159 brouard 1149:
1150: #ifdef _WIN32
1151: #include <io.h>
1.172 brouard 1152: #include <windows.h>
1153: #include <tchar.h>
1.159 brouard 1154: #else
1.126 brouard 1155: #include <unistd.h>
1.159 brouard 1156: #endif
1.126 brouard 1157:
1158: #include <limits.h>
1159: #include <sys/types.h>
1.171 brouard 1160:
1161: #if defined(__GNUC__)
1162: #include <sys/utsname.h> /* Doesn't work on Windows */
1163: #endif
1164:
1.126 brouard 1165: #include <sys/stat.h>
1166: #include <errno.h>
1.159 brouard 1167: /* extern int errno; */
1.126 brouard 1168:
1.157 brouard 1169: /* #ifdef LINUX */
1170: /* #include <time.h> */
1171: /* #include "timeval.h" */
1172: /* #else */
1173: /* #include <sys/time.h> */
1174: /* #endif */
1175:
1.126 brouard 1176: #include <time.h>
1177:
1.136 brouard 1178: #ifdef GSL
1179: #include <gsl/gsl_errno.h>
1180: #include <gsl/gsl_multimin.h>
1181: #endif
1182:
1.167 brouard 1183:
1.162 brouard 1184: #ifdef NLOPT
1185: #include <nlopt.h>
1186: typedef struct {
1187: double (* function)(double [] );
1188: } myfunc_data ;
1189: #endif
1190:
1.126 brouard 1191: /* #include <libintl.h> */
1192: /* #define _(String) gettext (String) */
1193:
1.251 brouard 1194: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1195:
1196: #define GNUPLOTPROGRAM "gnuplot"
1197: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1198: #define FILENAMELENGTH 256
1.126 brouard 1199:
1200: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1201: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1202:
1.144 brouard 1203: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1204: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1205:
1206: #define NINTERVMAX 8
1.144 brouard 1207: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1208: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1209: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1210: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1211: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1212: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1213: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1214: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1215: /* #define AGESUP 130 */
1.288 brouard 1216: /* #define AGESUP 150 */
1217: #define AGESUP 200
1.268 brouard 1218: #define AGEINF 0
1.218 brouard 1219: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1220: #define AGEBASE 40
1.194 brouard 1221: #define AGEOVERFLOW 1.e20
1.164 brouard 1222: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1223: #ifdef _WIN32
1224: #define DIRSEPARATOR '\\'
1225: #define CHARSEPARATOR "\\"
1226: #define ODIRSEPARATOR '/'
1227: #else
1.126 brouard 1228: #define DIRSEPARATOR '/'
1229: #define CHARSEPARATOR "/"
1230: #define ODIRSEPARATOR '\\'
1231: #endif
1232:
1.330 ! brouard 1233: /* $Id: imach.c,v 1.329 2022/08/03 17:29:54 brouard Exp $ */
1.126 brouard 1234: /* $State: Exp $ */
1.196 brouard 1235: #include "version.h"
1236: char version[]=__IMACH_VERSION__;
1.323 brouard 1237: 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";
1.330 ! brouard 1238: char fullversion[]="$Revision: 1.329 $ $Date: 2022/08/03 17:29:54 $";
1.126 brouard 1239: char strstart[80];
1240: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1241: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1242: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 ! brouard 1243: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
! 1244: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
! 1245: int cptcovn=0; /**< cptcovn decodemodel: number of covariates k of the models excluding age*products =6 and age*age */
! 1246: int cptcovt=0; /**< cptcovt: total number of covariates of the model (2) nbocc(+)+1 = 8 excepting constant and age and age*age */
! 1247: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1.225 brouard 1248: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1249: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1250: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.330 ! brouard 1251: int cptcoveff=0; /* Total number of covariates to vary for printing results (2**cptcoveff combinations of dummies)(computed in tricode as cptcov) */
1.233 brouard 1252: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1253: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1254: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1255: int nsd=0; /**< Total number of single dummy variables (output) */
1256: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1257: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1258: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1259: int ntveff=0; /**< ntveff number of effective time varying variables */
1260: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1261: int cptcov=0; /* Working variable */
1.290 brouard 1262: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1263: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1264: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1265: int nlstate=2; /* Number of live states */
1266: int ndeath=1; /* Number of dead states */
1.130 brouard 1267: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1268: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1269: int popbased=0;
1270:
1271: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1272: int maxwav=0; /* Maxim number of waves */
1273: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1274: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1275: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1276: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1277: int mle=1, weightopt=0;
1.126 brouard 1278: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1279: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1280: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1281: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1282: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1283: int selected(int kvar); /* Is covariate kvar selected for printing results */
1284:
1.130 brouard 1285: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1286: double **matprod2(); /* test */
1.126 brouard 1287: double **oldm, **newm, **savm; /* Working pointers to matrices */
1288: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1289: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1290:
1.136 brouard 1291: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1292: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1293: FILE *ficlog, *ficrespow;
1.130 brouard 1294: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1295: double fretone; /* Only one call to likelihood */
1.130 brouard 1296: long ipmx=0; /* Number of contributions */
1.126 brouard 1297: double sw; /* Sum of weights */
1298: char filerespow[FILENAMELENGTH];
1299: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1300: FILE *ficresilk;
1301: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1302: FILE *ficresprobmorprev;
1303: FILE *fichtm, *fichtmcov; /* Html File */
1304: FILE *ficreseij;
1305: char filerese[FILENAMELENGTH];
1306: FILE *ficresstdeij;
1307: char fileresstde[FILENAMELENGTH];
1308: FILE *ficrescveij;
1309: char filerescve[FILENAMELENGTH];
1310: FILE *ficresvij;
1311: char fileresv[FILENAMELENGTH];
1.269 brouard 1312:
1.126 brouard 1313: char title[MAXLINE];
1.234 brouard 1314: char model[MAXLINE]; /**< The model line */
1.217 brouard 1315: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1316: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1317: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1318: char command[FILENAMELENGTH];
1319: int outcmd=0;
1320:
1.217 brouard 1321: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1322: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1323: char filelog[FILENAMELENGTH]; /* Log file */
1324: char filerest[FILENAMELENGTH];
1325: char fileregp[FILENAMELENGTH];
1326: char popfile[FILENAMELENGTH];
1327:
1328: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1329:
1.157 brouard 1330: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1331: /* struct timezone tzp; */
1332: /* extern int gettimeofday(); */
1333: struct tm tml, *gmtime(), *localtime();
1334:
1335: extern time_t time();
1336:
1337: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1338: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1339: struct tm tm;
1340:
1.126 brouard 1341: char strcurr[80], strfor[80];
1342:
1343: char *endptr;
1344: long lval;
1345: double dval;
1346:
1347: #define NR_END 1
1348: #define FREE_ARG char*
1349: #define FTOL 1.0e-10
1350:
1351: #define NRANSI
1.240 brouard 1352: #define ITMAX 200
1353: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1354:
1355: #define TOL 2.0e-4
1356:
1357: #define CGOLD 0.3819660
1358: #define ZEPS 1.0e-10
1359: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1360:
1361: #define GOLD 1.618034
1362: #define GLIMIT 100.0
1363: #define TINY 1.0e-20
1364:
1365: static double maxarg1,maxarg2;
1366: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1367: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1368:
1369: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1370: #define rint(a) floor(a+0.5)
1.166 brouard 1371: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1372: #define mytinydouble 1.0e-16
1.166 brouard 1373: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1374: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1375: /* static double dsqrarg; */
1376: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1377: static double sqrarg;
1378: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1379: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1380: int agegomp= AGEGOMP;
1381:
1382: int imx;
1383: int stepm=1;
1384: /* Stepm, step in month: minimum step interpolation*/
1385:
1386: int estepm;
1387: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1388:
1389: int m,nb;
1390: long *num;
1.197 brouard 1391: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1392: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1393: covariate for which somebody answered excluding
1394: undefined. Usually 2: 0 and 1. */
1395: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1396: covariate for which somebody answered including
1397: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1398: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1399: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1400: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1401: double *ageexmed,*agecens;
1402: double dateintmean=0;
1.296 brouard 1403: double anprojd, mprojd, jprojd; /* For eventual projections */
1404: double anprojf, mprojf, jprojf;
1.126 brouard 1405:
1.296 brouard 1406: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1407: double anbackf, mbackf, jbackf;
1408: double jintmean,mintmean,aintmean;
1.126 brouard 1409: double *weight;
1410: int **s; /* Status */
1.141 brouard 1411: double *agedc;
1.145 brouard 1412: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1413: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1414: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1415: double **coqvar; /* Fixed quantitative covariate nqv */
1416: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1417: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1418: double idx;
1419: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1420: /* Some documentation */
1421: /* Design original data
1422: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1423: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1424: * ntv=3 nqtv=1
1.330 ! brouard 1425: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1426: * For time varying covariate, quanti or dummies
1427: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1428: * cotvar[wav][ntv+iv][i]= [3+(1 to nqtv)][i]=(V12) quanti
1429: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1430: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1431: * covar[k,i], value of kth fixed covariate dummy or quanti :
1432: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1433: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1434: * k= 1 2 3 4 5 6 7 8 9 10 11
1435: */
1436: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1437: /* 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
1438: # States 1=Coresidence, 2 Living alone, 3 Institution
1439: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1440: */
1.319 brouard 1441: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1442: /* k 1 2 3 4 5 6 7 8 9 */
1443: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1444: /* fixed or varying), 1 for age product, 2 for*/
1445: /* product */
1446: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1447: /*(single or product without age), 2 dummy*/
1448: /* with age product, 3 quant with age product*/
1449: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1450: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 ! brouard 1451: /*TnsdVar[Tvar] 1 2 3 */
1.319 brouard 1452: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1453: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1454: /* nsq 1 2 */ /* Counting single quantit tv */
1455: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1456: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1457: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1458: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1459: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1460: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 ! brouard 1461: /* Tvardk[4][1]=4;Tvardk[4][2]=3;Tvardk[7][1]=1;Tvardk[7][2]=2 */ /* Variables of a prod at position in the model equation*/
1.319 brouard 1462: /* 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 1463: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1464: /* Type */
1465: /* V 1 2 3 4 5 */
1466: /* F F V V V */
1467: /* D Q D D Q */
1468: /* */
1469: int *TvarsD;
1.330 ! brouard 1470: int *TnsdVar;
1.234 brouard 1471: int *TvarsDind;
1472: int *TvarsQ;
1473: int *TvarsQind;
1474:
1.318 brouard 1475: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1476: int nresult=0;
1.258 brouard 1477: int parameterline=0; /* # of the parameter (type) line */
1.318 brouard 1478: int TKresult[MAXRESULTLINESPONE];
1.330 ! brouard 1479: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
1.318 brouard 1480: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1481: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1.330 ! brouard 1482: int TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable or quanti value (output) */
1.318 brouard 1483: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For dummy variable , variable # (output) */
1484: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1485: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1486: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , variable # (output) */
1487:
1488: /* 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
1489: # States 1=Coresidence, 2 Living alone, 3 Institution
1490: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1491: */
1.234 brouard 1492: /* 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 1493: 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 */
1494: 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 */
1495: 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 */
1496: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1497: 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 */
1498: 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 1499: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1500: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1501: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1502: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1503: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1504: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1505: 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 */
1506: 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 */
1507:
1.230 brouard 1508: int *Tvarsel; /**< Selected covariates for output */
1509: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1510: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1511: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1512: 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 1513: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1514: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1515: int *Tage;
1.227 brouard 1516: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1517: 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 1518: 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*/
1519: 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 1520: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1521: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1522: int **Tvard;
1.330 ! brouard 1523: int **Tvardk;
1.227 brouard 1524: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1525: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1526: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1527: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1528: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1529: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1530: double *lsurv, *lpop, *tpop;
1531:
1.231 brouard 1532: #define FD 1; /* Fixed dummy covariate */
1533: #define FQ 2; /* Fixed quantitative covariate */
1534: #define FP 3; /* Fixed product covariate */
1535: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1536: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1537: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1538: #define VD 10; /* Varying dummy covariate */
1539: #define VQ 11; /* Varying quantitative covariate */
1540: #define VP 12; /* Varying product covariate */
1541: #define VPDD 13; /* Varying product dummy*dummy covariate */
1542: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1543: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1544: #define APFD 16; /* Age product * fixed dummy covariate */
1545: #define APFQ 17; /* Age product * fixed quantitative covariate */
1546: #define APVD 18; /* Age product * varying dummy covariate */
1547: #define APVQ 19; /* Age product * varying quantitative covariate */
1548:
1549: #define FTYPE 1; /* Fixed covariate */
1550: #define VTYPE 2; /* Varying covariate (loop in wave) */
1551: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1552:
1553: struct kmodel{
1554: int maintype; /* main type */
1555: int subtype; /* subtype */
1556: };
1557: struct kmodel modell[NCOVMAX];
1558:
1.143 brouard 1559: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1560: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1561:
1562: /**************** split *************************/
1563: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1564: {
1565: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1566: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1567: */
1568: char *ss; /* pointer */
1.186 brouard 1569: int l1=0, l2=0; /* length counters */
1.126 brouard 1570:
1571: l1 = strlen(path ); /* length of path */
1572: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1573: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1574: if ( ss == NULL ) { /* no directory, so determine current directory */
1575: strcpy( name, path ); /* we got the fullname name because no directory */
1576: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1577: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1578: /* get current working directory */
1579: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1580: #ifdef WIN32
1581: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1582: #else
1583: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1584: #endif
1.126 brouard 1585: return( GLOCK_ERROR_GETCWD );
1586: }
1587: /* got dirc from getcwd*/
1588: printf(" DIRC = %s \n",dirc);
1.205 brouard 1589: } else { /* strip directory from path */
1.126 brouard 1590: ss++; /* after this, the filename */
1591: l2 = strlen( ss ); /* length of filename */
1592: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1593: strcpy( name, ss ); /* save file name */
1594: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1595: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1596: printf(" DIRC2 = %s \n",dirc);
1597: }
1598: /* We add a separator at the end of dirc if not exists */
1599: l1 = strlen( dirc ); /* length of directory */
1600: if( dirc[l1-1] != DIRSEPARATOR ){
1601: dirc[l1] = DIRSEPARATOR;
1602: dirc[l1+1] = 0;
1603: printf(" DIRC3 = %s \n",dirc);
1604: }
1605: ss = strrchr( name, '.' ); /* find last / */
1606: if (ss >0){
1607: ss++;
1608: strcpy(ext,ss); /* save extension */
1609: l1= strlen( name);
1610: l2= strlen(ss)+1;
1611: strncpy( finame, name, l1-l2);
1612: finame[l1-l2]= 0;
1613: }
1614:
1615: return( 0 ); /* we're done */
1616: }
1617:
1618:
1619: /******************************************/
1620:
1621: void replace_back_to_slash(char *s, char*t)
1622: {
1623: int i;
1624: int lg=0;
1625: i=0;
1626: lg=strlen(t);
1627: for(i=0; i<= lg; i++) {
1628: (s[i] = t[i]);
1629: if (t[i]== '\\') s[i]='/';
1630: }
1631: }
1632:
1.132 brouard 1633: char *trimbb(char *out, char *in)
1.137 brouard 1634: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1635: char *s;
1636: s=out;
1637: while (*in != '\0'){
1.137 brouard 1638: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1639: in++;
1640: }
1641: *out++ = *in++;
1642: }
1643: *out='\0';
1644: return s;
1645: }
1646:
1.187 brouard 1647: /* char *substrchaine(char *out, char *in, char *chain) */
1648: /* { */
1649: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1650: /* char *s, *t; */
1651: /* t=in;s=out; */
1652: /* while ((*in != *chain) && (*in != '\0')){ */
1653: /* *out++ = *in++; */
1654: /* } */
1655:
1656: /* /\* *in matches *chain *\/ */
1657: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1658: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1659: /* } */
1660: /* in--; chain--; */
1661: /* while ( (*in != '\0')){ */
1662: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1663: /* *out++ = *in++; */
1664: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1665: /* } */
1666: /* *out='\0'; */
1667: /* out=s; */
1668: /* return out; */
1669: /* } */
1670: char *substrchaine(char *out, char *in, char *chain)
1671: {
1672: /* Substract chain 'chain' from 'in', return and output 'out' */
1673: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1674:
1675: char *strloc;
1676:
1677: strcpy (out, in);
1678: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1679: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1680: if(strloc != NULL){
1681: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1682: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1683: /* strcpy (strloc, strloc +strlen(chain));*/
1684: }
1685: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1686: return out;
1687: }
1688:
1689:
1.145 brouard 1690: char *cutl(char *blocc, char *alocc, char *in, char occ)
1691: {
1.187 brouard 1692: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1693: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1694: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1695: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1696: */
1.160 brouard 1697: char *s, *t;
1.145 brouard 1698: t=in;s=in;
1699: while ((*in != occ) && (*in != '\0')){
1700: *alocc++ = *in++;
1701: }
1702: if( *in == occ){
1703: *(alocc)='\0';
1704: s=++in;
1705: }
1706:
1707: if (s == t) {/* occ not found */
1708: *(alocc-(in-s))='\0';
1709: in=s;
1710: }
1711: while ( *in != '\0'){
1712: *blocc++ = *in++;
1713: }
1714:
1715: *blocc='\0';
1716: return t;
1717: }
1.137 brouard 1718: char *cutv(char *blocc, char *alocc, char *in, char occ)
1719: {
1.187 brouard 1720: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1721: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1722: gives blocc="abcdef2ghi" and alocc="j".
1723: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1724: */
1725: char *s, *t;
1726: t=in;s=in;
1727: while (*in != '\0'){
1728: while( *in == occ){
1729: *blocc++ = *in++;
1730: s=in;
1731: }
1732: *blocc++ = *in++;
1733: }
1734: if (s == t) /* occ not found */
1735: *(blocc-(in-s))='\0';
1736: else
1737: *(blocc-(in-s)-1)='\0';
1738: in=s;
1739: while ( *in != '\0'){
1740: *alocc++ = *in++;
1741: }
1742:
1743: *alocc='\0';
1744: return s;
1745: }
1746:
1.126 brouard 1747: int nbocc(char *s, char occ)
1748: {
1749: int i,j=0;
1750: int lg=20;
1751: i=0;
1752: lg=strlen(s);
1753: for(i=0; i<= lg; i++) {
1.234 brouard 1754: if (s[i] == occ ) j++;
1.126 brouard 1755: }
1756: return j;
1757: }
1758:
1.137 brouard 1759: /* void cutv(char *u,char *v, char*t, char occ) */
1760: /* { */
1761: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1762: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1763: /* gives u="abcdef2ghi" and v="j" *\/ */
1764: /* int i,lg,j,p=0; */
1765: /* i=0; */
1766: /* lg=strlen(t); */
1767: /* for(j=0; j<=lg-1; j++) { */
1768: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1769: /* } */
1.126 brouard 1770:
1.137 brouard 1771: /* for(j=0; j<p; j++) { */
1772: /* (u[j] = t[j]); */
1773: /* } */
1774: /* u[p]='\0'; */
1.126 brouard 1775:
1.137 brouard 1776: /* for(j=0; j<= lg; j++) { */
1777: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1778: /* } */
1779: /* } */
1.126 brouard 1780:
1.160 brouard 1781: #ifdef _WIN32
1782: char * strsep(char **pp, const char *delim)
1783: {
1784: char *p, *q;
1785:
1786: if ((p = *pp) == NULL)
1787: return 0;
1788: if ((q = strpbrk (p, delim)) != NULL)
1789: {
1790: *pp = q + 1;
1791: *q = '\0';
1792: }
1793: else
1794: *pp = 0;
1795: return p;
1796: }
1797: #endif
1798:
1.126 brouard 1799: /********************** nrerror ********************/
1800:
1801: void nrerror(char error_text[])
1802: {
1803: fprintf(stderr,"ERREUR ...\n");
1804: fprintf(stderr,"%s\n",error_text);
1805: exit(EXIT_FAILURE);
1806: }
1807: /*********************** vector *******************/
1808: double *vector(int nl, int nh)
1809: {
1810: double *v;
1811: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1812: if (!v) nrerror("allocation failure in vector");
1813: return v-nl+NR_END;
1814: }
1815:
1816: /************************ free vector ******************/
1817: void free_vector(double*v, int nl, int nh)
1818: {
1819: free((FREE_ARG)(v+nl-NR_END));
1820: }
1821:
1822: /************************ivector *******************************/
1823: int *ivector(long nl,long nh)
1824: {
1825: int *v;
1826: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1827: if (!v) nrerror("allocation failure in ivector");
1828: return v-nl+NR_END;
1829: }
1830:
1831: /******************free ivector **************************/
1832: void free_ivector(int *v, long nl, long nh)
1833: {
1834: free((FREE_ARG)(v+nl-NR_END));
1835: }
1836:
1837: /************************lvector *******************************/
1838: long *lvector(long nl,long nh)
1839: {
1840: long *v;
1841: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1842: if (!v) nrerror("allocation failure in ivector");
1843: return v-nl+NR_END;
1844: }
1845:
1846: /******************free lvector **************************/
1847: void free_lvector(long *v, long nl, long nh)
1848: {
1849: free((FREE_ARG)(v+nl-NR_END));
1850: }
1851:
1852: /******************* imatrix *******************************/
1853: int **imatrix(long nrl, long nrh, long ncl, long nch)
1854: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1855: {
1856: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1857: int **m;
1858:
1859: /* allocate pointers to rows */
1860: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1861: if (!m) nrerror("allocation failure 1 in matrix()");
1862: m += NR_END;
1863: m -= nrl;
1864:
1865:
1866: /* allocate rows and set pointers to them */
1867: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1868: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1869: m[nrl] += NR_END;
1870: m[nrl] -= ncl;
1871:
1872: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1873:
1874: /* return pointer to array of pointers to rows */
1875: return m;
1876: }
1877:
1878: /****************** free_imatrix *************************/
1879: void free_imatrix(m,nrl,nrh,ncl,nch)
1880: int **m;
1881: long nch,ncl,nrh,nrl;
1882: /* free an int matrix allocated by imatrix() */
1883: {
1884: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1885: free((FREE_ARG) (m+nrl-NR_END));
1886: }
1887:
1888: /******************* matrix *******************************/
1889: double **matrix(long nrl, long nrh, long ncl, long nch)
1890: {
1891: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1892: double **m;
1893:
1894: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1895: if (!m) nrerror("allocation failure 1 in matrix()");
1896: m += NR_END;
1897: m -= nrl;
1898:
1899: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1900: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1901: m[nrl] += NR_END;
1902: m[nrl] -= ncl;
1903:
1904: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1905: return m;
1.145 brouard 1906: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1907: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1908: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1909: */
1910: }
1911:
1912: /*************************free matrix ************************/
1913: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1914: {
1915: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1916: free((FREE_ARG)(m+nrl-NR_END));
1917: }
1918:
1919: /******************* ma3x *******************************/
1920: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1921: {
1922: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1923: double ***m;
1924:
1925: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1926: if (!m) nrerror("allocation failure 1 in matrix()");
1927: m += NR_END;
1928: m -= nrl;
1929:
1930: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1931: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1932: m[nrl] += NR_END;
1933: m[nrl] -= ncl;
1934:
1935: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1936:
1937: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1938: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1939: m[nrl][ncl] += NR_END;
1940: m[nrl][ncl] -= nll;
1941: for (j=ncl+1; j<=nch; j++)
1942: m[nrl][j]=m[nrl][j-1]+nlay;
1943:
1944: for (i=nrl+1; i<=nrh; i++) {
1945: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1946: for (j=ncl+1; j<=nch; j++)
1947: m[i][j]=m[i][j-1]+nlay;
1948: }
1949: return m;
1950: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1951: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1952: */
1953: }
1954:
1955: /*************************free ma3x ************************/
1956: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1957: {
1958: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1959: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1960: free((FREE_ARG)(m+nrl-NR_END));
1961: }
1962:
1963: /*************** function subdirf ***********/
1964: char *subdirf(char fileres[])
1965: {
1966: /* Caution optionfilefiname is hidden */
1967: strcpy(tmpout,optionfilefiname);
1968: strcat(tmpout,"/"); /* Add to the right */
1969: strcat(tmpout,fileres);
1970: return tmpout;
1971: }
1972:
1973: /*************** function subdirf2 ***********/
1974: char *subdirf2(char fileres[], char *preop)
1975: {
1.314 brouard 1976: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
1977: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 1978: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 1979: /* Caution optionfilefiname is hidden */
1980: strcpy(tmpout,optionfilefiname);
1981: strcat(tmpout,"/");
1982: strcat(tmpout,preop);
1983: strcat(tmpout,fileres);
1984: return tmpout;
1985: }
1986:
1987: /*************** function subdirf3 ***********/
1988: char *subdirf3(char fileres[], char *preop, char *preop2)
1989: {
1990:
1991: /* Caution optionfilefiname is hidden */
1992: strcpy(tmpout,optionfilefiname);
1993: strcat(tmpout,"/");
1994: strcat(tmpout,preop);
1995: strcat(tmpout,preop2);
1996: strcat(tmpout,fileres);
1997: return tmpout;
1998: }
1.213 brouard 1999:
2000: /*************** function subdirfext ***********/
2001: char *subdirfext(char fileres[], char *preop, char *postop)
2002: {
2003:
2004: strcpy(tmpout,preop);
2005: strcat(tmpout,fileres);
2006: strcat(tmpout,postop);
2007: return tmpout;
2008: }
1.126 brouard 2009:
1.213 brouard 2010: /*************** function subdirfext3 ***********/
2011: char *subdirfext3(char fileres[], char *preop, char *postop)
2012: {
2013:
2014: /* Caution optionfilefiname is hidden */
2015: strcpy(tmpout,optionfilefiname);
2016: strcat(tmpout,"/");
2017: strcat(tmpout,preop);
2018: strcat(tmpout,fileres);
2019: strcat(tmpout,postop);
2020: return tmpout;
2021: }
2022:
1.162 brouard 2023: char *asc_diff_time(long time_sec, char ascdiff[])
2024: {
2025: long sec_left, days, hours, minutes;
2026: days = (time_sec) / (60*60*24);
2027: sec_left = (time_sec) % (60*60*24);
2028: hours = (sec_left) / (60*60) ;
2029: sec_left = (sec_left) %(60*60);
2030: minutes = (sec_left) /60;
2031: sec_left = (sec_left) % (60);
2032: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2033: return ascdiff;
2034: }
2035:
1.126 brouard 2036: /***************** f1dim *************************/
2037: extern int ncom;
2038: extern double *pcom,*xicom;
2039: extern double (*nrfunc)(double []);
2040:
2041: double f1dim(double x)
2042: {
2043: int j;
2044: double f;
2045: double *xt;
2046:
2047: xt=vector(1,ncom);
2048: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2049: f=(*nrfunc)(xt);
2050: free_vector(xt,1,ncom);
2051: return f;
2052: }
2053:
2054: /*****************brent *************************/
2055: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2056: {
2057: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2058: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2059: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2060: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2061: * returned function value.
2062: */
1.126 brouard 2063: int iter;
2064: double a,b,d,etemp;
1.159 brouard 2065: double fu=0,fv,fw,fx;
1.164 brouard 2066: double ftemp=0.;
1.126 brouard 2067: double p,q,r,tol1,tol2,u,v,w,x,xm;
2068: double e=0.0;
2069:
2070: a=(ax < cx ? ax : cx);
2071: b=(ax > cx ? ax : cx);
2072: x=w=v=bx;
2073: fw=fv=fx=(*f)(x);
2074: for (iter=1;iter<=ITMAX;iter++) {
2075: xm=0.5*(a+b);
2076: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2077: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2078: printf(".");fflush(stdout);
2079: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2080: #ifdef DEBUGBRENT
1.126 brouard 2081: 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);
2082: 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);
2083: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2084: #endif
2085: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2086: *xmin=x;
2087: return fx;
2088: }
2089: ftemp=fu;
2090: if (fabs(e) > tol1) {
2091: r=(x-w)*(fx-fv);
2092: q=(x-v)*(fx-fw);
2093: p=(x-v)*q-(x-w)*r;
2094: q=2.0*(q-r);
2095: if (q > 0.0) p = -p;
2096: q=fabs(q);
2097: etemp=e;
2098: e=d;
2099: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2100: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2101: else {
1.224 brouard 2102: d=p/q;
2103: u=x+d;
2104: if (u-a < tol2 || b-u < tol2)
2105: d=SIGN(tol1,xm-x);
1.126 brouard 2106: }
2107: } else {
2108: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2109: }
2110: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2111: fu=(*f)(u);
2112: if (fu <= fx) {
2113: if (u >= x) a=x; else b=x;
2114: SHFT(v,w,x,u)
1.183 brouard 2115: SHFT(fv,fw,fx,fu)
2116: } else {
2117: if (u < x) a=u; else b=u;
2118: if (fu <= fw || w == x) {
1.224 brouard 2119: v=w;
2120: w=u;
2121: fv=fw;
2122: fw=fu;
1.183 brouard 2123: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2124: v=u;
2125: fv=fu;
1.183 brouard 2126: }
2127: }
1.126 brouard 2128: }
2129: nrerror("Too many iterations in brent");
2130: *xmin=x;
2131: return fx;
2132: }
2133:
2134: /****************** mnbrak ***********************/
2135:
2136: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2137: double (*func)(double))
1.183 brouard 2138: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2139: the downhill direction (defined by the function as evaluated at the initial points) and returns
2140: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2141: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2142: */
1.126 brouard 2143: double ulim,u,r,q, dum;
2144: double fu;
1.187 brouard 2145:
2146: double scale=10.;
2147: int iterscale=0;
2148:
2149: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2150: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2151:
2152:
2153: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2154: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2155: /* *bx = *ax - (*ax - *bx)/scale; */
2156: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2157: /* } */
2158:
1.126 brouard 2159: if (*fb > *fa) {
2160: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2161: SHFT(dum,*fb,*fa,dum)
2162: }
1.126 brouard 2163: *cx=(*bx)+GOLD*(*bx-*ax);
2164: *fc=(*func)(*cx);
1.183 brouard 2165: #ifdef DEBUG
1.224 brouard 2166: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2167: 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 2168: #endif
1.224 brouard 2169: 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 2170: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2171: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2172: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2173: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2174: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2175: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2176: fu=(*func)(u);
1.163 brouard 2177: #ifdef DEBUG
2178: /* f(x)=A(x-u)**2+f(u) */
2179: double A, fparabu;
2180: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2181: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2182: 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);
2183: 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 2184: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2185: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2186: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2187: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2188: #endif
1.184 brouard 2189: #ifdef MNBRAKORIGINAL
1.183 brouard 2190: #else
1.191 brouard 2191: /* if (fu > *fc) { */
2192: /* #ifdef DEBUG */
2193: /* printf("mnbrak4 fu > fc \n"); */
2194: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2195: /* #endif */
2196: /* /\* 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 *\\/ *\/ */
2197: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2198: /* dum=u; /\* Shifting c and u *\/ */
2199: /* u = *cx; */
2200: /* *cx = dum; */
2201: /* dum = fu; */
2202: /* fu = *fc; */
2203: /* *fc =dum; */
2204: /* } else { /\* end *\/ */
2205: /* #ifdef DEBUG */
2206: /* printf("mnbrak3 fu < fc \n"); */
2207: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2208: /* #endif */
2209: /* dum=u; /\* Shifting c and u *\/ */
2210: /* u = *cx; */
2211: /* *cx = dum; */
2212: /* dum = fu; */
2213: /* fu = *fc; */
2214: /* *fc =dum; */
2215: /* } */
1.224 brouard 2216: #ifdef DEBUGMNBRAK
2217: double A, fparabu;
2218: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2219: fparabu= *fa - A*(*ax-u)*(*ax-u);
2220: 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);
2221: 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 2222: #endif
1.191 brouard 2223: dum=u; /* Shifting c and u */
2224: u = *cx;
2225: *cx = dum;
2226: dum = fu;
2227: fu = *fc;
2228: *fc =dum;
1.183 brouard 2229: #endif
1.162 brouard 2230: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2231: #ifdef DEBUG
1.224 brouard 2232: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2233: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2234: #endif
1.126 brouard 2235: fu=(*func)(u);
2236: if (fu < *fc) {
1.183 brouard 2237: #ifdef DEBUG
1.224 brouard 2238: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2239: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2240: #endif
2241: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2242: SHFT(*fb,*fc,fu,(*func)(u))
2243: #ifdef DEBUG
2244: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2245: #endif
2246: }
1.162 brouard 2247: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2248: #ifdef DEBUG
1.224 brouard 2249: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2250: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2251: #endif
1.126 brouard 2252: u=ulim;
2253: fu=(*func)(u);
1.183 brouard 2254: } else { /* u could be left to b (if r > q parabola has a maximum) */
2255: #ifdef DEBUG
1.224 brouard 2256: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2257: 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 2258: #endif
1.126 brouard 2259: u=(*cx)+GOLD*(*cx-*bx);
2260: fu=(*func)(u);
1.224 brouard 2261: #ifdef DEBUG
2262: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2263: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2264: #endif
1.183 brouard 2265: } /* end tests */
1.126 brouard 2266: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2267: SHFT(*fa,*fb,*fc,fu)
2268: #ifdef DEBUG
1.224 brouard 2269: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2270: 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 2271: #endif
2272: } /* 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 2273: }
2274:
2275: /*************** linmin ************************/
1.162 brouard 2276: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2277: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2278: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2279: the value of func at the returned location p . This is actually all accomplished by calling the
2280: routines mnbrak and brent .*/
1.126 brouard 2281: int ncom;
2282: double *pcom,*xicom;
2283: double (*nrfunc)(double []);
2284:
1.224 brouard 2285: #ifdef LINMINORIGINAL
1.126 brouard 2286: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2287: #else
2288: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2289: #endif
1.126 brouard 2290: {
2291: double brent(double ax, double bx, double cx,
2292: double (*f)(double), double tol, double *xmin);
2293: double f1dim(double x);
2294: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2295: double *fc, double (*func)(double));
2296: int j;
2297: double xx,xmin,bx,ax;
2298: double fx,fb,fa;
1.187 brouard 2299:
1.203 brouard 2300: #ifdef LINMINORIGINAL
2301: #else
2302: double scale=10., axs, xxs; /* Scale added for infinity */
2303: #endif
2304:
1.126 brouard 2305: ncom=n;
2306: pcom=vector(1,n);
2307: xicom=vector(1,n);
2308: nrfunc=func;
2309: for (j=1;j<=n;j++) {
2310: pcom[j]=p[j];
1.202 brouard 2311: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2312: }
1.187 brouard 2313:
1.203 brouard 2314: #ifdef LINMINORIGINAL
2315: xx=1.;
2316: #else
2317: axs=0.0;
2318: xxs=1.;
2319: do{
2320: xx= xxs;
2321: #endif
1.187 brouard 2322: ax=0.;
2323: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2324: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2325: /* 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)) */
2326: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2327: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2328: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2329: /* 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 2330: #ifdef LINMINORIGINAL
2331: #else
2332: if (fx != fx){
1.224 brouard 2333: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2334: printf("|");
2335: fprintf(ficlog,"|");
1.203 brouard 2336: #ifdef DEBUGLINMIN
1.224 brouard 2337: 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 2338: #endif
2339: }
1.224 brouard 2340: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2341: #endif
2342:
1.191 brouard 2343: #ifdef DEBUGLINMIN
2344: 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 2345: 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 2346: #endif
1.224 brouard 2347: #ifdef LINMINORIGINAL
2348: #else
1.317 brouard 2349: if(fb == fx){ /* Flat function in the direction */
2350: xmin=xx;
1.224 brouard 2351: *flat=1;
1.317 brouard 2352: }else{
1.224 brouard 2353: *flat=0;
2354: #endif
2355: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2356: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2357: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2358: /* fmin = f(p[j] + xmin * xi[j]) */
2359: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2360: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2361: #ifdef DEBUG
1.224 brouard 2362: 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);
2363: 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);
2364: #endif
2365: #ifdef LINMINORIGINAL
2366: #else
2367: }
1.126 brouard 2368: #endif
1.191 brouard 2369: #ifdef DEBUGLINMIN
2370: printf("linmin end ");
1.202 brouard 2371: fprintf(ficlog,"linmin end ");
1.191 brouard 2372: #endif
1.126 brouard 2373: for (j=1;j<=n;j++) {
1.203 brouard 2374: #ifdef LINMINORIGINAL
2375: xi[j] *= xmin;
2376: #else
2377: #ifdef DEBUGLINMIN
2378: if(xxs <1.0)
2379: printf(" before xi[%d]=%12.8f", j,xi[j]);
2380: #endif
2381: 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) */
2382: #ifdef DEBUGLINMIN
2383: if(xxs <1.0)
2384: 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 );
2385: #endif
2386: #endif
1.187 brouard 2387: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2388: }
1.191 brouard 2389: #ifdef DEBUGLINMIN
1.203 brouard 2390: printf("\n");
1.191 brouard 2391: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2392: 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 2393: for (j=1;j<=n;j++) {
1.202 brouard 2394: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2395: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2396: if(j % ncovmodel == 0){
1.191 brouard 2397: printf("\n");
1.202 brouard 2398: fprintf(ficlog,"\n");
2399: }
1.191 brouard 2400: }
1.203 brouard 2401: #else
1.191 brouard 2402: #endif
1.126 brouard 2403: free_vector(xicom,1,n);
2404: free_vector(pcom,1,n);
2405: }
2406:
2407:
2408: /*************** powell ************************/
1.162 brouard 2409: /*
1.317 brouard 2410: Minimization of a function func of n variables. Input consists in an initial starting point
2411: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2412: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2413: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2414: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2415: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2416: */
1.224 brouard 2417: #ifdef LINMINORIGINAL
2418: #else
2419: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2420: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2421: #endif
1.126 brouard 2422: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2423: double (*func)(double []))
2424: {
1.224 brouard 2425: #ifdef LINMINORIGINAL
2426: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2427: double (*func)(double []));
1.224 brouard 2428: #else
1.241 brouard 2429: void linmin(double p[], double xi[], int n, double *fret,
2430: double (*func)(double []),int *flat);
1.224 brouard 2431: #endif
1.239 brouard 2432: int i,ibig,j,jk,k;
1.126 brouard 2433: double del,t,*pt,*ptt,*xit;
1.181 brouard 2434: double directest;
1.126 brouard 2435: double fp,fptt;
2436: double *xits;
2437: int niterf, itmp;
2438:
2439: pt=vector(1,n);
2440: ptt=vector(1,n);
2441: xit=vector(1,n);
2442: xits=vector(1,n);
2443: *fret=(*func)(p);
2444: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2445: rcurr_time = time(NULL);
1.126 brouard 2446: for (*iter=1;;++(*iter)) {
2447: ibig=0;
2448: del=0.0;
1.157 brouard 2449: rlast_time=rcurr_time;
2450: /* (void) gettimeofday(&curr_time,&tzp); */
2451: rcurr_time = time(NULL);
2452: curr_time = *localtime(&rcurr_time);
1.324 brouard 2453: 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);
2454: 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 2455: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2456: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2457: for (i=1;i<=n;i++) {
1.126 brouard 2458: fprintf(ficrespow," %.12lf", p[i]);
2459: }
1.239 brouard 2460: fprintf(ficrespow,"\n");fflush(ficrespow);
2461: printf("\n#model= 1 + age ");
2462: fprintf(ficlog,"\n#model= 1 + age ");
2463: if(nagesqr==1){
1.241 brouard 2464: printf(" + age*age ");
2465: fprintf(ficlog," + age*age ");
1.239 brouard 2466: }
2467: for(j=1;j <=ncovmodel-2;j++){
2468: if(Typevar[j]==0) {
2469: printf(" + V%d ",Tvar[j]);
2470: fprintf(ficlog," + V%d ",Tvar[j]);
2471: }else if(Typevar[j]==1) {
2472: printf(" + V%d*age ",Tvar[j]);
2473: fprintf(ficlog," + V%d*age ",Tvar[j]);
2474: }else if(Typevar[j]==2) {
2475: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2476: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2477: }
2478: }
1.126 brouard 2479: printf("\n");
1.239 brouard 2480: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2481: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2482: fprintf(ficlog,"\n");
1.239 brouard 2483: for(i=1,jk=1; i <=nlstate; i++){
2484: for(k=1; k <=(nlstate+ndeath); k++){
2485: if (k != i) {
2486: printf("%d%d ",i,k);
2487: fprintf(ficlog,"%d%d ",i,k);
2488: for(j=1; j <=ncovmodel; j++){
2489: printf("%12.7f ",p[jk]);
2490: fprintf(ficlog,"%12.7f ",p[jk]);
2491: jk++;
2492: }
2493: printf("\n");
2494: fprintf(ficlog,"\n");
2495: }
2496: }
2497: }
1.241 brouard 2498: if(*iter <=3 && *iter >1){
1.157 brouard 2499: tml = *localtime(&rcurr_time);
2500: strcpy(strcurr,asctime(&tml));
2501: rforecast_time=rcurr_time;
1.126 brouard 2502: itmp = strlen(strcurr);
2503: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2504: strcurr[itmp-1]='\0';
1.162 brouard 2505: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2506: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2507: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2508: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2509: forecast_time = *localtime(&rforecast_time);
2510: strcpy(strfor,asctime(&forecast_time));
2511: itmp = strlen(strfor);
2512: if(strfor[itmp-1]=='\n')
2513: strfor[itmp-1]='\0';
2514: 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);
2515: 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 2516: }
2517: }
1.187 brouard 2518: for (i=1;i<=n;i++) { /* For each direction i */
2519: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2520: fptt=(*fret);
2521: #ifdef DEBUG
1.203 brouard 2522: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2523: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2524: #endif
1.203 brouard 2525: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2526: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2527: #ifdef LINMINORIGINAL
1.188 brouard 2528: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2529: #else
2530: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2531: flatdir[i]=flat; /* Function is vanishing in that direction i */
2532: #endif
2533: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2534: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2535: /* because that direction will be replaced unless the gain del is small */
2536: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2537: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2538: /* with the new direction. */
2539: del=fabs(fptt-(*fret));
2540: ibig=i;
1.126 brouard 2541: }
2542: #ifdef DEBUG
2543: printf("%d %.12e",i,(*fret));
2544: fprintf(ficlog,"%d %.12e",i,(*fret));
2545: for (j=1;j<=n;j++) {
1.224 brouard 2546: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2547: printf(" x(%d)=%.12e",j,xit[j]);
2548: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2549: }
2550: for(j=1;j<=n;j++) {
1.225 brouard 2551: printf(" p(%d)=%.12e",j,p[j]);
2552: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2553: }
2554: printf("\n");
2555: fprintf(ficlog,"\n");
2556: #endif
1.187 brouard 2557: } /* end loop on each direction i */
2558: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2559: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2560: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2561: for(j=1;j<=n;j++) {
2562: if(flatdir[j] >0){
2563: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2564: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2565: }
1.319 brouard 2566: /* printf("\n"); */
2567: /* fprintf(ficlog,"\n"); */
2568: }
1.243 brouard 2569: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2570: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2571: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2572: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2573: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2574: /* decreased of more than 3.84 */
2575: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2576: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2577: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2578:
1.188 brouard 2579: /* Starting the program with initial values given by a former maximization will simply change */
2580: /* the scales of the directions and the directions, because the are reset to canonical directions */
2581: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2582: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2583: #ifdef DEBUG
2584: int k[2],l;
2585: k[0]=1;
2586: k[1]=-1;
2587: printf("Max: %.12e",(*func)(p));
2588: fprintf(ficlog,"Max: %.12e",(*func)(p));
2589: for (j=1;j<=n;j++) {
2590: printf(" %.12e",p[j]);
2591: fprintf(ficlog," %.12e",p[j]);
2592: }
2593: printf("\n");
2594: fprintf(ficlog,"\n");
2595: for(l=0;l<=1;l++) {
2596: for (j=1;j<=n;j++) {
2597: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2598: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2599: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2600: }
2601: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2602: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2603: }
2604: #endif
2605:
2606: free_vector(xit,1,n);
2607: free_vector(xits,1,n);
2608: free_vector(ptt,1,n);
2609: free_vector(pt,1,n);
2610: return;
1.192 brouard 2611: } /* enough precision */
1.240 brouard 2612: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2613: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2614: ptt[j]=2.0*p[j]-pt[j];
2615: xit[j]=p[j]-pt[j];
2616: pt[j]=p[j];
2617: }
1.181 brouard 2618: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2619: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2620: if (*iter <=4) {
1.225 brouard 2621: #else
2622: #endif
1.224 brouard 2623: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2624: #else
1.161 brouard 2625: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2626: #endif
1.162 brouard 2627: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2628: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2629: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2630: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2631: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2632: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2633: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2634: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2635: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2636: /* Even if f3 <f1, directest can be negative and t >0 */
2637: /* mu² and del² are equal when f3=f1 */
2638: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2639: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2640: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2641: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2642: #ifdef NRCORIGINAL
2643: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2644: #else
2645: 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 2646: t= t- del*SQR(fp-fptt);
1.183 brouard 2647: #endif
1.202 brouard 2648: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2649: #ifdef DEBUG
1.181 brouard 2650: 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);
2651: 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 2652: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2653: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2654: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2655: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2656: 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);
2657: 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);
2658: #endif
1.183 brouard 2659: #ifdef POWELLORIGINAL
2660: if (t < 0.0) { /* Then we use it for new direction */
2661: #else
1.182 brouard 2662: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2663: 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 2664: 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 2665: 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 2666: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2667: }
1.181 brouard 2668: if (directest < 0.0) { /* Then we use it for new direction */
2669: #endif
1.191 brouard 2670: #ifdef DEBUGLINMIN
1.234 brouard 2671: printf("Before linmin in direction P%d-P0\n",n);
2672: for (j=1;j<=n;j++) {
2673: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2674: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2675: if(j % ncovmodel == 0){
2676: printf("\n");
2677: fprintf(ficlog,"\n");
2678: }
2679: }
1.224 brouard 2680: #endif
2681: #ifdef LINMINORIGINAL
1.234 brouard 2682: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2683: #else
1.234 brouard 2684: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2685: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2686: #endif
1.234 brouard 2687:
1.191 brouard 2688: #ifdef DEBUGLINMIN
1.234 brouard 2689: for (j=1;j<=n;j++) {
2690: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2691: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2692: if(j % ncovmodel == 0){
2693: printf("\n");
2694: fprintf(ficlog,"\n");
2695: }
2696: }
1.224 brouard 2697: #endif
1.234 brouard 2698: for (j=1;j<=n;j++) {
2699: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2700: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2701: }
1.224 brouard 2702: #ifdef LINMINORIGINAL
2703: #else
1.234 brouard 2704: for (j=1, flatd=0;j<=n;j++) {
2705: if(flatdir[j]>0)
2706: flatd++;
2707: }
2708: if(flatd >0){
1.255 brouard 2709: printf("%d flat directions: ",flatd);
2710: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2711: for (j=1;j<=n;j++) {
2712: if(flatdir[j]>0){
2713: printf("%d ",j);
2714: fprintf(ficlog,"%d ",j);
2715: }
2716: }
2717: printf("\n");
2718: fprintf(ficlog,"\n");
1.319 brouard 2719: #ifdef FLATSUP
2720: free_vector(xit,1,n);
2721: free_vector(xits,1,n);
2722: free_vector(ptt,1,n);
2723: free_vector(pt,1,n);
2724: return;
2725: #endif
1.234 brouard 2726: }
1.191 brouard 2727: #endif
1.234 brouard 2728: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2729: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2730:
1.126 brouard 2731: #ifdef DEBUG
1.234 brouard 2732: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2733: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2734: for(j=1;j<=n;j++){
2735: printf(" %lf",xit[j]);
2736: fprintf(ficlog," %lf",xit[j]);
2737: }
2738: printf("\n");
2739: fprintf(ficlog,"\n");
1.126 brouard 2740: #endif
1.192 brouard 2741: } /* end of t or directest negative */
1.224 brouard 2742: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2743: #else
1.234 brouard 2744: } /* end if (fptt < fp) */
1.192 brouard 2745: #endif
1.225 brouard 2746: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2747: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2748: #else
1.224 brouard 2749: #endif
1.234 brouard 2750: } /* loop iteration */
1.126 brouard 2751: }
1.234 brouard 2752:
1.126 brouard 2753: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2754:
1.235 brouard 2755: 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 2756: {
1.279 brouard 2757: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2758: * (and selected quantitative values in nres)
2759: * by left multiplying the unit
2760: * matrix by transitions matrix until convergence is reached with precision ftolpl
2761: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2762: * Wx is row vector: population in state 1, population in state 2, population dead
2763: * or prevalence in state 1, prevalence in state 2, 0
2764: * newm is the matrix after multiplications, its rows are identical at a factor.
2765: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2766: * Output is prlim.
2767: * Initial matrix pimij
2768: */
1.206 brouard 2769: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2770: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2771: /* 0, 0 , 1} */
2772: /*
2773: * and after some iteration: */
2774: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2775: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2776: /* 0, 0 , 1} */
2777: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2778: /* {0.51571254859325999, 0.4842874514067399, */
2779: /* 0.51326036147820708, 0.48673963852179264} */
2780: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2781:
1.126 brouard 2782: int i, ii,j,k;
1.209 brouard 2783: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2784: /* double **matprod2(); */ /* test */
1.218 brouard 2785: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2786: double **newm;
1.209 brouard 2787: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2788: int ncvloop=0;
1.288 brouard 2789: int first=0;
1.169 brouard 2790:
1.209 brouard 2791: min=vector(1,nlstate);
2792: max=vector(1,nlstate);
2793: meandiff=vector(1,nlstate);
2794:
1.218 brouard 2795: /* Starting with matrix unity */
1.126 brouard 2796: for (ii=1;ii<=nlstate+ndeath;ii++)
2797: for (j=1;j<=nlstate+ndeath;j++){
2798: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2799: }
1.169 brouard 2800:
2801: cov[1]=1.;
2802:
2803: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2804: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2805: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2806: ncvloop++;
1.126 brouard 2807: newm=savm;
2808: /* Covariates have to be included here again */
1.138 brouard 2809: cov[2]=agefin;
1.319 brouard 2810: if(nagesqr==1){
2811: cov[3]= agefin*agefin;
2812: }
1.234 brouard 2813: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2814: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
1.330 ! brouard 2815: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];
1.319 brouard 2816: /* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; */
1.235 brouard 2817: /* 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 2818: }
2819: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2820: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.319 brouard 2821: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2822: /* cov[++k1]=Tqresult[nres][k]; */
1.235 brouard 2823: /* 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 2824: }
1.237 brouard 2825: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 2826: if(Dummy[Tage[k]]==2){ /* dummy with age */
1.330 ! brouard 2827: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2];
1.319 brouard 2828: /* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
2829: } else if(Dummy[Tage[k]]==3){ /* quantitative with age */
2830: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2831: /* cov[++k1]=Tqresult[nres][k]; */
1.234 brouard 2832: }
1.235 brouard 2833: /* 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 2834: }
1.237 brouard 2835: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2836: /* 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]); */
1.329 brouard 2837: if(Dummy[Tvard[k][1]]==0){
2838: if(Dummy[Tvard[k][2]]==0){
1.330 ! brouard 2839: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])];
1.319 brouard 2840: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.237 brouard 2841: }else{
1.330 ! brouard 2842: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k];
1.319 brouard 2843: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; */
1.237 brouard 2844: }
2845: }else{
1.329 brouard 2846: if(Dummy[Tvard[k][2]]==0){
1.330 ! brouard 2847: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]];
1.319 brouard 2848: /* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; */
1.237 brouard 2849: }else{
2850: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
1.319 brouard 2851: /* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
1.237 brouard 2852: }
2853: }
1.234 brouard 2854: }
1.138 brouard 2855: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2856: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2857: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2858: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2859: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 2860: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2861: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2862:
1.126 brouard 2863: savm=oldm;
2864: oldm=newm;
1.209 brouard 2865:
2866: for(j=1; j<=nlstate; j++){
2867: max[j]=0.;
2868: min[j]=1.;
2869: }
2870: for(i=1;i<=nlstate;i++){
2871: sumnew=0;
2872: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2873: for(j=1; j<=nlstate; j++){
2874: prlim[i][j]= newm[i][j]/(1-sumnew);
2875: max[j]=FMAX(max[j],prlim[i][j]);
2876: min[j]=FMIN(min[j],prlim[i][j]);
2877: }
2878: }
2879:
1.126 brouard 2880: maxmax=0.;
1.209 brouard 2881: for(j=1; j<=nlstate; j++){
2882: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2883: maxmax=FMAX(maxmax,meandiff[j]);
2884: /* 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 2885: } /* j loop */
1.203 brouard 2886: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2887: /* 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 2888: if(maxmax < ftolpl){
1.209 brouard 2889: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2890: free_vector(min,1,nlstate);
2891: free_vector(max,1,nlstate);
2892: free_vector(meandiff,1,nlstate);
1.126 brouard 2893: return prlim;
2894: }
1.288 brouard 2895: } /* agefin loop */
1.208 brouard 2896: /* After some age loop it doesn't converge */
1.288 brouard 2897: if(!first){
2898: first=1;
2899: 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 2900: 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);
2901: }else if (first >=1 && first <10){
2902: 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);
2903: first++;
2904: }else if (first ==10){
2905: 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);
2906: 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");
2907: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
2908: first++;
1.288 brouard 2909: }
2910:
1.209 brouard 2911: /* 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); */
2912: free_vector(min,1,nlstate);
2913: free_vector(max,1,nlstate);
2914: free_vector(meandiff,1,nlstate);
1.208 brouard 2915:
1.169 brouard 2916: return prlim; /* should not reach here */
1.126 brouard 2917: }
2918:
1.217 brouard 2919:
2920: /**** Back Prevalence limit (stable or period prevalence) ****************/
2921:
1.218 brouard 2922: /* 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) */
2923: /* 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 2924: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2925: {
1.264 brouard 2926: /* 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 2927: matrix by transitions matrix until convergence is reached with precision ftolpl */
2928: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2929: /* Wx is row vector: population in state 1, population in state 2, population dead */
2930: /* or prevalence in state 1, prevalence in state 2, 0 */
2931: /* newm is the matrix after multiplications, its rows are identical at a factor */
2932: /* Initial matrix pimij */
2933: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2934: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2935: /* 0, 0 , 1} */
2936: /*
2937: * and after some iteration: */
2938: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2939: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2940: /* 0, 0 , 1} */
2941: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2942: /* {0.51571254859325999, 0.4842874514067399, */
2943: /* 0.51326036147820708, 0.48673963852179264} */
2944: /* If we start from prlim again, prlim tends to a constant matrix */
2945:
2946: int i, ii,j,k;
1.247 brouard 2947: int first=0;
1.217 brouard 2948: double *min, *max, *meandiff, maxmax,sumnew=0.;
2949: /* double **matprod2(); */ /* test */
2950: double **out, cov[NCOVMAX+1], **bmij();
2951: double **newm;
1.218 brouard 2952: double **dnewm, **doldm, **dsavm; /* for use */
2953: double **oldm, **savm; /* for use */
2954:
1.217 brouard 2955: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2956: int ncvloop=0;
2957:
2958: min=vector(1,nlstate);
2959: max=vector(1,nlstate);
2960: meandiff=vector(1,nlstate);
2961:
1.266 brouard 2962: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2963: oldm=oldms; savm=savms;
2964:
2965: /* Starting with matrix unity */
2966: for (ii=1;ii<=nlstate+ndeath;ii++)
2967: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2968: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2969: }
2970:
2971: cov[1]=1.;
2972:
2973: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2974: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2975: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2976: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2977: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2978: ncvloop++;
1.218 brouard 2979: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2980: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2981: /* Covariates have to be included here again */
2982: cov[2]=agefin;
1.319 brouard 2983: if(nagesqr==1){
1.217 brouard 2984: cov[3]= agefin*agefin;;
1.319 brouard 2985: }
1.242 brouard 2986: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2987: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
1.330 ! brouard 2988: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];
1.264 brouard 2989: /* 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 2990: }
2991: /* for (k=1; k<=cptcovn;k++) { */
2992: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2993: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2994: /* /\* 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])]); *\/ */
2995: /* } */
2996: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2997: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2998: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2999: /* 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]); */
3000: }
3001: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
3002: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
3003: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
3004: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
3005: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 3006: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age *\/ ERROR ???*/
3007: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.330 ! brouard 3008: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2];
1.319 brouard 3009: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
3010: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.242 brouard 3011: }
3012: /* 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]); */
3013: }
3014: for (k=1; k<=cptcovprod;k++){ /* For product without age */
3015: /* 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]); */
1.329 brouard 3016: if(Dummy[Tvard[k][1]]==0){
3017: if(Dummy[Tvard[k][2]]==0){
1.330 ! brouard 3018: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])];
1.242 brouard 3019: }else{
1.330 ! brouard 3020: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k];
1.242 brouard 3021: }
3022: }else{
1.329 brouard 3023: if(Dummy[Tvard[k][2]]==0){
1.330 ! brouard 3024: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]];
1.242 brouard 3025: }else{
3026: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3027: }
3028: }
1.217 brouard 3029: }
3030:
3031: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3032: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3033: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3034: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3035: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3036: /* ij should be linked to the correct index of cov */
3037: /* age and covariate values ij are in 'cov', but we need to pass
3038: * ij for the observed prevalence at age and status and covariate
3039: * number: prevacurrent[(int)agefin][ii][ij]
3040: */
3041: /* 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 *\/ */
3042: /* 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 *\/ */
3043: 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 3044: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3045: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3046: /* for(i=1; i<=nlstate+ndeath; i++) { */
3047: /* printf("%d newm= ",i); */
3048: /* for(j=1;j<=nlstate+ndeath;j++) { */
3049: /* printf("%f ",newm[i][j]); */
3050: /* } */
3051: /* printf("oldm * "); */
3052: /* for(j=1;j<=nlstate+ndeath;j++) { */
3053: /* printf("%f ",oldm[i][j]); */
3054: /* } */
1.268 brouard 3055: /* printf(" bmmij "); */
1.266 brouard 3056: /* for(j=1;j<=nlstate+ndeath;j++) { */
3057: /* printf("%f ",pmmij[i][j]); */
3058: /* } */
3059: /* printf("\n"); */
3060: /* } */
3061: /* } */
1.217 brouard 3062: savm=oldm;
3063: oldm=newm;
1.266 brouard 3064:
1.217 brouard 3065: for(j=1; j<=nlstate; j++){
3066: max[j]=0.;
3067: min[j]=1.;
3068: }
3069: for(j=1; j<=nlstate; j++){
3070: for(i=1;i<=nlstate;i++){
1.234 brouard 3071: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3072: bprlim[i][j]= newm[i][j];
3073: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3074: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3075: }
3076: }
1.218 brouard 3077:
1.217 brouard 3078: maxmax=0.;
3079: for(i=1; i<=nlstate; i++){
1.318 brouard 3080: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3081: maxmax=FMAX(maxmax,meandiff[i]);
3082: /* 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 3083: } /* i loop */
1.217 brouard 3084: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3085: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3086: if(maxmax < ftolpl){
1.220 brouard 3087: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3088: free_vector(min,1,nlstate);
3089: free_vector(max,1,nlstate);
3090: free_vector(meandiff,1,nlstate);
3091: return bprlim;
3092: }
1.288 brouard 3093: } /* agefin loop */
1.217 brouard 3094: /* After some age loop it doesn't converge */
1.288 brouard 3095: if(!first){
1.247 brouard 3096: first=1;
3097: 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\
3098: 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);
3099: }
3100: 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 3101: 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);
3102: /* 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); */
3103: free_vector(min,1,nlstate);
3104: free_vector(max,1,nlstate);
3105: free_vector(meandiff,1,nlstate);
3106:
3107: return bprlim; /* should not reach here */
3108: }
3109:
1.126 brouard 3110: /*************** transition probabilities ***************/
3111:
3112: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3113: {
1.138 brouard 3114: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3115: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3116: model to the ncovmodel covariates (including constant and age).
3117: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3118: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3119: ncth covariate in the global vector x is given by the formula:
3120: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3121: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3122: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3123: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3124: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3125: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3126: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3127: */
3128: double s1, lnpijopii;
1.126 brouard 3129: /*double t34;*/
1.164 brouard 3130: int i,j, nc, ii, jj;
1.126 brouard 3131:
1.223 brouard 3132: for(i=1; i<= nlstate; i++){
3133: for(j=1; j<i;j++){
3134: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3135: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3136: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3137: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3138: }
3139: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 ! brouard 3140: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3141: }
3142: for(j=i+1; j<=nlstate+ndeath;j++){
3143: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3144: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3145: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3146: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3147: }
3148: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 ! brouard 3149: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3150: }
3151: }
1.218 brouard 3152:
1.223 brouard 3153: for(i=1; i<= nlstate; i++){
3154: s1=0;
3155: for(j=1; j<i; j++){
3156: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 ! brouard 3157: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
1.223 brouard 3158: }
3159: for(j=i+1; j<=nlstate+ndeath; j++){
3160: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 ! brouard 3161: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
1.223 brouard 3162: }
3163: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3164: ps[i][i]=1./(s1+1.);
3165: /* Computing other pijs */
3166: for(j=1; j<i; j++)
1.325 brouard 3167: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3168: for(j=i+1; j<=nlstate+ndeath; j++)
3169: ps[i][j]= exp(ps[i][j])*ps[i][i];
3170: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3171: } /* end i */
1.218 brouard 3172:
1.223 brouard 3173: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3174: for(jj=1; jj<= nlstate+ndeath; jj++){
3175: ps[ii][jj]=0;
3176: ps[ii][ii]=1;
3177: }
3178: }
1.294 brouard 3179:
3180:
1.223 brouard 3181: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3182: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3183: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3184: /* } */
3185: /* printf("\n "); */
3186: /* } */
3187: /* printf("\n ");printf("%lf ",cov[2]);*/
3188: /*
3189: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3190: goto end;*/
1.266 brouard 3191: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3192: }
3193:
1.218 brouard 3194: /*************** backward transition probabilities ***************/
3195:
3196: /* 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 ) */
3197: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3198: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3199: {
1.302 brouard 3200: /* 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 3201: * 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 3202: */
1.218 brouard 3203: int i, ii, j,k;
1.222 brouard 3204:
3205: double **out, **pmij();
3206: double sumnew=0.;
1.218 brouard 3207: double agefin;
1.292 brouard 3208: 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 3209: double **dnewm, **dsavm, **doldm;
3210: double **bbmij;
3211:
1.218 brouard 3212: doldm=ddoldms; /* global pointers */
1.222 brouard 3213: dnewm=ddnewms;
3214: dsavm=ddsavms;
1.318 brouard 3215:
3216: /* Debug */
3217: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3218: agefin=cov[2];
1.268 brouard 3219: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3220: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3221: the observed prevalence (with this covariate ij) at beginning of transition */
3222: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3223:
3224: /* P_x */
1.325 brouard 3225: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3226: /* outputs pmmij which is a stochastic matrix in row */
3227:
3228: /* Diag(w_x) */
1.292 brouard 3229: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3230: sumnew=0.;
1.269 brouard 3231: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3232: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3233: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3234: sumnew+=prevacurrent[(int)agefin][ii][ij];
3235: }
3236: if(sumnew >0.01){ /* At least some value in the prevalence */
3237: for (ii=1;ii<=nlstate+ndeath;ii++){
3238: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3239: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3240: }
3241: }else{
3242: for (ii=1;ii<=nlstate+ndeath;ii++){
3243: for (j=1;j<=nlstate+ndeath;j++)
3244: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3245: }
3246: /* if(sumnew <0.9){ */
3247: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3248: /* } */
3249: }
3250: k3=0.0; /* We put the last diagonal to 0 */
3251: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3252: doldm[ii][ii]= k3;
3253: }
3254: /* End doldm, At the end doldm is diag[(w_i)] */
3255:
1.292 brouard 3256: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3257: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3258:
1.292 brouard 3259: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3260: /* 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 3261: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3262: sumnew=0.;
1.222 brouard 3263: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3264: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3265: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3266: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3267: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3268: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3269: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3270: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3271: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3272: /* }else */
1.268 brouard 3273: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3274: } /*End ii */
3275: } /* 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 */
3276:
1.292 brouard 3277: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3278: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3279: /* end bmij */
1.266 brouard 3280: return ps; /*pointer is unchanged */
1.218 brouard 3281: }
1.217 brouard 3282: /*************** transition probabilities ***************/
3283:
1.218 brouard 3284: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3285: {
3286: /* According to parameters values stored in x and the covariate's values stored in cov,
3287: computes the probability to be observed in state j being in state i by appying the
3288: model to the ncovmodel covariates (including constant and age).
3289: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3290: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3291: ncth covariate in the global vector x is given by the formula:
3292: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3293: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3294: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3295: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3296: Outputs ps[i][j] the probability to be observed in j being in j according to
3297: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3298: */
3299: double s1, lnpijopii;
3300: /*double t34;*/
3301: int i,j, nc, ii, jj;
3302:
1.234 brouard 3303: for(i=1; i<= nlstate; i++){
3304: for(j=1; j<i;j++){
3305: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3306: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3307: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3308: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3309: }
3310: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3311: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3312: }
3313: for(j=i+1; j<=nlstate+ndeath;j++){
3314: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3315: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3316: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3317: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3318: }
3319: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3320: }
3321: }
3322:
3323: for(i=1; i<= nlstate; i++){
3324: s1=0;
3325: for(j=1; j<i; j++){
3326: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3327: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3328: }
3329: for(j=i+1; j<=nlstate+ndeath; j++){
3330: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3331: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3332: }
3333: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3334: ps[i][i]=1./(s1+1.);
3335: /* Computing other pijs */
3336: for(j=1; j<i; j++)
3337: ps[i][j]= exp(ps[i][j])*ps[i][i];
3338: for(j=i+1; j<=nlstate+ndeath; j++)
3339: ps[i][j]= exp(ps[i][j])*ps[i][i];
3340: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3341: } /* end i */
3342:
3343: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3344: for(jj=1; jj<= nlstate+ndeath; jj++){
3345: ps[ii][jj]=0;
3346: ps[ii][ii]=1;
3347: }
3348: }
1.296 brouard 3349: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3350: for(jj=1; jj<= nlstate+ndeath; jj++){
3351: s1=0.;
3352: for(ii=1; ii<= nlstate+ndeath; ii++){
3353: s1+=ps[ii][jj];
3354: }
3355: for(ii=1; ii<= nlstate; ii++){
3356: ps[ii][jj]=ps[ii][jj]/s1;
3357: }
3358: }
3359: /* Transposition */
3360: for(jj=1; jj<= nlstate+ndeath; jj++){
3361: for(ii=jj; ii<= nlstate+ndeath; ii++){
3362: s1=ps[ii][jj];
3363: ps[ii][jj]=ps[jj][ii];
3364: ps[jj][ii]=s1;
3365: }
3366: }
3367: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3368: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3369: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3370: /* } */
3371: /* printf("\n "); */
3372: /* } */
3373: /* printf("\n ");printf("%lf ",cov[2]);*/
3374: /*
3375: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3376: goto end;*/
3377: return ps;
1.217 brouard 3378: }
3379:
3380:
1.126 brouard 3381: /**************** Product of 2 matrices ******************/
3382:
1.145 brouard 3383: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3384: {
3385: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3386: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3387: /* in, b, out are matrice of pointers which should have been initialized
3388: before: only the contents of out is modified. The function returns
3389: a pointer to pointers identical to out */
1.145 brouard 3390: int i, j, k;
1.126 brouard 3391: for(i=nrl; i<= nrh; i++)
1.145 brouard 3392: for(k=ncolol; k<=ncoloh; k++){
3393: out[i][k]=0.;
3394: for(j=ncl; j<=nch; j++)
3395: out[i][k] +=in[i][j]*b[j][k];
3396: }
1.126 brouard 3397: return out;
3398: }
3399:
3400:
3401: /************* Higher Matrix Product ***************/
3402:
1.235 brouard 3403: 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 3404: {
1.218 brouard 3405: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3406: 'nhstepm*hstepm*stepm' months (i.e. until
3407: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3408: nhstepm*hstepm matrices.
3409: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3410: (typically every 2 years instead of every month which is too big
3411: for the memory).
3412: Model is determined by parameters x and covariates have to be
3413: included manually here.
3414:
3415: */
3416:
1.330 ! brouard 3417: int i, j, d, h, k, k1;
1.131 brouard 3418: double **out, cov[NCOVMAX+1];
1.126 brouard 3419: double **newm;
1.187 brouard 3420: double agexact;
1.214 brouard 3421: double agebegin, ageend;
1.126 brouard 3422:
3423: /* Hstepm could be zero and should return the unit matrix */
3424: for (i=1;i<=nlstate+ndeath;i++)
3425: for (j=1;j<=nlstate+ndeath;j++){
3426: oldm[i][j]=(i==j ? 1.0 : 0.0);
3427: po[i][j][0]=(i==j ? 1.0 : 0.0);
3428: }
3429: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3430: for(h=1; h <=nhstepm; h++){
3431: for(d=1; d <=hstepm; d++){
3432: newm=savm;
3433: /* Covariates have to be included here again */
3434: cov[1]=1.;
1.214 brouard 3435: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3436: cov[2]=agexact;
1.319 brouard 3437: if(nagesqr==1){
1.227 brouard 3438: cov[3]= agexact*agexact;
1.319 brouard 3439: }
1.330 ! brouard 3440: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
! 3441: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
! 3442: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
! 3443: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
! 3444: /* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) */
! 3445: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
! 3446: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
1.319 brouard 3447: /* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 */
3448: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3449: /* k 1 2 3 4 5 6 7 8 9 */
3450: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
3451: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
3452: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
3453: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1.330 ! brouard 3454: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] */
! 3455: cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]];
! 3456: /* printf("hpxij Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,TnsdVar[TvarsD[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,TnsdVar[TvarsD[k]])); */
! 3457: printf("hpxij Dummy combi=%d k1=%d Tvar[%d]=V%d cov[2+%d+%d]=%lf resultmodel[nres][%d]=%d nres/nresult=%d/%d \n",ij,k1,k1, Tvar[k1],nagesqr,k1,cov[2+nagesqr+k1],k1,resultmodel[nres][k1],nres,nresult);
! 3458: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative variables */
! 3459: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
! 3460: cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]];
! 3461: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
! 3462: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
! 3463: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
! 3464: printf("hPxij Quantitative k1=%d resultmodel[nres][%d]=%d,Tqresult[%d][%d]=%f\n",k1,k1,resultmodel[nres][k1],nres,resultmodel[nres][k1],Tqresult[nres][resultmodel[nres][k1]]);
! 3465: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
! 3466: /* Tvar[k1] Variable in the age product age*V1 is 1 */
! 3467: /* [Tinvresult[nres][V1] is its value in the resultline nres */
! 3468: cov[2+nagesqr+k1]=Tinvresult[nres][Tvar[k1]];
! 3469: printf("DhPxij Dummy with age k1=%d Tvar[%d]=%d Tinvresult[nres][%d]=%d,cov[2+%d+%d]=%.3f\n",k1,k1,Tvar[k1],Tinvresult[nres][Tvar[k1]],nagesqr,k1,cov[2+nagesqr+k1]);
! 3470: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
! 3471: /* for (k=1; k<=cptcovage;k++){ /\* For product with age V1+V1*age +V4 +age*V3 *\/ */
! 3472: /* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*/
! 3473: /* */
! 3474: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
! 3475: /* k 1 2 3 4 5 6 7 8 9 */
! 3476: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
! 3477: /*cptcovage=2 1 2 */
! 3478: /*Tage[k]= 5 8 */
! 3479: }else if( Dummy[k1]==2 ){ /* For quant with age product */
! 3480: cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]];
! 3481: printf("QhPxij Quant with age k1=%d resultmodel[nres][%d]=%d,Tqresult[%d][%d]=%f\n",k1,k1,resultmodel[nres][k1],nres,resultmodel[nres][k1],Tqresult[nres][resultmodel[nres][k1]]);
! 3482: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
! 3483: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ *\/ */
! 3484: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
! 3485: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
! 3486: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; */
! 3487: /* printf("hPxij Age combi=%d k=%d cptcovage=%d Tage[%d]=%d Tvar[Tage[%d]]=V%d nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[Tvar[Tage[k]]]])]=%d nres=%d\n",ij,k,cptcovage,k,Tage[k],k,Tvar[Tage[k]], nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[Tvar[Tage[k]]])],nres); */
! 3488: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
! 3489: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
! 3490: /* } */
! 3491: /* 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]); */
! 3492: }else if(Typevar[k1]==2 ){ /* For product (not with age) */
! 3493: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
! 3494: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
! 3495: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
! 3496: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
! 3497: /* /\*cptcovprod=1 1 2 *\/ */
! 3498: /* /\*Tprod[]= 4 7 *\/ */
! 3499: /* /\*Tvard[][1] 4 1 *\/ */
! 3500: /* /\*Tvard[][2] 3 2 *\/ */
! 3501:
! 3502: /* printf("hPxij Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]=%d nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]=%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2],nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])],nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]); */
! 3503: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
! 3504: cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
! 3505: printf("hPxij Prod ij=%d k1=%d cov[2+%d+%d]=%.5f Tvard[%d][1]=V%d * Tvard[%d][2]=V%d ; TinvDoQresult[nres][Tvardk[k1][1]]=%.4f * TinvDoQresult[nres][Tvardk[k1][1]]=%.4f\n",ij,k1,nagesqr,k1,cov[2+nagesqr+k1],k1,Tvard[k1][1], k1,Tvard[k1][2], TinvDoQresult[nres][Tvardk[k1][1]], TinvDoQresult[nres][Tvardk[k1][2]]);
! 3506: /* if(Dummy[Tvardk[k1][1]]==0){ */
! 3507: /* if(Dummy[Tvardk[k1][2]]==0){ /\* Product of dummies *\/ */
! 3508: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
! 3509: /* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; */
! 3510: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])]; */
! 3511: /* }else{ /\* Product of dummy by quantitative *\/ */
! 3512: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; */
! 3513: /* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; */
! 3514: /* } */
! 3515: /* }else{ /\* Product of quantitative by...*\/ */
! 3516: /* if(Dummy[Tvard[k][2]]==0){ /\* quant by dummy *\/ */
! 3517: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
! 3518: /* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; */
! 3519: /* }else{ /\* Product of two quant *\/ */
! 3520: /* /\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
! 3521: /* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; */
! 3522: /* } */
! 3523: /* }/\*end of products quantitative *\/ */
! 3524: }/*end of products */
! 3525: } /* End of loop on model equation */
1.235 brouard 3526: /* for (k=1; k<=cptcovn;k++) */
3527: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3528: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3529: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3530: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3531: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3532:
3533:
1.126 brouard 3534: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3535: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3536: /* right multiplication of oldm by the current matrix */
1.126 brouard 3537: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3538: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3539: /* if((int)age == 70){ */
3540: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3541: /* for(i=1; i<=nlstate+ndeath; i++) { */
3542: /* printf("%d pmmij ",i); */
3543: /* for(j=1;j<=nlstate+ndeath;j++) { */
3544: /* printf("%f ",pmmij[i][j]); */
3545: /* } */
3546: /* printf(" oldm "); */
3547: /* for(j=1;j<=nlstate+ndeath;j++) { */
3548: /* printf("%f ",oldm[i][j]); */
3549: /* } */
3550: /* printf("\n"); */
3551: /* } */
3552: /* } */
1.126 brouard 3553: savm=oldm;
3554: oldm=newm;
3555: }
3556: for(i=1; i<=nlstate+ndeath; i++)
3557: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3558: po[i][j][h]=newm[i][j];
3559: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3560: }
1.128 brouard 3561: /*printf("h=%d ",h);*/
1.126 brouard 3562: } /* end h */
1.267 brouard 3563: /* printf("\n H=%d \n",h); */
1.126 brouard 3564: return po;
3565: }
3566:
1.217 brouard 3567: /************* Higher Back Matrix Product ***************/
1.218 brouard 3568: /* 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 3569: 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 3570: {
1.266 brouard 3571: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3572: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3573: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3574: nhstepm*hstepm matrices.
3575: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3576: (typically every 2 years instead of every month which is too big
1.217 brouard 3577: for the memory).
1.218 brouard 3578: Model is determined by parameters x and covariates have to be
1.266 brouard 3579: included manually here. Then we use a call to bmij(x and cov)
3580: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3581: */
1.217 brouard 3582:
3583: int i, j, d, h, k;
1.266 brouard 3584: double **out, cov[NCOVMAX+1], **bmij();
3585: double **newm, ***newmm;
1.217 brouard 3586: double agexact;
3587: double agebegin, ageend;
1.222 brouard 3588: double **oldm, **savm;
1.217 brouard 3589:
1.266 brouard 3590: newmm=po; /* To be saved */
3591: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3592: /* Hstepm could be zero and should return the unit matrix */
3593: for (i=1;i<=nlstate+ndeath;i++)
3594: for (j=1;j<=nlstate+ndeath;j++){
3595: oldm[i][j]=(i==j ? 1.0 : 0.0);
3596: po[i][j][0]=(i==j ? 1.0 : 0.0);
3597: }
3598: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3599: for(h=1; h <=nhstepm; h++){
3600: for(d=1; d <=hstepm; d++){
3601: newm=savm;
3602: /* Covariates have to be included here again */
3603: cov[1]=1.;
1.271 brouard 3604: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3605: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3606: /* Debug */
3607: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3608: cov[2]=agexact;
3609: if(nagesqr==1)
1.222 brouard 3610: cov[3]= agexact*agexact;
1.325 brouard 3611: for (k=1; k<=nsd;k++){ /* For single dummy covariates only *//* cptcovn error */
1.266 brouard 3612: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3613: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
1.330 ! brouard 3614: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/* Bug valgrind */
1.266 brouard 3615: /* 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)); */
3616: }
1.267 brouard 3617: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3618: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3619: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3620: /* 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]); */
3621: }
1.319 brouard 3622: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 *//* For product with age */
3623: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age error!!!*\/ */
3624: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.330 ! brouard 3625: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2];
1.319 brouard 3626: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
1.267 brouard 3627: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3628: }
3629: /* 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]); */
3630: }
3631: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.330 ! brouard 3632: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])];
1.329 brouard 3633: if(Dummy[Tvard[k][1]]==0){
3634: if(Dummy[Tvard[k][2]]==0){
1.330 ! brouard 3635: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])];
1.325 brouard 3636: }else{
1.330 ! brouard 3637: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k];
1.325 brouard 3638: }
3639: }else{
1.329 brouard 3640: if(Dummy[Tvard[k][2]]==0){
1.330 ! brouard 3641: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]];
1.325 brouard 3642: }else{
3643: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3644: }
3645: }
1.267 brouard 3646: }
1.217 brouard 3647: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3648: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3649:
1.218 brouard 3650: /* Careful transposed matrix */
1.266 brouard 3651: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3652: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3653: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3654: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3655: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3656: /* if((int)age == 70){ */
3657: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3658: /* for(i=1; i<=nlstate+ndeath; i++) { */
3659: /* printf("%d pmmij ",i); */
3660: /* for(j=1;j<=nlstate+ndeath;j++) { */
3661: /* printf("%f ",pmmij[i][j]); */
3662: /* } */
3663: /* printf(" oldm "); */
3664: /* for(j=1;j<=nlstate+ndeath;j++) { */
3665: /* printf("%f ",oldm[i][j]); */
3666: /* } */
3667: /* printf("\n"); */
3668: /* } */
3669: /* } */
3670: savm=oldm;
3671: oldm=newm;
3672: }
3673: for(i=1; i<=nlstate+ndeath; i++)
3674: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3675: po[i][j][h]=newm[i][j];
1.268 brouard 3676: /* if(h==nhstepm) */
3677: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3678: }
1.268 brouard 3679: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3680: } /* end h */
1.268 brouard 3681: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3682: return po;
3683: }
3684:
3685:
1.162 brouard 3686: #ifdef NLOPT
3687: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3688: double fret;
3689: double *xt;
3690: int j;
3691: myfunc_data *d2 = (myfunc_data *) pd;
3692: /* xt = (p1-1); */
3693: xt=vector(1,n);
3694: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3695:
3696: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3697: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3698: printf("Function = %.12lf ",fret);
3699: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3700: printf("\n");
3701: free_vector(xt,1,n);
3702: return fret;
3703: }
3704: #endif
1.126 brouard 3705:
3706: /*************** log-likelihood *************/
3707: double func( double *x)
3708: {
1.226 brouard 3709: int i, ii, j, k, mi, d, kk;
3710: int ioffset=0;
3711: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3712: double **out;
3713: double lli; /* Individual log likelihood */
3714: int s1, s2;
1.228 brouard 3715: 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 3716: double bbh, survp;
3717: long ipmx;
3718: double agexact;
3719: /*extern weight */
3720: /* We are differentiating ll according to initial status */
3721: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3722: /*for(i=1;i<imx;i++)
3723: printf(" %d\n",s[4][i]);
3724: */
1.162 brouard 3725:
1.226 brouard 3726: ++countcallfunc;
1.162 brouard 3727:
1.226 brouard 3728: cov[1]=1.;
1.126 brouard 3729:
1.226 brouard 3730: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3731: ioffset=0;
1.226 brouard 3732: if(mle==1){
3733: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3734: /* Computes the values of the ncovmodel covariates of the model
3735: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3736: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3737: to be observed in j being in i according to the model.
3738: */
1.243 brouard 3739: ioffset=2+nagesqr ;
1.233 brouard 3740: /* Fixed */
1.319 brouard 3741: for (k=1; k<=ncovf;k++){ /* For each fixed covariate dummu or quant or prod */
3742: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3743: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3744: /* 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 3745: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.319 brouard 3746: 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)*/
3747: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3748: }
1.226 brouard 3749: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3750: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3751: has been calculated etc */
3752: /* For an individual i, wav[i] gives the number of effective waves */
3753: /* We compute the contribution to Likelihood of each effective transition
3754: mw[mi][i] is real wave of the mi th effectve wave */
3755: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3756: s2=s[mw[mi+1][i]][i];
3757: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3758: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3759: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3760: */
3761: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3762: 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*/
3763: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? */
1.242 brouard 3764: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3765: }
3766: for (ii=1;ii<=nlstate+ndeath;ii++)
3767: for (j=1;j<=nlstate+ndeath;j++){
3768: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3769: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3770: }
3771: for(d=0; d<dh[mi][i]; d++){
3772: newm=savm;
3773: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3774: cov[2]=agexact;
3775: if(nagesqr==1)
3776: cov[3]= agexact*agexact; /* Should be changed here */
3777: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3778: if(!FixedV[Tvar[Tage[kk]]])
3779: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3780: else
3781: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3782: }
3783: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3784: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3785: savm=oldm;
3786: oldm=newm;
3787: } /* end mult */
3788:
3789: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3790: /* But now since version 0.9 we anticipate for bias at large stepm.
3791: * If stepm is larger than one month (smallest stepm) and if the exact delay
3792: * (in months) between two waves is not a multiple of stepm, we rounded to
3793: * the nearest (and in case of equal distance, to the lowest) interval but now
3794: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3795: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3796: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3797: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3798: * -stepm/2 to stepm/2 .
3799: * For stepm=1 the results are the same as for previous versions of Imach.
3800: * For stepm > 1 the results are less biased than in previous versions.
3801: */
1.234 brouard 3802: s1=s[mw[mi][i]][i];
3803: s2=s[mw[mi+1][i]][i];
3804: bbh=(double)bh[mi][i]/(double)stepm;
3805: /* bias bh is positive if real duration
3806: * is higher than the multiple of stepm and negative otherwise.
3807: */
3808: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3809: if( s2 > nlstate){
3810: /* i.e. if s2 is a death state and if the date of death is known
3811: then the contribution to the likelihood is the probability to
3812: die between last step unit time and current step unit time,
3813: which is also equal to probability to die before dh
3814: minus probability to die before dh-stepm .
3815: In version up to 0.92 likelihood was computed
3816: as if date of death was unknown. Death was treated as any other
3817: health state: the date of the interview describes the actual state
3818: and not the date of a change in health state. The former idea was
3819: to consider that at each interview the state was recorded
3820: (healthy, disable or death) and IMaCh was corrected; but when we
3821: introduced the exact date of death then we should have modified
3822: the contribution of an exact death to the likelihood. This new
3823: contribution is smaller and very dependent of the step unit
3824: stepm. It is no more the probability to die between last interview
3825: and month of death but the probability to survive from last
3826: interview up to one month before death multiplied by the
3827: probability to die within a month. Thanks to Chris
3828: Jackson for correcting this bug. Former versions increased
3829: mortality artificially. The bad side is that we add another loop
3830: which slows down the processing. The difference can be up to 10%
3831: lower mortality.
3832: */
3833: /* If, at the beginning of the maximization mostly, the
3834: cumulative probability or probability to be dead is
3835: constant (ie = 1) over time d, the difference is equal to
3836: 0. out[s1][3] = savm[s1][3]: probability, being at state
3837: s1 at precedent wave, to be dead a month before current
3838: wave is equal to probability, being at state s1 at
3839: precedent wave, to be dead at mont of the current
3840: wave. Then the observed probability (that this person died)
3841: is null according to current estimated parameter. In fact,
3842: it should be very low but not zero otherwise the log go to
3843: infinity.
3844: */
1.183 brouard 3845: /* #ifdef INFINITYORIGINAL */
3846: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3847: /* #else */
3848: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3849: /* lli=log(mytinydouble); */
3850: /* else */
3851: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3852: /* #endif */
1.226 brouard 3853: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3854:
1.226 brouard 3855: } else if ( s2==-1 ) { /* alive */
3856: for (j=1,survp=0. ; j<=nlstate; j++)
3857: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3858: /*survp += out[s1][j]; */
3859: lli= log(survp);
3860: }
3861: else if (s2==-4) {
3862: for (j=3,survp=0. ; j<=nlstate; j++)
3863: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3864: lli= log(survp);
3865: }
3866: else if (s2==-5) {
3867: for (j=1,survp=0. ; j<=2; j++)
3868: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3869: lli= log(survp);
3870: }
3871: else{
3872: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3873: /* 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 */
3874: }
3875: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3876: /*if(lli ==000.0)*/
3877: /*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); */
3878: ipmx +=1;
3879: sw += weight[i];
3880: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3881: /* if (lli < log(mytinydouble)){ */
3882: /* 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); */
3883: /* 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]); */
3884: /* } */
3885: } /* end of wave */
3886: } /* end of individual */
3887: } else if(mle==2){
3888: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 3889: ioffset=2+nagesqr ;
3890: for (k=1; k<=ncovf;k++)
3891: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 3892: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3893: for(k=1; k <= ncovv ; k++){
3894: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3895: }
1.226 brouard 3896: for (ii=1;ii<=nlstate+ndeath;ii++)
3897: for (j=1;j<=nlstate+ndeath;j++){
3898: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3899: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3900: }
3901: for(d=0; d<=dh[mi][i]; d++){
3902: newm=savm;
3903: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3904: cov[2]=agexact;
3905: if(nagesqr==1)
3906: cov[3]= agexact*agexact;
3907: for (kk=1; kk<=cptcovage;kk++) {
3908: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3909: }
3910: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3911: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3912: savm=oldm;
3913: oldm=newm;
3914: } /* end mult */
3915:
3916: s1=s[mw[mi][i]][i];
3917: s2=s[mw[mi+1][i]][i];
3918: bbh=(double)bh[mi][i]/(double)stepm;
3919: 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 */
3920: ipmx +=1;
3921: sw += weight[i];
3922: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3923: } /* end of wave */
3924: } /* end of individual */
3925: } else if(mle==3){ /* exponential inter-extrapolation */
3926: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3927: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3928: for(mi=1; mi<= wav[i]-1; mi++){
3929: for (ii=1;ii<=nlstate+ndeath;ii++)
3930: for (j=1;j<=nlstate+ndeath;j++){
3931: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3932: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3933: }
3934: for(d=0; d<dh[mi][i]; d++){
3935: newm=savm;
3936: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3937: cov[2]=agexact;
3938: if(nagesqr==1)
3939: cov[3]= agexact*agexact;
3940: for (kk=1; kk<=cptcovage;kk++) {
3941: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3942: }
3943: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3944: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3945: savm=oldm;
3946: oldm=newm;
3947: } /* end mult */
3948:
3949: s1=s[mw[mi][i]][i];
3950: s2=s[mw[mi+1][i]][i];
3951: bbh=(double)bh[mi][i]/(double)stepm;
3952: 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 */
3953: ipmx +=1;
3954: sw += weight[i];
3955: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3956: } /* end of wave */
3957: } /* end of individual */
3958: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3959: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3960: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3961: for(mi=1; mi<= wav[i]-1; mi++){
3962: for (ii=1;ii<=nlstate+ndeath;ii++)
3963: for (j=1;j<=nlstate+ndeath;j++){
3964: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3965: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3966: }
3967: for(d=0; d<dh[mi][i]; d++){
3968: newm=savm;
3969: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3970: cov[2]=agexact;
3971: if(nagesqr==1)
3972: cov[3]= agexact*agexact;
3973: for (kk=1; kk<=cptcovage;kk++) {
3974: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3975: }
1.126 brouard 3976:
1.226 brouard 3977: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3978: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3979: savm=oldm;
3980: oldm=newm;
3981: } /* end mult */
3982:
3983: s1=s[mw[mi][i]][i];
3984: s2=s[mw[mi+1][i]][i];
3985: if( s2 > nlstate){
3986: lli=log(out[s1][s2] - savm[s1][s2]);
3987: } else if ( s2==-1 ) { /* alive */
3988: for (j=1,survp=0. ; j<=nlstate; j++)
3989: survp += out[s1][j];
3990: lli= log(survp);
3991: }else{
3992: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3993: }
3994: ipmx +=1;
3995: sw += weight[i];
3996: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3997: /* 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 3998: } /* end of wave */
3999: } /* end of individual */
4000: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4001: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4002: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4003: for(mi=1; mi<= wav[i]-1; mi++){
4004: for (ii=1;ii<=nlstate+ndeath;ii++)
4005: for (j=1;j<=nlstate+ndeath;j++){
4006: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4007: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4008: }
4009: for(d=0; d<dh[mi][i]; d++){
4010: newm=savm;
4011: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4012: cov[2]=agexact;
4013: if(nagesqr==1)
4014: cov[3]= agexact*agexact;
4015: for (kk=1; kk<=cptcovage;kk++) {
4016: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4017: }
1.126 brouard 4018:
1.226 brouard 4019: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4020: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4021: savm=oldm;
4022: oldm=newm;
4023: } /* end mult */
4024:
4025: s1=s[mw[mi][i]][i];
4026: s2=s[mw[mi+1][i]][i];
4027: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4028: ipmx +=1;
4029: sw += weight[i];
4030: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4031: /*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]);*/
4032: } /* end of wave */
4033: } /* end of individual */
4034: } /* End of if */
4035: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4036: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4037: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4038: return -l;
1.126 brouard 4039: }
4040:
4041: /*************** log-likelihood *************/
4042: double funcone( double *x)
4043: {
1.228 brouard 4044: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 4045: int i, ii, j, k, mi, d, kk;
1.228 brouard 4046: int ioffset=0;
1.131 brouard 4047: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4048: double **out;
4049: double lli; /* Individual log likelihood */
4050: double llt;
4051: int s1, s2;
1.228 brouard 4052: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4053:
1.126 brouard 4054: double bbh, survp;
1.187 brouard 4055: double agexact;
1.214 brouard 4056: double agebegin, ageend;
1.126 brouard 4057: /*extern weight */
4058: /* We are differentiating ll according to initial status */
4059: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4060: /*for(i=1;i<imx;i++)
4061: printf(" %d\n",s[4][i]);
4062: */
4063: cov[1]=1.;
4064:
4065: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4066: ioffset=0;
4067: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 4068: /* ioffset=2+nagesqr+cptcovage; */
4069: ioffset=2+nagesqr;
1.232 brouard 4070: /* Fixed */
1.224 brouard 4071: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4072: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.311 brouard 4073: 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 4074: 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)*/
4075: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4076: /* cov[2+6]=covar[Tvar[6]][i]; */
4077: /* cov[2+6]=covar[2][i]; V2 */
4078: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4079: /* cov[2+7]=covar[Tvar[7]][i]; */
4080: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4081: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4082: /* cov[2+9]=covar[Tvar[9]][i]; */
4083: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4084: }
1.232 brouard 4085: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4086: /* 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?)*\/ */
4087: /* } */
1.231 brouard 4088: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4089: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4090: /* } */
1.225 brouard 4091:
1.233 brouard 4092:
4093: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 4094: /* Wave varying (but not age varying) */
4095: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 4096: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
4097: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4098: }
1.232 brouard 4099: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 4100: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4101: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4102: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4103: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4104: /* 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 4105: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4106: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4107: /* /\* 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]); *\/ */
4108: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4109: /* } */
1.126 brouard 4110: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4111: for (j=1;j<=nlstate+ndeath;j++){
4112: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4113: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4114: }
1.214 brouard 4115:
4116: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4117: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4118: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4119: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4120: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4121: and mw[mi+1][i]. dh depends on stepm.*/
4122: newm=savm;
1.247 brouard 4123: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4124: cov[2]=agexact;
4125: if(nagesqr==1)
4126: cov[3]= agexact*agexact;
4127: for (kk=1; kk<=cptcovage;kk++) {
4128: if(!FixedV[Tvar[Tage[kk]]])
4129: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4130: else
4131: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
4132: }
4133: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4134: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4135: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4136: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4137: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4138: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4139: savm=oldm;
4140: oldm=newm;
1.126 brouard 4141: } /* end mult */
4142:
4143: s1=s[mw[mi][i]][i];
4144: s2=s[mw[mi+1][i]][i];
1.217 brouard 4145: /* if(s2==-1){ */
1.268 brouard 4146: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4147: /* /\* exit(1); *\/ */
4148: /* } */
1.126 brouard 4149: bbh=(double)bh[mi][i]/(double)stepm;
4150: /* bias is positive if real duration
4151: * is higher than the multiple of stepm and negative otherwise.
4152: */
4153: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4154: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4155: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4156: for (j=1,survp=0. ; j<=nlstate; j++)
4157: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4158: lli= log(survp);
1.126 brouard 4159: }else if (mle==1){
1.242 brouard 4160: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4161: } else if(mle==2){
1.242 brouard 4162: 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 4163: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4164: 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 4165: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4166: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4167: } else{ /* mle=0 back to 1 */
1.242 brouard 4168: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4169: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4170: } /* End of if */
4171: ipmx +=1;
4172: sw += weight[i];
4173: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 4174: /*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 4175: if(globpr){
1.246 brouard 4176: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4177: %11.6f %11.6f %11.6f ", \
1.242 brouard 4178: 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 4179: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 4180: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4181: llt +=ll[k]*gipmx/gsw;
4182: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
4183: }
4184: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 4185: }
1.232 brouard 4186: } /* end of wave */
4187: } /* end of individual */
4188: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4189: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4190: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4191: if(globpr==0){ /* First time we count the contributions and weights */
4192: gipmx=ipmx;
4193: gsw=sw;
4194: }
4195: return -l;
1.126 brouard 4196: }
4197:
4198:
4199: /*************** function likelione ***********/
1.292 brouard 4200: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4201: {
4202: /* This routine should help understanding what is done with
4203: the selection of individuals/waves and
4204: to check the exact contribution to the likelihood.
4205: Plotting could be done.
4206: */
4207: int k;
4208:
4209: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4210: strcpy(fileresilk,"ILK_");
1.202 brouard 4211: strcat(fileresilk,fileresu);
1.126 brouard 4212: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4213: printf("Problem with resultfile: %s\n", fileresilk);
4214: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4215: }
1.214 brouard 4216: 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");
4217: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4218: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4219: for(k=1; k<=nlstate; k++)
4220: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4221: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4222: }
4223:
1.292 brouard 4224: *fretone=(*func)(p);
1.126 brouard 4225: if(*globpri !=0){
4226: fclose(ficresilk);
1.205 brouard 4227: if (mle ==0)
4228: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4229: else if(mle >=1)
4230: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4231: 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 4232: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4233:
4234: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4235: 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 4236: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4237: }
1.207 brouard 4238: 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 4239: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4240: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4241: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4242: fflush(fichtm);
1.205 brouard 4243: }
1.126 brouard 4244: return;
4245: }
4246:
4247:
4248: /*********** Maximum Likelihood Estimation ***************/
4249:
4250: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4251: {
1.319 brouard 4252: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4253: double **xi;
4254: double fret;
4255: double fretone; /* Only one call to likelihood */
4256: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4257:
4258: #ifdef NLOPT
4259: int creturn;
4260: nlopt_opt opt;
4261: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4262: double *lb;
4263: double minf; /* the minimum objective value, upon return */
4264: double * p1; /* Shifted parameters from 0 instead of 1 */
4265: myfunc_data dinst, *d = &dinst;
4266: #endif
4267:
4268:
1.126 brouard 4269: xi=matrix(1,npar,1,npar);
4270: for (i=1;i<=npar;i++)
4271: for (j=1;j<=npar;j++)
4272: xi[i][j]=(i==j ? 1.0 : 0.0);
4273: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4274: strcpy(filerespow,"POW_");
1.126 brouard 4275: strcat(filerespow,fileres);
4276: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4277: printf("Problem with resultfile: %s\n", filerespow);
4278: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4279: }
4280: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4281: for (i=1;i<=nlstate;i++)
4282: for(j=1;j<=nlstate+ndeath;j++)
4283: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4284: fprintf(ficrespow,"\n");
1.162 brouard 4285: #ifdef POWELL
1.319 brouard 4286: #ifdef LINMINORIGINAL
4287: #else /* LINMINORIGINAL */
4288:
4289: flatdir=ivector(1,npar);
4290: for (j=1;j<=npar;j++) flatdir[j]=0;
4291: #endif /*LINMINORIGINAL */
4292:
4293: #ifdef FLATSUP
4294: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4295: /* reorganizing p by suppressing flat directions */
4296: for(i=1, jk=1; i <=nlstate; i++){
4297: for(k=1; k <=(nlstate+ndeath); k++){
4298: if (k != i) {
4299: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4300: if(flatdir[jk]==1){
4301: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4302: }
4303: for(j=1; j <=ncovmodel; j++){
4304: printf("%12.7f ",p[jk]);
4305: jk++;
4306: }
4307: printf("\n");
4308: }
4309: }
4310: }
4311: /* skipping */
4312: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4313: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4314: for(k=1; k <=(nlstate+ndeath); k++){
4315: if (k != i) {
4316: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4317: if(flatdir[jk]==1){
4318: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4319: for(j=1; j <=ncovmodel; jk++,j++){
4320: printf(" p[%d]=%12.7f",jk, p[jk]);
4321: /*q[jjk]=p[jk];*/
4322: }
4323: }else{
4324: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4325: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4326: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4327: /*q[jjk]=p[jk];*/
4328: }
4329: }
4330: printf("\n");
4331: }
4332: fflush(stdout);
4333: }
4334: }
4335: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4336: #else /* FLATSUP */
1.126 brouard 4337: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4338: #endif /* FLATSUP */
4339:
4340: #ifdef LINMINORIGINAL
4341: #else
4342: free_ivector(flatdir,1,npar);
4343: #endif /* LINMINORIGINAL*/
4344: #endif /* POWELL */
1.126 brouard 4345:
1.162 brouard 4346: #ifdef NLOPT
4347: #ifdef NEWUOA
4348: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4349: #else
4350: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4351: #endif
4352: lb=vector(0,npar-1);
4353: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4354: nlopt_set_lower_bounds(opt, lb);
4355: nlopt_set_initial_step1(opt, 0.1);
4356:
4357: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4358: d->function = func;
4359: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4360: nlopt_set_min_objective(opt, myfunc, d);
4361: nlopt_set_xtol_rel(opt, ftol);
4362: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4363: printf("nlopt failed! %d\n",creturn);
4364: }
4365: else {
4366: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4367: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4368: iter=1; /* not equal */
4369: }
4370: nlopt_destroy(opt);
4371: #endif
1.319 brouard 4372: #ifdef FLATSUP
4373: /* npared = npar -flatd/ncovmodel; */
4374: /* xired= matrix(1,npared,1,npared); */
4375: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4376: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4377: /* free_matrix(xire,1,npared,1,npared); */
4378: #else /* FLATSUP */
4379: #endif /* FLATSUP */
1.126 brouard 4380: free_matrix(xi,1,npar,1,npar);
4381: fclose(ficrespow);
1.203 brouard 4382: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4383: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4384: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4385:
4386: }
4387:
4388: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4389: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4390: {
4391: double **a,**y,*x,pd;
1.203 brouard 4392: /* double **hess; */
1.164 brouard 4393: int i, j;
1.126 brouard 4394: int *indx;
4395:
4396: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4397: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4398: void lubksb(double **a, int npar, int *indx, double b[]) ;
4399: void ludcmp(double **a, int npar, int *indx, double *d) ;
4400: double gompertz(double p[]);
1.203 brouard 4401: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4402:
4403: printf("\nCalculation of the hessian matrix. Wait...\n");
4404: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4405: for (i=1;i<=npar;i++){
1.203 brouard 4406: printf("%d-",i);fflush(stdout);
4407: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4408:
4409: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4410:
4411: /* printf(" %f ",p[i]);
4412: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4413: }
4414:
4415: for (i=1;i<=npar;i++) {
4416: for (j=1;j<=npar;j++) {
4417: if (j>i) {
1.203 brouard 4418: printf(".%d-%d",i,j);fflush(stdout);
4419: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4420: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4421:
4422: hess[j][i]=hess[i][j];
4423: /*printf(" %lf ",hess[i][j]);*/
4424: }
4425: }
4426: }
4427: printf("\n");
4428: fprintf(ficlog,"\n");
4429:
4430: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4431: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4432:
4433: a=matrix(1,npar,1,npar);
4434: y=matrix(1,npar,1,npar);
4435: x=vector(1,npar);
4436: indx=ivector(1,npar);
4437: for (i=1;i<=npar;i++)
4438: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4439: ludcmp(a,npar,indx,&pd);
4440:
4441: for (j=1;j<=npar;j++) {
4442: for (i=1;i<=npar;i++) x[i]=0;
4443: x[j]=1;
4444: lubksb(a,npar,indx,x);
4445: for (i=1;i<=npar;i++){
4446: matcov[i][j]=x[i];
4447: }
4448: }
4449:
4450: printf("\n#Hessian matrix#\n");
4451: fprintf(ficlog,"\n#Hessian matrix#\n");
4452: for (i=1;i<=npar;i++) {
4453: for (j=1;j<=npar;j++) {
1.203 brouard 4454: printf("%.6e ",hess[i][j]);
4455: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4456: }
4457: printf("\n");
4458: fprintf(ficlog,"\n");
4459: }
4460:
1.203 brouard 4461: /* printf("\n#Covariance matrix#\n"); */
4462: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4463: /* for (i=1;i<=npar;i++) { */
4464: /* for (j=1;j<=npar;j++) { */
4465: /* printf("%.6e ",matcov[i][j]); */
4466: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4467: /* } */
4468: /* printf("\n"); */
4469: /* fprintf(ficlog,"\n"); */
4470: /* } */
4471:
1.126 brouard 4472: /* Recompute Inverse */
1.203 brouard 4473: /* for (i=1;i<=npar;i++) */
4474: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4475: /* ludcmp(a,npar,indx,&pd); */
4476:
4477: /* printf("\n#Hessian matrix recomputed#\n"); */
4478:
4479: /* for (j=1;j<=npar;j++) { */
4480: /* for (i=1;i<=npar;i++) x[i]=0; */
4481: /* x[j]=1; */
4482: /* lubksb(a,npar,indx,x); */
4483: /* for (i=1;i<=npar;i++){ */
4484: /* y[i][j]=x[i]; */
4485: /* printf("%.3e ",y[i][j]); */
4486: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4487: /* } */
4488: /* printf("\n"); */
4489: /* fprintf(ficlog,"\n"); */
4490: /* } */
4491:
4492: /* Verifying the inverse matrix */
4493: #ifdef DEBUGHESS
4494: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4495:
1.203 brouard 4496: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4497: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4498:
4499: for (j=1;j<=npar;j++) {
4500: for (i=1;i<=npar;i++){
1.203 brouard 4501: printf("%.2f ",y[i][j]);
4502: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4503: }
4504: printf("\n");
4505: fprintf(ficlog,"\n");
4506: }
1.203 brouard 4507: #endif
1.126 brouard 4508:
4509: free_matrix(a,1,npar,1,npar);
4510: free_matrix(y,1,npar,1,npar);
4511: free_vector(x,1,npar);
4512: free_ivector(indx,1,npar);
1.203 brouard 4513: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4514:
4515:
4516: }
4517:
4518: /*************** hessian matrix ****************/
4519: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4520: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4521: int i;
4522: int l=1, lmax=20;
1.203 brouard 4523: double k1,k2, res, fx;
1.132 brouard 4524: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4525: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4526: int k=0,kmax=10;
4527: double l1;
4528:
4529: fx=func(x);
4530: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4531: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4532: l1=pow(10,l);
4533: delts=delt;
4534: for(k=1 ; k <kmax; k=k+1){
4535: delt = delta*(l1*k);
4536: p2[theta]=x[theta] +delt;
1.145 brouard 4537: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4538: p2[theta]=x[theta]-delt;
4539: k2=func(p2)-fx;
4540: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4541: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4542:
1.203 brouard 4543: #ifdef DEBUGHESSII
1.126 brouard 4544: 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);
4545: 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);
4546: #endif
4547: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4548: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4549: k=kmax;
4550: }
4551: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4552: k=kmax; l=lmax*10;
1.126 brouard 4553: }
4554: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4555: delts=delt;
4556: }
1.203 brouard 4557: } /* End loop k */
1.126 brouard 4558: }
4559: delti[theta]=delts;
4560: return res;
4561:
4562: }
4563:
1.203 brouard 4564: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4565: {
4566: int i;
1.164 brouard 4567: int l=1, lmax=20;
1.126 brouard 4568: double k1,k2,k3,k4,res,fx;
1.132 brouard 4569: double p2[MAXPARM+1];
1.203 brouard 4570: int k, kmax=1;
4571: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4572:
4573: int firstime=0;
1.203 brouard 4574:
1.126 brouard 4575: fx=func(x);
1.203 brouard 4576: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4577: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4578: p2[thetai]=x[thetai]+delti[thetai]*k;
4579: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4580: k1=func(p2)-fx;
4581:
1.203 brouard 4582: p2[thetai]=x[thetai]+delti[thetai]*k;
4583: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4584: k2=func(p2)-fx;
4585:
1.203 brouard 4586: p2[thetai]=x[thetai]-delti[thetai]*k;
4587: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4588: k3=func(p2)-fx;
4589:
1.203 brouard 4590: p2[thetai]=x[thetai]-delti[thetai]*k;
4591: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4592: k4=func(p2)-fx;
1.203 brouard 4593: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4594: if(k1*k2*k3*k4 <0.){
1.208 brouard 4595: firstime=1;
1.203 brouard 4596: kmax=kmax+10;
1.208 brouard 4597: }
4598: if(kmax >=10 || firstime ==1){
1.246 brouard 4599: 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);
4600: 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 4601: 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);
4602: 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);
4603: }
4604: #ifdef DEBUGHESSIJ
4605: v1=hess[thetai][thetai];
4606: v2=hess[thetaj][thetaj];
4607: cv12=res;
4608: /* Computing eigen value of Hessian matrix */
4609: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4610: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4611: if ((lc2 <0) || (lc1 <0) ){
4612: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4613: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4614: 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);
4615: 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);
4616: }
1.126 brouard 4617: #endif
4618: }
4619: return res;
4620: }
4621:
1.203 brouard 4622: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4623: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4624: /* { */
4625: /* int i; */
4626: /* int l=1, lmax=20; */
4627: /* double k1,k2,k3,k4,res,fx; */
4628: /* double p2[MAXPARM+1]; */
4629: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4630: /* int k=0,kmax=10; */
4631: /* double l1; */
4632:
4633: /* fx=func(x); */
4634: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4635: /* l1=pow(10,l); */
4636: /* delts=delt; */
4637: /* for(k=1 ; k <kmax; k=k+1){ */
4638: /* delt = delti*(l1*k); */
4639: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4640: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4641: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4642: /* k1=func(p2)-fx; */
4643:
4644: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4645: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4646: /* k2=func(p2)-fx; */
4647:
4648: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4649: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4650: /* k3=func(p2)-fx; */
4651:
4652: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4653: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4654: /* k4=func(p2)-fx; */
4655: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4656: /* #ifdef DEBUGHESSIJ */
4657: /* 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); */
4658: /* 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); */
4659: /* #endif */
4660: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4661: /* k=kmax; */
4662: /* } */
4663: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4664: /* k=kmax; l=lmax*10; */
4665: /* } */
4666: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4667: /* delts=delt; */
4668: /* } */
4669: /* } /\* End loop k *\/ */
4670: /* } */
4671: /* delti[theta]=delts; */
4672: /* return res; */
4673: /* } */
4674:
4675:
1.126 brouard 4676: /************** Inverse of matrix **************/
4677: void ludcmp(double **a, int n, int *indx, double *d)
4678: {
4679: int i,imax,j,k;
4680: double big,dum,sum,temp;
4681: double *vv;
4682:
4683: vv=vector(1,n);
4684: *d=1.0;
4685: for (i=1;i<=n;i++) {
4686: big=0.0;
4687: for (j=1;j<=n;j++)
4688: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4689: if (big == 0.0){
4690: printf(" Singular Hessian matrix at row %d:\n",i);
4691: for (j=1;j<=n;j++) {
4692: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4693: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4694: }
4695: fflush(ficlog);
4696: fclose(ficlog);
4697: nrerror("Singular matrix in routine ludcmp");
4698: }
1.126 brouard 4699: vv[i]=1.0/big;
4700: }
4701: for (j=1;j<=n;j++) {
4702: for (i=1;i<j;i++) {
4703: sum=a[i][j];
4704: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4705: a[i][j]=sum;
4706: }
4707: big=0.0;
4708: for (i=j;i<=n;i++) {
4709: sum=a[i][j];
4710: for (k=1;k<j;k++)
4711: sum -= a[i][k]*a[k][j];
4712: a[i][j]=sum;
4713: if ( (dum=vv[i]*fabs(sum)) >= big) {
4714: big=dum;
4715: imax=i;
4716: }
4717: }
4718: if (j != imax) {
4719: for (k=1;k<=n;k++) {
4720: dum=a[imax][k];
4721: a[imax][k]=a[j][k];
4722: a[j][k]=dum;
4723: }
4724: *d = -(*d);
4725: vv[imax]=vv[j];
4726: }
4727: indx[j]=imax;
4728: if (a[j][j] == 0.0) a[j][j]=TINY;
4729: if (j != n) {
4730: dum=1.0/(a[j][j]);
4731: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4732: }
4733: }
4734: free_vector(vv,1,n); /* Doesn't work */
4735: ;
4736: }
4737:
4738: void lubksb(double **a, int n, int *indx, double b[])
4739: {
4740: int i,ii=0,ip,j;
4741: double sum;
4742:
4743: for (i=1;i<=n;i++) {
4744: ip=indx[i];
4745: sum=b[ip];
4746: b[ip]=b[i];
4747: if (ii)
4748: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4749: else if (sum) ii=i;
4750: b[i]=sum;
4751: }
4752: for (i=n;i>=1;i--) {
4753: sum=b[i];
4754: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4755: b[i]=sum/a[i][i];
4756: }
4757: }
4758:
4759: void pstamp(FILE *fichier)
4760: {
1.196 brouard 4761: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4762: }
4763:
1.297 brouard 4764: void date2dmy(double date,double *day, double *month, double *year){
4765: double yp=0., yp1=0., yp2=0.;
4766:
4767: yp1=modf(date,&yp);/* extracts integral of date in yp and
4768: fractional in yp1 */
4769: *year=yp;
4770: yp2=modf((yp1*12),&yp);
4771: *month=yp;
4772: yp1=modf((yp2*30.5),&yp);
4773: *day=yp;
4774: if(*day==0) *day=1;
4775: if(*month==0) *month=1;
4776: }
4777:
1.253 brouard 4778:
4779:
1.126 brouard 4780: /************ Frequencies ********************/
1.251 brouard 4781: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4782: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4783: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4784: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4785:
1.265 brouard 4786: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4787: int iind=0, iage=0;
4788: int mi; /* Effective wave */
4789: int first;
4790: double ***freq; /* Frequencies */
1.268 brouard 4791: 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 */
4792: 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 4793: double *meanq, *stdq, *idq;
1.226 brouard 4794: double **meanqt;
4795: double *pp, **prop, *posprop, *pospropt;
4796: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4797: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4798: double agebegin, ageend;
4799:
4800: pp=vector(1,nlstate);
1.251 brouard 4801: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4802: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4803: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4804: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4805: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4806: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4807: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4808: meanqt=matrix(1,lastpass,1,nqtveff);
4809: strcpy(fileresp,"P_");
4810: strcat(fileresp,fileresu);
4811: /*strcat(fileresphtm,fileresu);*/
4812: if((ficresp=fopen(fileresp,"w"))==NULL) {
4813: printf("Problem with prevalence resultfile: %s\n", fileresp);
4814: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4815: exit(0);
4816: }
1.240 brouard 4817:
1.226 brouard 4818: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4819: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4820: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4821: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4822: fflush(ficlog);
4823: exit(70);
4824: }
4825: else{
4826: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4827: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4828: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4829: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4830: }
1.319 brouard 4831: 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 4832:
1.226 brouard 4833: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4834: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4835: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4836: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4837: fflush(ficlog);
4838: exit(70);
1.240 brouard 4839: } else{
1.226 brouard 4840: 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 4841: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4842: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4843: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4844: }
1.319 brouard 4845: 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 4846:
1.253 brouard 4847: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4848: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4849: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4850: j1=0;
1.126 brouard 4851:
1.227 brouard 4852: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4853: j=cptcoveff; /* Only dummy covariates of the model */
1.330 ! brouard 4854: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 4855: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4856:
4857:
1.226 brouard 4858: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4859: reference=low_education V1=0,V2=0
4860: med_educ V1=1 V2=0,
4861: high_educ V1=0 V2=1
1.330 ! brouard 4862: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 4863: */
1.249 brouard 4864: dateintsum=0;
4865: k2cpt=0;
4866:
1.253 brouard 4867: if(cptcoveff == 0 )
1.265 brouard 4868: nl=1; /* Constant and age model only */
1.253 brouard 4869: else
4870: nl=2;
1.265 brouard 4871:
4872: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4873: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.330 ! brouard 4874: * Loop on j1(1 to 2**cptcovn) covariate combination
1.265 brouard 4875: * freq[s1][s2][iage] =0.
4876: * Loop on iind
4877: * ++freq[s1][s2][iage] weighted
4878: * end iind
4879: * if covariate and j!0
4880: * headers Variable on one line
4881: * endif cov j!=0
4882: * header of frequency table by age
4883: * Loop on age
4884: * pp[s1]+=freq[s1][s2][iage] weighted
4885: * pos+=freq[s1][s2][iage] weighted
4886: * Loop on s1 initial state
4887: * fprintf(ficresp
4888: * end s1
4889: * end age
4890: * if j!=0 computes starting values
4891: * end compute starting values
4892: * end j1
4893: * end nl
4894: */
1.253 brouard 4895: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4896: if(nj==1)
4897: j=0; /* First pass for the constant */
1.265 brouard 4898: else{
1.330 ! brouard 4899: j=cptcovs; /* Other passes for the covariate values */
1.265 brouard 4900: }
1.251 brouard 4901: first=1;
1.265 brouard 4902: 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 4903: posproptt=0.;
1.330 ! brouard 4904: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 4905: scanf("%d", i);*/
4906: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4907: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4908: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4909: freq[i][s2][m]=0;
1.251 brouard 4910:
4911: for (i=1; i<=nlstate; i++) {
1.240 brouard 4912: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4913: prop[i][m]=0;
4914: posprop[i]=0;
4915: pospropt[i]=0;
4916: }
1.283 brouard 4917: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4918: idq[z1]=0.;
4919: meanq[z1]=0.;
4920: stdq[z1]=0.;
1.283 brouard 4921: }
4922: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4923: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4924: /* meanqt[m][z1]=0.; */
4925: /* } */
4926: /* } */
1.251 brouard 4927: /* dateintsum=0; */
4928: /* k2cpt=0; */
4929:
1.265 brouard 4930: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4931: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4932: bool=1;
4933: if(j !=0){
4934: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.330 ! brouard 4935: if (cptcovn >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
! 4936: for (z1=1; z1<=cptcovn; z1++) { /* loops on covariates in the model */
1.251 brouard 4937: /* if(Tvaraff[z1] ==-20){ */
4938: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4939: /* }else if(Tvaraff[z1] ==-10){ */
4940: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 ! brouard 4941: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
! 4942: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]){ /* for combination j1 of covariates */
1.265 brouard 4943: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4944: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4945: /* 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",
4946: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4947: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4948: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4949: } /* Onlyf fixed */
4950: } /* end z1 */
4951: } /* cptcovn > 0 */
4952: } /* end any */
4953: }/* end j==0 */
1.265 brouard 4954: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4955: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4956: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4957: m=mw[mi][iind];
4958: if(j!=0){
4959: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.330 ! brouard 4960: for (z1=1; z1<=cptcovn; z1++) {
1.251 brouard 4961: if( Fixed[Tmodelind[z1]]==1){
4962: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.330 ! brouard 4963: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]) /* iv=1 to ntv, right modality. If covariate's
1.251 brouard 4964: value is -1, we don't select. It differs from the
4965: constant and age model which counts them. */
4966: bool=0; /* not selected */
4967: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.330 ! brouard 4968: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]) {
1.251 brouard 4969: bool=0;
4970: }
4971: }
4972: }
4973: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4974: } /* end j==0 */
4975: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4976: if(bool==1){ /*Selected */
1.251 brouard 4977: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4978: and mw[mi+1][iind]. dh depends on stepm. */
4979: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4980: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4981: if(m >=firstpass && m <=lastpass){
4982: k2=anint[m][iind]+(mint[m][iind]/12.);
4983: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4984: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4985: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4986: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4987: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4988: if (m<lastpass) {
4989: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4990: /* 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]); */
4991: if(s[m][iind]==-1)
4992: 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.));
4993: 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 4994: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
4995: if(!isnan(covar[ncovcol+z1][iind])){
4996: idq[z1]=idq[z1]+weight[iind];
4997: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4998: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
4999: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
5000: }
1.284 brouard 5001: }
1.251 brouard 5002: /* if((int)agev[m][iind] == 55) */
5003: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5004: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5005: 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 5006: }
1.251 brouard 5007: } /* end if between passes */
5008: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5009: dateintsum=dateintsum+k2; /* on all covariates ?*/
5010: k2cpt++;
5011: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5012: }
1.251 brouard 5013: }else{
5014: bool=1;
5015: }/* end bool 2 */
5016: } /* end m */
1.284 brouard 5017: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5018: /* idq[z1]=idq[z1]+weight[iind]; */
5019: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5020: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5021: /* } */
1.251 brouard 5022: } /* end bool */
5023: } /* end iind = 1 to imx */
1.319 brouard 5024: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5025: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5026:
5027:
5028: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.330 ! brouard 5029: if(cptcovn==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5030: pstamp(ficresp);
1.330 ! brouard 5031: if (cptcovn>0 && j!=0){
1.265 brouard 5032: pstamp(ficresp);
1.251 brouard 5033: printf( "\n#********** Variable ");
5034: fprintf(ficresp, "\n#********** Variable ");
5035: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5036: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5037: fprintf(ficlog, "\n#********** Variable ");
1.330 ! brouard 5038: for (z1=1; z1<=cptcovs; z1++){
1.251 brouard 5039: if(!FixedV[Tvaraff[z1]]){
1.330 ! brouard 5040: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
! 5041: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
! 5042: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
! 5043: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
! 5044: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5045: }else{
1.330 ! brouard 5046: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
! 5047: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
! 5048: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
! 5049: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
! 5050: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5051: }
5052: }
5053: printf( "**********\n#");
5054: fprintf(ficresp, "**********\n#");
5055: fprintf(ficresphtm, "**********</h3>\n");
5056: fprintf(ficresphtmfr, "**********</h3>\n");
5057: fprintf(ficlog, "**********\n");
5058: }
1.284 brouard 5059: /*
5060: Printing means of quantitative variables if any
5061: */
5062: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5063: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5064: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5065: if(weightopt==1){
5066: printf(" Weighted mean and standard deviation of");
5067: fprintf(ficlog," Weighted mean and standard deviation of");
5068: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5069: }
1.311 brouard 5070: /* mu = \frac{w x}{\sum w}
5071: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5072: */
5073: 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]));
5074: 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]));
5075: 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 5076: }
5077: /* for (z1=1; z1<= nqtveff; z1++) { */
5078: /* for(m=1;m<=lastpass;m++){ */
5079: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5080: /* } */
5081: /* } */
1.283 brouard 5082:
1.251 brouard 5083: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.330 ! brouard 5084: if((cptcovn==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5085: fprintf(ficresp, " Age");
1.330 ! brouard 5086: if(nj==2) for (z1=1; z1<=cptcovn; z1++) fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]);
1.251 brouard 5087: for(i=1; i<=nlstate;i++) {
1.330 ! brouard 5088: if((cptcovn==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5089: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5090: }
1.330 ! brouard 5091: if((cptcovn==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5092: fprintf(ficresphtm, "\n");
5093:
5094: /* Header of frequency table by age */
5095: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5096: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5097: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5098: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5099: if(s2!=0 && m!=0)
5100: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5101: }
1.226 brouard 5102: }
1.251 brouard 5103: fprintf(ficresphtmfr, "\n");
5104:
5105: /* For each age */
5106: for(iage=iagemin; iage <= iagemax+3; iage++){
5107: fprintf(ficresphtm,"<tr>");
5108: if(iage==iagemax+1){
5109: fprintf(ficlog,"1");
5110: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5111: }else if(iage==iagemax+2){
5112: fprintf(ficlog,"0");
5113: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5114: }else if(iage==iagemax+3){
5115: fprintf(ficlog,"Total");
5116: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5117: }else{
1.240 brouard 5118: if(first==1){
1.251 brouard 5119: first=0;
5120: printf("See log file for details...\n");
5121: }
5122: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5123: fprintf(ficlog,"Age %d", iage);
5124: }
1.265 brouard 5125: for(s1=1; s1 <=nlstate ; s1++){
5126: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5127: pp[s1] += freq[s1][m][iage];
1.251 brouard 5128: }
1.265 brouard 5129: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5130: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5131: pos += freq[s1][m][iage];
5132: if(pp[s1]>=1.e-10){
1.251 brouard 5133: if(first==1){
1.265 brouard 5134: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5135: }
1.265 brouard 5136: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5137: }else{
5138: if(first==1)
1.265 brouard 5139: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5140: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5141: }
5142: }
5143:
1.265 brouard 5144: for(s1=1; s1 <=nlstate ; s1++){
5145: /* posprop[s1]=0; */
5146: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5147: pp[s1] += freq[s1][m][iage];
5148: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5149:
5150: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5151: pos += pp[s1]; /* pos is the total number of transitions until this age */
5152: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5153: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5154: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5155: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5156: }
5157:
5158: /* Writing ficresp */
1.330 ! brouard 5159: if(cptcovn==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5160: if( iage <= iagemax){
5161: fprintf(ficresp," %d",iage);
5162: }
5163: }else if( nj==2){
5164: if( iage <= iagemax){
5165: fprintf(ficresp," %d",iage);
1.330 ! brouard 5166: for (z1=1; z1<=cptcovn; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]);
1.265 brouard 5167: }
1.240 brouard 5168: }
1.265 brouard 5169: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5170: if(pos>=1.e-5){
1.251 brouard 5171: if(first==1)
1.265 brouard 5172: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5173: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5174: }else{
5175: if(first==1)
1.265 brouard 5176: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5177: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5178: }
5179: if( iage <= iagemax){
5180: if(pos>=1.e-5){
1.330 ! brouard 5181: if(cptcovn==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5182: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5183: }else if( nj==2){
5184: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5185: }
5186: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5187: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5188: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5189: } else{
1.330 ! brouard 5190: if((cptcovn==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5191: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5192: }
1.240 brouard 5193: }
1.265 brouard 5194: pospropt[s1] +=posprop[s1];
5195: } /* end loop s1 */
1.251 brouard 5196: /* pospropt=0.; */
1.265 brouard 5197: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5198: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5199: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5200: if(first==1){
1.265 brouard 5201: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5202: }
1.265 brouard 5203: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5204: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5205: }
1.265 brouard 5206: if(s1!=0 && m!=0)
5207: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5208: }
1.265 brouard 5209: } /* end loop s1 */
1.251 brouard 5210: posproptt=0.;
1.265 brouard 5211: for(s1=1; s1 <=nlstate; s1++){
5212: posproptt += pospropt[s1];
1.251 brouard 5213: }
5214: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5215: fprintf(ficresphtm,"</tr>\n");
1.330 ! brouard 5216: if((cptcovn==0 && nj==1)|| nj==2 ) {
1.265 brouard 5217: if(iage <= iagemax)
5218: fprintf(ficresp,"\n");
1.240 brouard 5219: }
1.251 brouard 5220: if(first==1)
5221: printf("Others in log...\n");
5222: fprintf(ficlog,"\n");
5223: } /* end loop age iage */
1.265 brouard 5224:
1.251 brouard 5225: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5226: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5227: if(posproptt < 1.e-5){
1.265 brouard 5228: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5229: }else{
1.265 brouard 5230: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5231: }
1.226 brouard 5232: }
1.251 brouard 5233: fprintf(ficresphtm,"</tr>\n");
5234: fprintf(ficresphtm,"</table>\n");
5235: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5236: if(posproptt < 1.e-5){
1.251 brouard 5237: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5238: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5239: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5240: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5241: invalidvarcomb[j1]=1;
1.226 brouard 5242: }else{
1.251 brouard 5243: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
5244: invalidvarcomb[j1]=0;
1.226 brouard 5245: }
1.251 brouard 5246: fprintf(ficresphtmfr,"</table>\n");
5247: fprintf(ficlog,"\n");
5248: if(j!=0){
5249: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5250: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5251: for(k=1; k <=(nlstate+ndeath); k++){
5252: if (k != i) {
1.265 brouard 5253: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5254: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5255: if(j1==1){ /* All dummy covariates to zero */
5256: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5257: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5258: printf("%d%d ",i,k);
5259: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5260: 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]));
5261: 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]));
5262: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5263: }
1.253 brouard 5264: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5265: for(iage=iagemin; iage <= iagemax+3; iage++){
5266: x[iage]= (double)iage;
5267: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5268: /* 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 5269: }
1.268 brouard 5270: /* Some are not finite, but linreg will ignore these ages */
5271: no=0;
1.253 brouard 5272: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5273: pstart[s1]=b;
5274: pstart[s1-1]=a;
1.252 brouard 5275: }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 */
5276: 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]);
5277: 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 5278: 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 5279: printf("%d%d ",i,k);
5280: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5281: 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 5282: }else{ /* Other cases, like quantitative fixed or varying covariates */
5283: ;
5284: }
5285: /* printf("%12.7f )", param[i][jj][k]); */
5286: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5287: s1++;
1.251 brouard 5288: } /* end jj */
5289: } /* end k!= i */
5290: } /* end k */
1.265 brouard 5291: } /* end i, s1 */
1.251 brouard 5292: } /* end j !=0 */
5293: } /* end selected combination of covariate j1 */
5294: if(j==0){ /* We can estimate starting values from the occurences in each case */
5295: printf("#Freqsummary: Starting values for the constants:\n");
5296: fprintf(ficlog,"\n");
1.265 brouard 5297: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5298: for(k=1; k <=(nlstate+ndeath); k++){
5299: if (k != i) {
5300: printf("%d%d ",i,k);
5301: fprintf(ficlog,"%d%d ",i,k);
5302: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5303: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5304: if(jj==1){ /* Age has to be done */
1.265 brouard 5305: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5306: 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]));
5307: 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 5308: }
5309: /* printf("%12.7f )", param[i][jj][k]); */
5310: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5311: s1++;
1.250 brouard 5312: }
1.251 brouard 5313: printf("\n");
5314: fprintf(ficlog,"\n");
1.250 brouard 5315: }
5316: }
1.284 brouard 5317: } /* end of state i */
1.251 brouard 5318: printf("#Freqsummary\n");
5319: fprintf(ficlog,"\n");
1.265 brouard 5320: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5321: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5322: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5323: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5324: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5325: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5326: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5327: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5328: /* } */
5329: }
1.265 brouard 5330: } /* end loop s1 */
1.251 brouard 5331:
5332: printf("\n");
5333: fprintf(ficlog,"\n");
5334: } /* end j=0 */
1.249 brouard 5335: } /* end j */
1.252 brouard 5336:
1.253 brouard 5337: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5338: for(i=1, jk=1; i <=nlstate; i++){
5339: for(j=1; j <=nlstate+ndeath; j++){
5340: if(j!=i){
5341: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5342: printf("%1d%1d",i,j);
5343: fprintf(ficparo,"%1d%1d",i,j);
5344: for(k=1; k<=ncovmodel;k++){
5345: /* printf(" %lf",param[i][j][k]); */
5346: /* fprintf(ficparo," %lf",param[i][j][k]); */
5347: p[jk]=pstart[jk];
5348: printf(" %f ",pstart[jk]);
5349: fprintf(ficparo," %f ",pstart[jk]);
5350: jk++;
5351: }
5352: printf("\n");
5353: fprintf(ficparo,"\n");
5354: }
5355: }
5356: }
5357: } /* end mle=-2 */
1.226 brouard 5358: dateintmean=dateintsum/k2cpt;
1.296 brouard 5359: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5360:
1.226 brouard 5361: fclose(ficresp);
5362: fclose(ficresphtm);
5363: fclose(ficresphtmfr);
1.283 brouard 5364: free_vector(idq,1,nqfveff);
1.226 brouard 5365: free_vector(meanq,1,nqfveff);
1.284 brouard 5366: free_vector(stdq,1,nqfveff);
1.226 brouard 5367: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5368: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5369: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5370: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5371: free_vector(pospropt,1,nlstate);
5372: free_vector(posprop,1,nlstate);
1.251 brouard 5373: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5374: free_vector(pp,1,nlstate);
5375: /* End of freqsummary */
5376: }
1.126 brouard 5377:
1.268 brouard 5378: /* Simple linear regression */
5379: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5380:
5381: /* y=a+bx regression */
5382: double sumx = 0.0; /* sum of x */
5383: double sumx2 = 0.0; /* sum of x**2 */
5384: double sumxy = 0.0; /* sum of x * y */
5385: double sumy = 0.0; /* sum of y */
5386: double sumy2 = 0.0; /* sum of y**2 */
5387: double sume2 = 0.0; /* sum of square or residuals */
5388: double yhat;
5389:
5390: double denom=0;
5391: int i;
5392: int ne=*no;
5393:
5394: for ( i=ifi, ne=0;i<=ila;i++) {
5395: if(!isfinite(x[i]) || !isfinite(y[i])){
5396: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5397: continue;
5398: }
5399: ne=ne+1;
5400: sumx += x[i];
5401: sumx2 += x[i]*x[i];
5402: sumxy += x[i] * y[i];
5403: sumy += y[i];
5404: sumy2 += y[i]*y[i];
5405: denom = (ne * sumx2 - sumx*sumx);
5406: /* 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); */
5407: }
5408:
5409: denom = (ne * sumx2 - sumx*sumx);
5410: if (denom == 0) {
5411: // vertical, slope m is infinity
5412: *b = INFINITY;
5413: *a = 0;
5414: if (r) *r = 0;
5415: return 1;
5416: }
5417:
5418: *b = (ne * sumxy - sumx * sumy) / denom;
5419: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5420: if (r!=NULL) {
5421: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5422: sqrt((sumx2 - sumx*sumx/ne) *
5423: (sumy2 - sumy*sumy/ne));
5424: }
5425: *no=ne;
5426: for ( i=ifi, ne=0;i<=ila;i++) {
5427: if(!isfinite(x[i]) || !isfinite(y[i])){
5428: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5429: continue;
5430: }
5431: ne=ne+1;
5432: yhat = y[i] - *a -*b* x[i];
5433: sume2 += yhat * yhat ;
5434:
5435: denom = (ne * sumx2 - sumx*sumx);
5436: /* 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); */
5437: }
5438: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5439: *sa= *sb * sqrt(sumx2/ne);
5440:
5441: return 0;
5442: }
5443:
1.126 brouard 5444: /************ Prevalence ********************/
1.227 brouard 5445: 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)
5446: {
5447: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5448: in each health status at the date of interview (if between dateprev1 and dateprev2).
5449: We still use firstpass and lastpass as another selection.
5450: */
1.126 brouard 5451:
1.227 brouard 5452: int i, m, jk, j1, bool, z1,j, iv;
5453: int mi; /* Effective wave */
5454: int iage;
5455: double agebegin, ageend;
5456:
5457: double **prop;
5458: double posprop;
5459: double y2; /* in fractional years */
5460: int iagemin, iagemax;
5461: int first; /** to stop verbosity which is redirected to log file */
5462:
5463: iagemin= (int) agemin;
5464: iagemax= (int) agemax;
5465: /*pp=vector(1,nlstate);*/
1.251 brouard 5466: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5467: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5468: j1=0;
1.222 brouard 5469:
1.227 brouard 5470: /*j=cptcoveff;*/
5471: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5472:
1.288 brouard 5473: first=0;
1.227 brouard 5474: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5475: for (i=1; i<=nlstate; i++)
1.251 brouard 5476: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5477: prop[i][iage]=0.0;
5478: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5479: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5480: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5481:
5482: for (i=1; i<=imx; i++) { /* Each individual */
5483: bool=1;
5484: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5485: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5486: m=mw[mi][i];
5487: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5488: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5489: for (z1=1; z1<=cptcoveff; z1++){
5490: if( Fixed[Tmodelind[z1]]==1){
5491: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.330 ! brouard 5492: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]) /* iv=1 to ntv, right modality */
1.227 brouard 5493: bool=0;
5494: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.330 ! brouard 5495: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]) {
1.227 brouard 5496: bool=0;
5497: }
5498: }
5499: if(bool==1){ /* Otherwise we skip that wave/person */
5500: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5501: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5502: if(m >=firstpass && m <=lastpass){
5503: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5504: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5505: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5506: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5507: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5508: 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);
5509: exit(1);
5510: }
5511: if (s[m][i]>0 && s[m][i]<=nlstate) {
5512: /*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]]);*/
5513: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5514: prop[s[m][i]][iagemax+3] += weight[i];
5515: } /* end valid statuses */
5516: } /* end selection of dates */
5517: } /* end selection of waves */
5518: } /* end bool */
5519: } /* end wave */
5520: } /* end individual */
5521: for(i=iagemin; i <= iagemax+3; i++){
5522: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5523: posprop += prop[jk][i];
5524: }
5525:
5526: for(jk=1; jk <=nlstate ; jk++){
5527: if( i <= iagemax){
5528: if(posprop>=1.e-5){
5529: probs[i][jk][j1]= prop[jk][i]/posprop;
5530: } else{
1.288 brouard 5531: if(!first){
5532: first=1;
1.266 brouard 5533: 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]);
5534: }else{
1.288 brouard 5535: 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 5536: }
5537: }
5538: }
5539: }/* end jk */
5540: }/* end i */
1.222 brouard 5541: /*} *//* end i1 */
1.227 brouard 5542: } /* end j1 */
1.222 brouard 5543:
1.227 brouard 5544: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5545: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5546: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5547: } /* End of prevalence */
1.126 brouard 5548:
5549: /************* Waves Concatenation ***************/
5550:
5551: 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)
5552: {
1.298 brouard 5553: /* 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 5554: Death is a valid wave (if date is known).
5555: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5556: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5557: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5558: */
1.126 brouard 5559:
1.224 brouard 5560: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5561: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5562: double sum=0., jmean=0.;*/
1.224 brouard 5563: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5564: int j, k=0,jk, ju, jl;
5565: double sum=0.;
5566: first=0;
1.214 brouard 5567: firstwo=0;
1.217 brouard 5568: firsthree=0;
1.218 brouard 5569: firstfour=0;
1.164 brouard 5570: jmin=100000;
1.126 brouard 5571: jmax=-1;
5572: jmean=0.;
1.224 brouard 5573:
5574: /* Treating live states */
1.214 brouard 5575: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5576: mi=0; /* First valid wave */
1.227 brouard 5577: mli=0; /* Last valid wave */
1.309 brouard 5578: m=firstpass; /* Loop on waves */
5579: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5580: 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 */
5581: mli=m-1;/* mw[++mi][i]=m-1; */
5582: }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 5583: 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 5584: mli=m;
1.224 brouard 5585: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5586: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5587: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5588: }
1.309 brouard 5589: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5590: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5591: break;
1.224 brouard 5592: #else
1.317 brouard 5593: 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 5594: if(firsthree == 0){
1.302 brouard 5595: 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 5596: firsthree=1;
1.317 brouard 5597: }else if(firsthree >=1 && firsthree < 10){
5598: 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);
5599: firsthree++;
5600: }else if(firsthree == 10){
5601: printf("Information, too many Information flags: no more reported to log either\n");
5602: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5603: firsthree++;
5604: }else{
5605: firsthree++;
1.227 brouard 5606: }
1.309 brouard 5607: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5608: mli=m;
5609: }
5610: if(s[m][i]==-2){ /* Vital status is really unknown */
5611: nbwarn++;
1.309 brouard 5612: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5613: 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);
5614: 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);
5615: }
5616: break;
5617: }
5618: break;
1.224 brouard 5619: #endif
1.227 brouard 5620: }/* End m >= lastpass */
1.126 brouard 5621: }/* end while */
1.224 brouard 5622:
1.227 brouard 5623: /* 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 5624: /* After last pass */
1.224 brouard 5625: /* Treating death states */
1.214 brouard 5626: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5627: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5628: /* } */
1.126 brouard 5629: mi++; /* Death is another wave */
5630: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5631: /* Only death is a correct wave */
1.126 brouard 5632: mw[mi][i]=m;
1.257 brouard 5633: } /* else not in a death state */
1.224 brouard 5634: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5635: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5636: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5637: 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 5638: nbwarn++;
5639: if(firstfiv==0){
1.309 brouard 5640: 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 5641: firstfiv=1;
5642: }else{
1.309 brouard 5643: 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 5644: }
1.309 brouard 5645: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5646: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5647: nberr++;
5648: if(firstwo==0){
1.309 brouard 5649: 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 5650: firstwo=1;
5651: }
1.309 brouard 5652: 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 5653: }
1.257 brouard 5654: }else{ /* if date of interview is unknown */
1.227 brouard 5655: /* death is known but not confirmed by death status at any wave */
5656: if(firstfour==0){
1.309 brouard 5657: 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 5658: firstfour=1;
5659: }
1.309 brouard 5660: 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 5661: }
1.224 brouard 5662: } /* end if date of death is known */
5663: #endif
1.309 brouard 5664: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5665: /* wav[i]=mw[mi][i]; */
1.126 brouard 5666: if(mi==0){
5667: nbwarn++;
5668: if(first==0){
1.227 brouard 5669: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5670: first=1;
1.126 brouard 5671: }
5672: if(first==1){
1.227 brouard 5673: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5674: }
5675: } /* end mi==0 */
5676: } /* End individuals */
1.214 brouard 5677: /* wav and mw are no more changed */
1.223 brouard 5678:
1.317 brouard 5679: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
5680: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
5681:
5682:
1.126 brouard 5683: for(i=1; i<=imx; i++){
5684: for(mi=1; mi<wav[i];mi++){
5685: if (stepm <=0)
1.227 brouard 5686: dh[mi][i]=1;
1.126 brouard 5687: else{
1.260 brouard 5688: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5689: if (agedc[i] < 2*AGESUP) {
5690: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5691: if(j==0) j=1; /* Survives at least one month after exam */
5692: else if(j<0){
5693: nberr++;
5694: 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]);
5695: j=1; /* Temporary Dangerous patch */
5696: 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);
5697: 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]);
5698: 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);
5699: }
5700: k=k+1;
5701: if (j >= jmax){
5702: jmax=j;
5703: ijmax=i;
5704: }
5705: if (j <= jmin){
5706: jmin=j;
5707: ijmin=i;
5708: }
5709: sum=sum+j;
5710: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5711: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5712: }
5713: }
5714: else{
5715: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5716: /* 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 5717:
1.227 brouard 5718: k=k+1;
5719: if (j >= jmax) {
5720: jmax=j;
5721: ijmax=i;
5722: }
5723: else if (j <= jmin){
5724: jmin=j;
5725: ijmin=i;
5726: }
5727: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5728: /*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]);*/
5729: if(j<0){
5730: nberr++;
5731: 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]);
5732: 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]);
5733: }
5734: sum=sum+j;
5735: }
5736: jk= j/stepm;
5737: jl= j -jk*stepm;
5738: ju= j -(jk+1)*stepm;
5739: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5740: if(jl==0){
5741: dh[mi][i]=jk;
5742: bh[mi][i]=0;
5743: }else{ /* We want a negative bias in order to only have interpolation ie
5744: * to avoid the price of an extra matrix product in likelihood */
5745: dh[mi][i]=jk+1;
5746: bh[mi][i]=ju;
5747: }
5748: }else{
5749: if(jl <= -ju){
5750: dh[mi][i]=jk;
5751: bh[mi][i]=jl; /* bias is positive if real duration
5752: * is higher than the multiple of stepm and negative otherwise.
5753: */
5754: }
5755: else{
5756: dh[mi][i]=jk+1;
5757: bh[mi][i]=ju;
5758: }
5759: if(dh[mi][i]==0){
5760: dh[mi][i]=1; /* At least one step */
5761: bh[mi][i]=ju; /* At least one step */
5762: /* 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);*/
5763: }
5764: } /* end if mle */
1.126 brouard 5765: }
5766: } /* end wave */
5767: }
5768: jmean=sum/k;
5769: 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 5770: 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 5771: }
1.126 brouard 5772:
5773: /*********** Tricode ****************************/
1.220 brouard 5774: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5775: {
5776: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5777: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5778: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5779: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5780: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5781: */
1.130 brouard 5782:
1.242 brouard 5783: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5784: int modmaxcovj=0; /* Modality max of covariates j */
5785: int cptcode=0; /* Modality max of covariates j */
5786: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5787:
5788:
1.242 brouard 5789: /* cptcoveff=0; */
5790: /* *cptcov=0; */
1.126 brouard 5791:
1.242 brouard 5792: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5793: for (k=1; k <= maxncov; k++)
5794: for(j=1; j<=2; j++)
5795: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5796:
1.242 brouard 5797: /* Loop on covariates without age and products and no quantitative variable */
5798: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5799: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5800: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5801: switch(Fixed[k]) {
5802: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5803: modmaxcovj=0;
5804: modmincovj=0;
1.242 brouard 5805: 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*/
5806: ij=(int)(covar[Tvar[k]][i]);
5807: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5808: * If product of Vn*Vm, still boolean *:
5809: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5810: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5811: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5812: modality of the nth covariate of individual i. */
5813: if (ij > modmaxcovj)
5814: modmaxcovj=ij;
5815: else if (ij < modmincovj)
5816: modmincovj=ij;
1.287 brouard 5817: if (ij <0 || ij >1 ){
1.311 brouard 5818: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5819: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5820: fflush(ficlog);
5821: exit(1);
1.287 brouard 5822: }
5823: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5824: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5825: exit(1);
5826: }else
5827: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5828: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5829: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5830: /* getting the maximum value of the modality of the covariate
5831: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5832: female ies 1, then modmaxcovj=1.
5833: */
5834: } /* end for loop on individuals i */
5835: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5836: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5837: cptcode=modmaxcovj;
5838: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5839: /*for (i=0; i<=cptcode; i++) {*/
5840: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5841: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5842: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5843: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5844: if( j != -1){
5845: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5846: covariate for which somebody answered excluding
5847: undefined. Usually 2: 0 and 1. */
5848: }
5849: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5850: covariate for which somebody answered including
5851: undefined. Usually 3: -1, 0 and 1. */
5852: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5853: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5854: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5855:
1.242 brouard 5856: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5857: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5858: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5859: /* modmincovj=3; modmaxcovj = 7; */
5860: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5861: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5862: /* defining two dummy variables: variables V1_1 and V1_2.*/
5863: /* nbcode[Tvar[j]][ij]=k; */
5864: /* nbcode[Tvar[j]][1]=0; */
5865: /* nbcode[Tvar[j]][2]=1; */
5866: /* nbcode[Tvar[j]][3]=2; */
5867: /* To be continued (not working yet). */
5868: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5869:
5870: /* 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*/
5871: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5872: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5873: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5874: /*, could be restored in the future */
5875: 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 5876: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5877: break;
5878: }
5879: ij++;
1.287 brouard 5880: 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 5881: cptcode = ij; /* New max modality for covar j */
5882: } /* end of loop on modality i=-1 to 1 or more */
5883: break;
5884: case 1: /* Testing on varying covariate, could be simple and
5885: * should look at waves or product of fixed *
5886: * varying. No time to test -1, assuming 0 and 1 only */
5887: ij=0;
5888: for(i=0; i<=1;i++){
5889: nbcode[Tvar[k]][++ij]=i;
5890: }
5891: break;
5892: default:
5893: break;
5894: } /* end switch */
5895: } /* end dummy test */
1.311 brouard 5896: if(Dummy[k]==1 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5897: 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*/
5898: if(isnan(covar[Tvar[k]][i])){
5899: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5900: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5901: fflush(ficlog);
5902: exit(1);
5903: }
5904: }
5905: }
1.287 brouard 5906: } /* 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 5907:
5908: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5909: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5910: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5911: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5912: 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 */
5913: 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 */
5914: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5915: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5916:
5917: ij=0;
5918: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5919: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5920: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5921: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5922: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5923: /* If product not in single variable we don't print results */
5924: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5925: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5926: 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*/
5927: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5928: 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 */
5929: if(Fixed[k]!=0)
5930: anyvaryingduminmodel=1;
5931: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5932: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5933: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5934: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5935: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5936: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5937: }
5938: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5939: /* ij--; */
5940: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.330 ! brouard 5941: *cptcov=ij; /* cptcov= Number of total real effective covariates: effective (used as cptcoveff in other functions)
1.242 brouard 5942: * because they can be excluded from the model and real
5943: * if in the model but excluded because missing values, but how to get k from ij?*/
5944: for(j=ij+1; j<= cptcovt; j++){
5945: Tvaraff[j]=0;
5946: Tmodelind[j]=0;
5947: }
5948: for(j=ntveff+1; j<= cptcovt; j++){
5949: TmodelInvind[j]=0;
5950: }
5951: /* To be sorted */
5952: ;
5953: }
1.126 brouard 5954:
1.145 brouard 5955:
1.126 brouard 5956: /*********** Health Expectancies ****************/
5957:
1.235 brouard 5958: 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 5959:
5960: {
5961: /* Health expectancies, no variances */
1.329 brouard 5962: /* cij is the combination in the list of combination of dummy covariates */
5963: /* strstart is a string of time at start of computing */
1.164 brouard 5964: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5965: int nhstepma, nstepma; /* Decreasing with age */
5966: double age, agelim, hf;
5967: double ***p3mat;
5968: double eip;
5969:
1.238 brouard 5970: /* pstamp(ficreseij); */
1.126 brouard 5971: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5972: fprintf(ficreseij,"# Age");
5973: for(i=1; i<=nlstate;i++){
5974: for(j=1; j<=nlstate;j++){
5975: fprintf(ficreseij," e%1d%1d ",i,j);
5976: }
5977: fprintf(ficreseij," e%1d. ",i);
5978: }
5979: fprintf(ficreseij,"\n");
5980:
5981:
5982: if(estepm < stepm){
5983: printf ("Problem %d lower than %d\n",estepm, stepm);
5984: }
5985: else hstepm=estepm;
5986: /* We compute the life expectancy from trapezoids spaced every estepm months
5987: * This is mainly to measure the difference between two models: for example
5988: * if stepm=24 months pijx are given only every 2 years and by summing them
5989: * we are calculating an estimate of the Life Expectancy assuming a linear
5990: * progression in between and thus overestimating or underestimating according
5991: * to the curvature of the survival function. If, for the same date, we
5992: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5993: * to compare the new estimate of Life expectancy with the same linear
5994: * hypothesis. A more precise result, taking into account a more precise
5995: * curvature will be obtained if estepm is as small as stepm. */
5996:
5997: /* For example we decided to compute the life expectancy with the smallest unit */
5998: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5999: nhstepm is the number of hstepm from age to agelim
6000: nstepm is the number of stepm from age to agelin.
1.270 brouard 6001: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6002: and note for a fixed period like estepm months */
6003: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6004: survival function given by stepm (the optimization length). Unfortunately it
6005: means that if the survival funtion is printed only each two years of age and if
6006: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6007: results. So we changed our mind and took the option of the best precision.
6008: */
6009: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6010:
6011: agelim=AGESUP;
6012: /* If stepm=6 months */
6013: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6014: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6015:
6016: /* nhstepm age range expressed in number of stepm */
6017: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6018: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6019: /* if (stepm >= YEARM) hstepm=1;*/
6020: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6021: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6022:
6023: for (age=bage; age<=fage; age ++){
6024: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6025: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6026: /* if (stepm >= YEARM) hstepm=1;*/
6027: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6028:
6029: /* If stepm=6 months */
6030: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6031: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 ! brouard 6032: /* printf("HELLO evsij Entering hpxij age=%d cij=%d hstepm=%d x[1]=%f nres=%d\n",(int) age, cij, hstepm, x[1], nres); */
1.235 brouard 6033: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6034:
6035: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6036:
6037: printf("%d|",(int)age);fflush(stdout);
6038: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6039:
6040: /* Computing expectancies */
6041: for(i=1; i<=nlstate;i++)
6042: for(j=1; j<=nlstate;j++)
6043: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6044: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6045:
6046: /* 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]);*/
6047:
6048: }
6049:
6050: fprintf(ficreseij,"%3.0f",age );
6051: for(i=1; i<=nlstate;i++){
6052: eip=0;
6053: for(j=1; j<=nlstate;j++){
6054: eip +=eij[i][j][(int)age];
6055: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6056: }
6057: fprintf(ficreseij,"%9.4f", eip );
6058: }
6059: fprintf(ficreseij,"\n");
6060:
6061: }
6062: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6063: printf("\n");
6064: fprintf(ficlog,"\n");
6065:
6066: }
6067:
1.235 brouard 6068: 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 6069:
6070: {
6071: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6072: to initial status i, ei. .
1.126 brouard 6073: */
6074: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6075: int nhstepma, nstepma; /* Decreasing with age */
6076: double age, agelim, hf;
6077: double ***p3matp, ***p3matm, ***varhe;
6078: double **dnewm,**doldm;
6079: double *xp, *xm;
6080: double **gp, **gm;
6081: double ***gradg, ***trgradg;
6082: int theta;
6083:
6084: double eip, vip;
6085:
6086: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6087: xp=vector(1,npar);
6088: xm=vector(1,npar);
6089: dnewm=matrix(1,nlstate*nlstate,1,npar);
6090: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6091:
6092: pstamp(ficresstdeij);
6093: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6094: fprintf(ficresstdeij,"# Age");
6095: for(i=1; i<=nlstate;i++){
6096: for(j=1; j<=nlstate;j++)
6097: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6098: fprintf(ficresstdeij," e%1d. ",i);
6099: }
6100: fprintf(ficresstdeij,"\n");
6101:
6102: pstamp(ficrescveij);
6103: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6104: fprintf(ficrescveij,"# Age");
6105: for(i=1; i<=nlstate;i++)
6106: for(j=1; j<=nlstate;j++){
6107: cptj= (j-1)*nlstate+i;
6108: for(i2=1; i2<=nlstate;i2++)
6109: for(j2=1; j2<=nlstate;j2++){
6110: cptj2= (j2-1)*nlstate+i2;
6111: if(cptj2 <= cptj)
6112: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6113: }
6114: }
6115: fprintf(ficrescveij,"\n");
6116:
6117: if(estepm < stepm){
6118: printf ("Problem %d lower than %d\n",estepm, stepm);
6119: }
6120: else hstepm=estepm;
6121: /* We compute the life expectancy from trapezoids spaced every estepm months
6122: * This is mainly to measure the difference between two models: for example
6123: * if stepm=24 months pijx are given only every 2 years and by summing them
6124: * we are calculating an estimate of the Life Expectancy assuming a linear
6125: * progression in between and thus overestimating or underestimating according
6126: * to the curvature of the survival function. If, for the same date, we
6127: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6128: * to compare the new estimate of Life expectancy with the same linear
6129: * hypothesis. A more precise result, taking into account a more precise
6130: * curvature will be obtained if estepm is as small as stepm. */
6131:
6132: /* For example we decided to compute the life expectancy with the smallest unit */
6133: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6134: nhstepm is the number of hstepm from age to agelim
6135: nstepm is the number of stepm from age to agelin.
6136: Look at hpijx to understand the reason of that which relies in memory size
6137: and note for a fixed period like estepm months */
6138: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6139: survival function given by stepm (the optimization length). Unfortunately it
6140: means that if the survival funtion is printed only each two years of age and if
6141: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6142: results. So we changed our mind and took the option of the best precision.
6143: */
6144: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6145:
6146: /* If stepm=6 months */
6147: /* nhstepm age range expressed in number of stepm */
6148: agelim=AGESUP;
6149: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6150: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6151: /* if (stepm >= YEARM) hstepm=1;*/
6152: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6153:
6154: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6155: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6156: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6157: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6158: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6159: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6160:
6161: for (age=bage; age<=fage; age ++){
6162: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6163: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6164: /* if (stepm >= YEARM) hstepm=1;*/
6165: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6166:
1.126 brouard 6167: /* If stepm=6 months */
6168: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6169: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6170:
6171: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6172:
1.126 brouard 6173: /* Computing Variances of health expectancies */
6174: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6175: decrease memory allocation */
6176: for(theta=1; theta <=npar; theta++){
6177: for(i=1; i<=npar; i++){
1.222 brouard 6178: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6179: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6180: }
1.235 brouard 6181: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6182: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6183:
1.126 brouard 6184: for(j=1; j<= nlstate; j++){
1.222 brouard 6185: for(i=1; i<=nlstate; i++){
6186: for(h=0; h<=nhstepm-1; h++){
6187: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6188: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6189: }
6190: }
1.126 brouard 6191: }
1.218 brouard 6192:
1.126 brouard 6193: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6194: for(h=0; h<=nhstepm-1; h++){
6195: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6196: }
1.126 brouard 6197: }/* End theta */
6198:
6199:
6200: for(h=0; h<=nhstepm-1; h++)
6201: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6202: for(theta=1; theta <=npar; theta++)
6203: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6204:
1.218 brouard 6205:
1.222 brouard 6206: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6207: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6208: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6209:
1.222 brouard 6210: printf("%d|",(int)age);fflush(stdout);
6211: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6212: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6213: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6214: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6215: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6216: for(ij=1;ij<=nlstate*nlstate;ij++)
6217: for(ji=1;ji<=nlstate*nlstate;ji++)
6218: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6219: }
6220: }
1.320 brouard 6221: /* if((int)age ==50){ */
6222: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6223: /* } */
1.126 brouard 6224: /* Computing expectancies */
1.235 brouard 6225: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6226: for(i=1; i<=nlstate;i++)
6227: for(j=1; j<=nlstate;j++)
1.222 brouard 6228: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6229: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6230:
1.222 brouard 6231: /* 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 6232:
1.222 brouard 6233: }
1.269 brouard 6234:
6235: /* Standard deviation of expectancies ij */
1.126 brouard 6236: fprintf(ficresstdeij,"%3.0f",age );
6237: for(i=1; i<=nlstate;i++){
6238: eip=0.;
6239: vip=0.;
6240: for(j=1; j<=nlstate;j++){
1.222 brouard 6241: eip += eij[i][j][(int)age];
6242: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6243: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6244: 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 6245: }
6246: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6247: }
6248: fprintf(ficresstdeij,"\n");
1.218 brouard 6249:
1.269 brouard 6250: /* Variance of expectancies ij */
1.126 brouard 6251: fprintf(ficrescveij,"%3.0f",age );
6252: for(i=1; i<=nlstate;i++)
6253: for(j=1; j<=nlstate;j++){
1.222 brouard 6254: cptj= (j-1)*nlstate+i;
6255: for(i2=1; i2<=nlstate;i2++)
6256: for(j2=1; j2<=nlstate;j2++){
6257: cptj2= (j2-1)*nlstate+i2;
6258: if(cptj2 <= cptj)
6259: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6260: }
1.126 brouard 6261: }
6262: fprintf(ficrescveij,"\n");
1.218 brouard 6263:
1.126 brouard 6264: }
6265: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6266: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6267: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6268: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6269: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6270: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6271: printf("\n");
6272: fprintf(ficlog,"\n");
1.218 brouard 6273:
1.126 brouard 6274: free_vector(xm,1,npar);
6275: free_vector(xp,1,npar);
6276: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6277: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6278: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6279: }
1.218 brouard 6280:
1.126 brouard 6281: /************ Variance ******************/
1.235 brouard 6282: 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 6283: {
1.279 brouard 6284: /** Variance of health expectancies
6285: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6286: * double **newm;
6287: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6288: */
1.218 brouard 6289:
6290: /* int movingaverage(); */
6291: double **dnewm,**doldm;
6292: double **dnewmp,**doldmp;
6293: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6294: int first=0;
1.218 brouard 6295: int k;
6296: double *xp;
1.279 brouard 6297: double **gp, **gm; /**< for var eij */
6298: double ***gradg, ***trgradg; /**< for var eij */
6299: double **gradgp, **trgradgp; /**< for var p point j */
6300: double *gpp, *gmp; /**< for var p point j */
6301: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6302: double ***p3mat;
6303: double age,agelim, hf;
6304: /* double ***mobaverage; */
6305: int theta;
6306: char digit[4];
6307: char digitp[25];
6308:
6309: char fileresprobmorprev[FILENAMELENGTH];
6310:
6311: if(popbased==1){
6312: if(mobilav!=0)
6313: strcpy(digitp,"-POPULBASED-MOBILAV_");
6314: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6315: }
6316: else
6317: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6318:
1.218 brouard 6319: /* if (mobilav!=0) { */
6320: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6321: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6322: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6323: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6324: /* } */
6325: /* } */
6326:
6327: strcpy(fileresprobmorprev,"PRMORPREV-");
6328: sprintf(digit,"%-d",ij);
6329: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6330: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6331: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6332: strcat(fileresprobmorprev,fileresu);
6333: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6334: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6335: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6336: }
6337: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6338: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6339: pstamp(ficresprobmorprev);
6340: 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 6341: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
6342: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
6343: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
6344: }
6345: for(j=1;j<=cptcoveff;j++)
1.330 ! brouard 6346: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,Tvaraff[j])]);
1.238 brouard 6347: fprintf(ficresprobmorprev,"\n");
6348:
1.218 brouard 6349: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6350: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6351: fprintf(ficresprobmorprev," p.%-d SE",j);
6352: for(i=1; i<=nlstate;i++)
6353: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6354: }
6355: fprintf(ficresprobmorprev,"\n");
6356:
6357: fprintf(ficgp,"\n# Routine varevsij");
6358: fprintf(ficgp,"\nunset title \n");
6359: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6360: 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");
6361: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6362:
1.218 brouard 6363: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6364: pstamp(ficresvij);
6365: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6366: if(popbased==1)
6367: 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);
6368: else
6369: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6370: fprintf(ficresvij,"# Age");
6371: for(i=1; i<=nlstate;i++)
6372: for(j=1; j<=nlstate;j++)
6373: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6374: fprintf(ficresvij,"\n");
6375:
6376: xp=vector(1,npar);
6377: dnewm=matrix(1,nlstate,1,npar);
6378: doldm=matrix(1,nlstate,1,nlstate);
6379: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6380: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6381:
6382: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6383: gpp=vector(nlstate+1,nlstate+ndeath);
6384: gmp=vector(nlstate+1,nlstate+ndeath);
6385: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6386:
1.218 brouard 6387: if(estepm < stepm){
6388: printf ("Problem %d lower than %d\n",estepm, stepm);
6389: }
6390: else hstepm=estepm;
6391: /* For example we decided to compute the life expectancy with the smallest unit */
6392: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6393: nhstepm is the number of hstepm from age to agelim
6394: nstepm is the number of stepm from age to agelim.
6395: Look at function hpijx to understand why because of memory size limitations,
6396: we decided (b) to get a life expectancy respecting the most precise curvature of the
6397: survival function given by stepm (the optimization length). Unfortunately it
6398: means that if the survival funtion is printed every two years of age and if
6399: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6400: results. So we changed our mind and took the option of the best precision.
6401: */
6402: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6403: agelim = AGESUP;
6404: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6405: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6406: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6407: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6408: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6409: gp=matrix(0,nhstepm,1,nlstate);
6410: gm=matrix(0,nhstepm,1,nlstate);
6411:
6412:
6413: for(theta=1; theta <=npar; theta++){
6414: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6415: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6416: }
1.279 brouard 6417: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6418: * returns into prlim .
1.288 brouard 6419: */
1.242 brouard 6420: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6421:
6422: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6423: if (popbased==1) {
6424: if(mobilav ==0){
6425: for(i=1; i<=nlstate;i++)
6426: prlim[i][i]=probs[(int)age][i][ij];
6427: }else{ /* mobilav */
6428: for(i=1; i<=nlstate;i++)
6429: prlim[i][i]=mobaverage[(int)age][i][ij];
6430: }
6431: }
1.295 brouard 6432: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6433: */
6434: 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 6435: /**< 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 6436: * at horizon h in state j including mortality.
6437: */
1.218 brouard 6438: for(j=1; j<= nlstate; j++){
6439: for(h=0; h<=nhstepm; h++){
6440: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6441: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6442: }
6443: }
1.279 brouard 6444: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6445: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6446: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6447: */
6448: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6449: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6450: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6451: }
6452:
6453: /* Again with minus shift */
1.218 brouard 6454:
6455: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6456: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6457:
1.242 brouard 6458: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6459:
6460: if (popbased==1) {
6461: if(mobilav ==0){
6462: for(i=1; i<=nlstate;i++)
6463: prlim[i][i]=probs[(int)age][i][ij];
6464: }else{ /* mobilav */
6465: for(i=1; i<=nlstate;i++)
6466: prlim[i][i]=mobaverage[(int)age][i][ij];
6467: }
6468: }
6469:
1.235 brouard 6470: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6471:
6472: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6473: for(h=0; h<=nhstepm; h++){
6474: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6475: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6476: }
6477: }
6478: /* This for computing probability of death (h=1 means
6479: computed over hstepm matrices product = hstepm*stepm months)
6480: as a weighted average of prlim.
6481: */
6482: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6483: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6484: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6485: }
1.279 brouard 6486: /* end shifting computations */
6487:
6488: /**< Computing gradient matrix at horizon h
6489: */
1.218 brouard 6490: for(j=1; j<= nlstate; j++) /* vareij */
6491: for(h=0; h<=nhstepm; h++){
6492: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6493: }
1.279 brouard 6494: /**< Gradient of overall mortality p.3 (or p.j)
6495: */
6496: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6497: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6498: }
6499:
6500: } /* End theta */
1.279 brouard 6501:
6502: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6503: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6504:
6505: for(h=0; h<=nhstepm; h++) /* veij */
6506: for(j=1; j<=nlstate;j++)
6507: for(theta=1; theta <=npar; theta++)
6508: trgradg[h][j][theta]=gradg[h][theta][j];
6509:
6510: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6511: for(theta=1; theta <=npar; theta++)
6512: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6513: /**< as well as its transposed matrix
6514: */
1.218 brouard 6515:
6516: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6517: for(i=1;i<=nlstate;i++)
6518: for(j=1;j<=nlstate;j++)
6519: vareij[i][j][(int)age] =0.;
1.279 brouard 6520:
6521: /* Computing trgradg by matcov by gradg at age and summing over h
6522: * and k (nhstepm) formula 15 of article
6523: * Lievre-Brouard-Heathcote
6524: */
6525:
1.218 brouard 6526: for(h=0;h<=nhstepm;h++){
6527: for(k=0;k<=nhstepm;k++){
6528: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6529: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6530: for(i=1;i<=nlstate;i++)
6531: for(j=1;j<=nlstate;j++)
6532: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6533: }
6534: }
6535:
1.279 brouard 6536: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6537: * p.j overall mortality formula 49 but computed directly because
6538: * we compute the grad (wix pijx) instead of grad (pijx),even if
6539: * wix is independent of theta.
6540: */
1.218 brouard 6541: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6542: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6543: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6544: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6545: varppt[j][i]=doldmp[j][i];
6546: /* end ppptj */
6547: /* x centered again */
6548:
1.242 brouard 6549: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6550:
6551: if (popbased==1) {
6552: if(mobilav ==0){
6553: for(i=1; i<=nlstate;i++)
6554: prlim[i][i]=probs[(int)age][i][ij];
6555: }else{ /* mobilav */
6556: for(i=1; i<=nlstate;i++)
6557: prlim[i][i]=mobaverage[(int)age][i][ij];
6558: }
6559: }
6560:
6561: /* This for computing probability of death (h=1 means
6562: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6563: as a weighted average of prlim.
6564: */
1.235 brouard 6565: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6566: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6567: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6568: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6569: }
6570: /* end probability of death */
6571:
6572: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6573: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6574: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6575: for(i=1; i<=nlstate;i++){
6576: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6577: }
6578: }
6579: fprintf(ficresprobmorprev,"\n");
6580:
6581: fprintf(ficresvij,"%.0f ",age );
6582: for(i=1; i<=nlstate;i++)
6583: for(j=1; j<=nlstate;j++){
6584: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6585: }
6586: fprintf(ficresvij,"\n");
6587: free_matrix(gp,0,nhstepm,1,nlstate);
6588: free_matrix(gm,0,nhstepm,1,nlstate);
6589: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6590: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6591: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6592: } /* End age */
6593: free_vector(gpp,nlstate+1,nlstate+ndeath);
6594: free_vector(gmp,nlstate+1,nlstate+ndeath);
6595: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6596: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6597: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6598: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6599: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6600: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6601: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6602: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6603: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6604: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6605: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6606: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6607: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6608: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6609: 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);
6610: /* 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 6611: */
1.218 brouard 6612: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6613: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6614:
1.218 brouard 6615: free_vector(xp,1,npar);
6616: free_matrix(doldm,1,nlstate,1,nlstate);
6617: free_matrix(dnewm,1,nlstate,1,npar);
6618: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6619: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6620: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6621: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6622: fclose(ficresprobmorprev);
6623: fflush(ficgp);
6624: fflush(fichtm);
6625: } /* end varevsij */
1.126 brouard 6626:
6627: /************ Variance of prevlim ******************/
1.269 brouard 6628: 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 6629: {
1.205 brouard 6630: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6631: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6632:
1.268 brouard 6633: double **dnewmpar,**doldm;
1.126 brouard 6634: int i, j, nhstepm, hstepm;
6635: double *xp;
6636: double *gp, *gm;
6637: double **gradg, **trgradg;
1.208 brouard 6638: double **mgm, **mgp;
1.126 brouard 6639: double age,agelim;
6640: int theta;
6641:
6642: pstamp(ficresvpl);
1.288 brouard 6643: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6644: fprintf(ficresvpl,"# Age ");
6645: if(nresult >=1)
6646: fprintf(ficresvpl," Result# ");
1.126 brouard 6647: for(i=1; i<=nlstate;i++)
6648: fprintf(ficresvpl," %1d-%1d",i,i);
6649: fprintf(ficresvpl,"\n");
6650:
6651: xp=vector(1,npar);
1.268 brouard 6652: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6653: doldm=matrix(1,nlstate,1,nlstate);
6654:
6655: hstepm=1*YEARM; /* Every year of age */
6656: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6657: agelim = AGESUP;
6658: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6659: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6660: if (stepm >= YEARM) hstepm=1;
6661: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6662: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6663: mgp=matrix(1,npar,1,nlstate);
6664: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6665: gp=vector(1,nlstate);
6666: gm=vector(1,nlstate);
6667:
6668: for(theta=1; theta <=npar; theta++){
6669: for(i=1; i<=npar; i++){ /* Computes gradient */
6670: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6671: }
1.288 brouard 6672: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6673: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6674: /* else */
6675: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6676: for(i=1;i<=nlstate;i++){
1.126 brouard 6677: gp[i] = prlim[i][i];
1.208 brouard 6678: mgp[theta][i] = prlim[i][i];
6679: }
1.126 brouard 6680: for(i=1; i<=npar; i++) /* Computes gradient */
6681: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6682: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6683: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6684: /* else */
6685: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6686: for(i=1;i<=nlstate;i++){
1.126 brouard 6687: gm[i] = prlim[i][i];
1.208 brouard 6688: mgm[theta][i] = prlim[i][i];
6689: }
1.126 brouard 6690: for(i=1;i<=nlstate;i++)
6691: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6692: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6693: } /* End theta */
6694:
6695: trgradg =matrix(1,nlstate,1,npar);
6696:
6697: for(j=1; j<=nlstate;j++)
6698: for(theta=1; theta <=npar; theta++)
6699: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6700: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6701: /* printf("\nmgm mgp %d ",(int)age); */
6702: /* for(j=1; j<=nlstate;j++){ */
6703: /* printf(" %d ",j); */
6704: /* for(theta=1; theta <=npar; theta++) */
6705: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6706: /* printf("\n "); */
6707: /* } */
6708: /* } */
6709: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6710: /* printf("\n gradg %d ",(int)age); */
6711: /* for(j=1; j<=nlstate;j++){ */
6712: /* printf("%d ",j); */
6713: /* for(theta=1; theta <=npar; theta++) */
6714: /* printf("%d %lf ",theta,gradg[theta][j]); */
6715: /* printf("\n "); */
6716: /* } */
6717: /* } */
1.126 brouard 6718:
6719: for(i=1;i<=nlstate;i++)
6720: varpl[i][(int)age] =0.;
1.209 brouard 6721: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6722: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6723: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6724: }else{
1.268 brouard 6725: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6726: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6727: }
1.126 brouard 6728: for(i=1;i<=nlstate;i++)
6729: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6730:
6731: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6732: if(nresult >=1)
6733: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6734: for(i=1; i<=nlstate;i++){
1.126 brouard 6735: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6736: /* for(j=1;j<=nlstate;j++) */
6737: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6738: }
1.126 brouard 6739: fprintf(ficresvpl,"\n");
6740: free_vector(gp,1,nlstate);
6741: free_vector(gm,1,nlstate);
1.208 brouard 6742: free_matrix(mgm,1,npar,1,nlstate);
6743: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6744: free_matrix(gradg,1,npar,1,nlstate);
6745: free_matrix(trgradg,1,nlstate,1,npar);
6746: } /* End age */
6747:
6748: free_vector(xp,1,npar);
6749: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6750: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6751:
6752: }
6753:
6754:
6755: /************ Variance of backprevalence limit ******************/
1.269 brouard 6756: 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 6757: {
6758: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6759: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6760:
6761: double **dnewmpar,**doldm;
6762: int i, j, nhstepm, hstepm;
6763: double *xp;
6764: double *gp, *gm;
6765: double **gradg, **trgradg;
6766: double **mgm, **mgp;
6767: double age,agelim;
6768: int theta;
6769:
6770: pstamp(ficresvbl);
6771: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6772: fprintf(ficresvbl,"# Age ");
6773: if(nresult >=1)
6774: fprintf(ficresvbl," Result# ");
6775: for(i=1; i<=nlstate;i++)
6776: fprintf(ficresvbl," %1d-%1d",i,i);
6777: fprintf(ficresvbl,"\n");
6778:
6779: xp=vector(1,npar);
6780: dnewmpar=matrix(1,nlstate,1,npar);
6781: doldm=matrix(1,nlstate,1,nlstate);
6782:
6783: hstepm=1*YEARM; /* Every year of age */
6784: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6785: agelim = AGEINF;
6786: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6787: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6788: if (stepm >= YEARM) hstepm=1;
6789: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6790: gradg=matrix(1,npar,1,nlstate);
6791: mgp=matrix(1,npar,1,nlstate);
6792: mgm=matrix(1,npar,1,nlstate);
6793: gp=vector(1,nlstate);
6794: gm=vector(1,nlstate);
6795:
6796: for(theta=1; theta <=npar; theta++){
6797: for(i=1; i<=npar; i++){ /* Computes gradient */
6798: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6799: }
6800: if(mobilavproj > 0 )
6801: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6802: else
6803: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6804: for(i=1;i<=nlstate;i++){
6805: gp[i] = bprlim[i][i];
6806: mgp[theta][i] = bprlim[i][i];
6807: }
6808: for(i=1; i<=npar; i++) /* Computes gradient */
6809: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6810: if(mobilavproj > 0 )
6811: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6812: else
6813: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6814: for(i=1;i<=nlstate;i++){
6815: gm[i] = bprlim[i][i];
6816: mgm[theta][i] = bprlim[i][i];
6817: }
6818: for(i=1;i<=nlstate;i++)
6819: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6820: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6821: } /* End theta */
6822:
6823: trgradg =matrix(1,nlstate,1,npar);
6824:
6825: for(j=1; j<=nlstate;j++)
6826: for(theta=1; theta <=npar; theta++)
6827: trgradg[j][theta]=gradg[theta][j];
6828: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6829: /* printf("\nmgm mgp %d ",(int)age); */
6830: /* for(j=1; j<=nlstate;j++){ */
6831: /* printf(" %d ",j); */
6832: /* for(theta=1; theta <=npar; theta++) */
6833: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6834: /* printf("\n "); */
6835: /* } */
6836: /* } */
6837: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6838: /* printf("\n gradg %d ",(int)age); */
6839: /* for(j=1; j<=nlstate;j++){ */
6840: /* printf("%d ",j); */
6841: /* for(theta=1; theta <=npar; theta++) */
6842: /* printf("%d %lf ",theta,gradg[theta][j]); */
6843: /* printf("\n "); */
6844: /* } */
6845: /* } */
6846:
6847: for(i=1;i<=nlstate;i++)
6848: varbpl[i][(int)age] =0.;
6849: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6850: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6851: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6852: }else{
6853: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6854: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6855: }
6856: for(i=1;i<=nlstate;i++)
6857: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6858:
6859: fprintf(ficresvbl,"%.0f ",age );
6860: if(nresult >=1)
6861: fprintf(ficresvbl,"%d ",nres );
6862: for(i=1; i<=nlstate;i++)
6863: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6864: fprintf(ficresvbl,"\n");
6865: free_vector(gp,1,nlstate);
6866: free_vector(gm,1,nlstate);
6867: free_matrix(mgm,1,npar,1,nlstate);
6868: free_matrix(mgp,1,npar,1,nlstate);
6869: free_matrix(gradg,1,npar,1,nlstate);
6870: free_matrix(trgradg,1,nlstate,1,npar);
6871: } /* End age */
6872:
6873: free_vector(xp,1,npar);
6874: free_matrix(doldm,1,nlstate,1,npar);
6875: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6876:
6877: }
6878:
6879: /************ Variance of one-step probabilities ******************/
6880: 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 6881: {
6882: int i, j=0, k1, l1, tj;
6883: int k2, l2, j1, z1;
6884: int k=0, l;
6885: int first=1, first1, first2;
1.326 brouard 6886: int nres=0; /* New */
1.222 brouard 6887: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6888: double **dnewm,**doldm;
6889: double *xp;
6890: double *gp, *gm;
6891: double **gradg, **trgradg;
6892: double **mu;
6893: double age, cov[NCOVMAX+1];
6894: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6895: int theta;
6896: char fileresprob[FILENAMELENGTH];
6897: char fileresprobcov[FILENAMELENGTH];
6898: char fileresprobcor[FILENAMELENGTH];
6899: double ***varpij;
6900:
6901: strcpy(fileresprob,"PROB_");
6902: strcat(fileresprob,fileres);
6903: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6904: printf("Problem with resultfile: %s\n", fileresprob);
6905: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6906: }
6907: strcpy(fileresprobcov,"PROBCOV_");
6908: strcat(fileresprobcov,fileresu);
6909: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6910: printf("Problem with resultfile: %s\n", fileresprobcov);
6911: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6912: }
6913: strcpy(fileresprobcor,"PROBCOR_");
6914: strcat(fileresprobcor,fileresu);
6915: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6916: printf("Problem with resultfile: %s\n", fileresprobcor);
6917: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6918: }
6919: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6920: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6921: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6922: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6923: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6924: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6925: pstamp(ficresprob);
6926: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6927: fprintf(ficresprob,"# Age");
6928: pstamp(ficresprobcov);
6929: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6930: fprintf(ficresprobcov,"# Age");
6931: pstamp(ficresprobcor);
6932: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6933: fprintf(ficresprobcor,"# Age");
1.126 brouard 6934:
6935:
1.222 brouard 6936: for(i=1; i<=nlstate;i++)
6937: for(j=1; j<=(nlstate+ndeath);j++){
6938: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6939: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6940: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6941: }
6942: /* fprintf(ficresprob,"\n");
6943: fprintf(ficresprobcov,"\n");
6944: fprintf(ficresprobcor,"\n");
6945: */
6946: xp=vector(1,npar);
6947: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6948: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6949: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6950: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6951: first=1;
6952: fprintf(ficgp,"\n# Routine varprob");
6953: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6954: fprintf(fichtm,"\n");
6955:
1.288 brouard 6956: 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 6957: 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);
6958: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6959: and drawn. It helps understanding how is the covariance between two incidences.\
6960: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6961: 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 6962: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6963: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6964: standard deviations wide on each axis. <br>\
6965: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6966: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6967: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6968:
1.222 brouard 6969: cov[1]=1;
6970: /* tj=cptcoveff; */
1.225 brouard 6971: tj = (int) pow(2,cptcoveff);
1.222 brouard 6972: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6973: j1=0;
1.224 brouard 6974: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.326 brouard 6975: for(nres=1;nres <=1; nres++){ /* For each resultline */
6976: /* for(nres=1;nres <=nresult; nres++){ /\* For each resultline *\/ */
1.222 brouard 6977: if (cptcovn>0) {
6978: fprintf(ficresprob, "\n#********** Variable ");
1.330 ! brouard 6979: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]);
1.222 brouard 6980: fprintf(ficresprob, "**********\n#\n");
6981: fprintf(ficresprobcov, "\n#********** Variable ");
1.330 ! brouard 6982: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]);
1.222 brouard 6983: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6984:
1.222 brouard 6985: fprintf(ficgp, "\n#********** Variable ");
1.330 ! brouard 6986: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]);
1.222 brouard 6987: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6988:
6989:
1.222 brouard 6990: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.319 brouard 6991: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]); */
1.330 ! brouard 6992: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtmcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]);
1.222 brouard 6993: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6994:
1.222 brouard 6995: fprintf(ficresprobcor, "\n#********** Variable ");
1.330 ! brouard 6996: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,Tvaraff[z1])]);
1.222 brouard 6997: fprintf(ficresprobcor, "**********\n#");
6998: if(invalidvarcomb[j1]){
6999: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7000: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7001: continue;
7002: }
7003: }
7004: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7005: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7006: gp=vector(1,(nlstate)*(nlstate+ndeath));
7007: gm=vector(1,(nlstate)*(nlstate+ndeath));
7008: for (age=bage; age<=fage; age ++){
7009: cov[2]=age;
7010: if(nagesqr==1)
7011: cov[3]= age*age;
1.326 brouard 7012: /* for (k=1; k<=cptcovn;k++) { */
7013: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; */
7014: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
7015: /* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates */
1.330 ! brouard 7016: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TvarsD[k])];
1.222 brouard 7017: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
7018: * 1 1 1 1 1
7019: * 2 2 1 1 1
7020: * 3 1 2 1 1
7021: */
7022: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
7023: }
1.319 brouard 7024: /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 */
7025: /* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] */
7026: /*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
1.326 brouard 7027: for (k=1; k<=cptcovage;k++){ /* For product with age */
7028: if(Dummy[Tage[k]]==2){ /* dummy with age */
1.330 ! brouard 7029: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,Tvar[Tage[k]])]*cov[2];
1.326 brouard 7030: /* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
7031: } else if(Dummy[Tage[k]]==3){ /* quantitative with age */
1.327 brouard 7032: printf("Internal IMaCh error, don't know which value for quantitative covariate with age, Tage[k]%d, k=%d, Tvar[Tage[k]]=V%d, age=%d\n",Tage[k],k ,Tvar[Tage[k]], (int)cov[2]);
7033: exit(1);
7034: /* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\* Using the mean of quantitative variable Tvar[Tage[k]] /\* Tqresult[nres][k]; *\/ */
1.326 brouard 7035: /* cov[++k1]=Tqresult[nres][k]; */
7036: }
7037: /* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
7038: }
7039: for (k=1; k<=cptcovprod;k++){/* For product without age */
1.329 brouard 7040: if(Dummy[Tvard[k][1]]==0){
7041: if(Dummy[Tvard[k][2]]==0){
1.330 ! brouard 7042: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(j1,Tvard[k][2])];
1.326 brouard 7043: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
7044: }else{ /* Should we use the mean of the quantitative variables? */
1.330 ! brouard 7045: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,Tvard[k][1])] * Tqresult[nres][k];
1.326 brouard 7046: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; */
7047: }
7048: }else{
1.329 brouard 7049: if(Dummy[Tvard[k][2]]==0){
1.330 ! brouard 7050: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]];
1.326 brouard 7051: /* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; */
7052: }else{
7053: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
7054: /* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
7055: }
7056: }
7057: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
7058: }
7059: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7060: for(theta=1; theta <=npar; theta++){
7061: for(i=1; i<=npar; i++)
7062: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7063:
1.222 brouard 7064: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7065:
1.222 brouard 7066: k=0;
7067: for(i=1; i<= (nlstate); i++){
7068: for(j=1; j<=(nlstate+ndeath);j++){
7069: k=k+1;
7070: gp[k]=pmmij[i][j];
7071: }
7072: }
1.220 brouard 7073:
1.222 brouard 7074: for(i=1; i<=npar; i++)
7075: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7076:
1.222 brouard 7077: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7078: k=0;
7079: for(i=1; i<=(nlstate); i++){
7080: for(j=1; j<=(nlstate+ndeath);j++){
7081: k=k+1;
7082: gm[k]=pmmij[i][j];
7083: }
7084: }
1.220 brouard 7085:
1.222 brouard 7086: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7087: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7088: }
1.126 brouard 7089:
1.222 brouard 7090: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7091: for(theta=1; theta <=npar; theta++)
7092: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7093:
1.222 brouard 7094: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7095: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7096:
1.222 brouard 7097: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7098:
1.222 brouard 7099: k=0;
7100: for(i=1; i<=(nlstate); i++){
7101: for(j=1; j<=(nlstate+ndeath);j++){
7102: k=k+1;
7103: mu[k][(int) age]=pmmij[i][j];
7104: }
7105: }
7106: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7107: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7108: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7109:
1.222 brouard 7110: /*printf("\n%d ",(int)age);
7111: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7112: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7113: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7114: }*/
1.220 brouard 7115:
1.222 brouard 7116: fprintf(ficresprob,"\n%d ",(int)age);
7117: fprintf(ficresprobcov,"\n%d ",(int)age);
7118: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7119:
1.222 brouard 7120: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7121: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7122: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7123: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7124: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7125: }
7126: i=0;
7127: for (k=1; k<=(nlstate);k++){
7128: for (l=1; l<=(nlstate+ndeath);l++){
7129: i++;
7130: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7131: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7132: for (j=1; j<=i;j++){
7133: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7134: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7135: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7136: }
7137: }
7138: }/* end of loop for state */
7139: } /* end of loop for age */
7140: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7141: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7142: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7143: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7144:
7145: /* Confidence intervalle of pij */
7146: /*
7147: fprintf(ficgp,"\nunset parametric;unset label");
7148: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7149: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7150: 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);
7151: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7152: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7153: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7154: */
7155:
7156: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7157: first1=1;first2=2;
7158: for (k2=1; k2<=(nlstate);k2++){
7159: for (l2=1; l2<=(nlstate+ndeath);l2++){
7160: if(l2==k2) continue;
7161: j=(k2-1)*(nlstate+ndeath)+l2;
7162: for (k1=1; k1<=(nlstate);k1++){
7163: for (l1=1; l1<=(nlstate+ndeath);l1++){
7164: if(l1==k1) continue;
7165: i=(k1-1)*(nlstate+ndeath)+l1;
7166: if(i<=j) continue;
7167: for (age=bage; age<=fage; age ++){
7168: if ((int)age %5==0){
7169: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7170: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7171: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7172: mu1=mu[i][(int) age]/stepm*YEARM ;
7173: mu2=mu[j][(int) age]/stepm*YEARM;
7174: c12=cv12/sqrt(v1*v2);
7175: /* Computing eigen value of matrix of covariance */
7176: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7177: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7178: if ((lc2 <0) || (lc1 <0) ){
7179: if(first2==1){
7180: first1=0;
7181: 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);
7182: }
7183: 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);
7184: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7185: /* lc2=fabs(lc2); */
7186: }
1.220 brouard 7187:
1.222 brouard 7188: /* Eigen vectors */
1.280 brouard 7189: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7190: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7191: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7192: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7193: }else
7194: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7195: /*v21=sqrt(1.-v11*v11); *//* error */
7196: v21=(lc1-v1)/cv12*v11;
7197: v12=-v21;
7198: v22=v11;
7199: tnalp=v21/v11;
7200: if(first1==1){
7201: first1=0;
7202: 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);
7203: }
7204: 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);
7205: /*printf(fignu*/
7206: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7207: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7208: if(first==1){
7209: first=0;
7210: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7211: fprintf(ficgp,"\nset parametric;unset label");
7212: 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);
7213: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7214: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7215: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7216: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7217: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7218: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7219: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7220: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7221: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7222: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7223: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7224: 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 7225: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7226: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7227: }else{
7228: first=0;
7229: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7230: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7231: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7232: 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 7233: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7234: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7235: }/* if first */
7236: } /* age mod 5 */
7237: } /* end loop age */
7238: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7239: first=1;
7240: } /*l12 */
7241: } /* k12 */
7242: } /*l1 */
7243: }/* k1 */
1.326 brouard 7244: } /* loop on nres */
1.222 brouard 7245: } /* loop on combination of covariates j1 */
7246: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7247: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7248: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7249: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7250: free_vector(xp,1,npar);
7251: fclose(ficresprob);
7252: fclose(ficresprobcov);
7253: fclose(ficresprobcor);
7254: fflush(ficgp);
7255: fflush(fichtmcov);
7256: }
1.126 brouard 7257:
7258:
7259: /******************* Printing html file ***********/
1.201 brouard 7260: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7261: int lastpass, int stepm, int weightopt, char model[],\
7262: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7263: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7264: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7265: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7266: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7267: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7268: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7269: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7270: </ul>");
1.319 brouard 7271: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7272: /* </ul>", model); */
1.214 brouard 7273: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7274: 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",
7275: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
7276: 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 7277: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7278: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7279: fprintf(fichtm,"\
7280: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7281: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7282: fprintf(fichtm,"\
1.217 brouard 7283: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7284: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7285: fprintf(fichtm,"\
1.288 brouard 7286: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7287: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7288: fprintf(fichtm,"\
1.288 brouard 7289: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7290: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7291: fprintf(fichtm,"\
1.211 brouard 7292: - (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 7293: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7294: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7295: if(prevfcast==1){
7296: fprintf(fichtm,"\
7297: - Prevalence projections by age and states: \
1.201 brouard 7298: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7299: }
1.126 brouard 7300:
7301:
1.225 brouard 7302: m=pow(2,cptcoveff);
1.222 brouard 7303: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7304:
1.317 brouard 7305: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7306:
7307: jj1=0;
7308:
7309: fprintf(fichtm," \n<ul>");
7310: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7311: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7312: if(m != 1 && TKresult[nres]!= k1)
7313: continue;
7314: jj1++;
7315: if (cptcovn > 0) {
7316: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
7317: for (cpt=1; cpt<=cptcoveff;cpt++){
7318: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7319: }
7320: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7321: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7322: }
7323: fprintf(fichtm,"\">");
7324:
7325: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7326: fprintf(fichtm,"************ Results for covariates");
7327: for (cpt=1; cpt<=cptcoveff;cpt++){
7328: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7329: }
7330: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7331: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7332: }
7333: if(invalidvarcomb[k1]){
7334: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7335: continue;
7336: }
7337: fprintf(fichtm,"</a></li>");
7338: } /* cptcovn >0 */
7339: }
1.317 brouard 7340: fprintf(fichtm," \n</ul>");
1.264 brouard 7341:
1.222 brouard 7342: jj1=0;
1.237 brouard 7343:
7344: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7345: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7346: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7347: continue;
1.220 brouard 7348:
1.222 brouard 7349: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7350: jj1++;
7351: if (cptcovn > 0) {
1.264 brouard 7352: fprintf(fichtm,"\n<p><a name=\"rescov");
7353: for (cpt=1; cpt<=cptcoveff;cpt++){
7354: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7355: }
7356: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7357: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7358: }
7359: fprintf(fichtm,"\"</a>");
7360:
1.222 brouard 7361: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7362: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7363: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7364: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7365: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7366: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7367: }
1.237 brouard 7368: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7369: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7370: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7371: }
7372:
1.230 brouard 7373: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.321 brouard 7374: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7375: if(invalidvarcomb[k1]){
7376: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7377: printf("\nCombination (%d) ignored because no cases \n",k1);
7378: continue;
7379: }
7380: }
7381: /* aij, bij */
1.259 brouard 7382: 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 7383: <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 7384: /* Pij */
1.241 brouard 7385: 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> \
7386: <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 7387: /* Quasi-incidences */
7388: 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 7389: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7390: 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 7391: 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> \
7392: <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 7393: /* Survival functions (period) in state j */
7394: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7395: fprintf(fichtm,"<br>\n- Survival functions in state %d. And probability to be observed in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, cpt, nlstate, subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
7396: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7397: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7398: }
7399: /* State specific survival functions (period) */
7400: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7401: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7402: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7403: <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> ", cpt, nlstate, cpt, subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
7404: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7405: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7406: }
1.288 brouard 7407: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7408: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7409: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability for a person being in state (1 to %d) at different ages, to be in state %d some years after. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
7410: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"P_"),subdirf2(optionfilefiname,"P_"));
7411: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7412: }
1.296 brouard 7413: if(prevbcast==1){
1.288 brouard 7414: /* Backward prevalence in each health state */
1.222 brouard 7415: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7416: 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 7417: <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 7418: }
1.217 brouard 7419: }
1.222 brouard 7420: if(prevfcast==1){
1.288 brouard 7421: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7422: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7423: 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);
7424: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7425: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7426: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7427: }
7428: }
1.296 brouard 7429: if(prevbcast==1){
1.268 brouard 7430: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7431: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7432: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7433: 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 \
7434: 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 7435: 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);
7436: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7437: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7438: }
7439: }
1.220 brouard 7440:
1.222 brouard 7441: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7442: 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);
7443: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7444: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7445: }
7446: /* } /\* end i1 *\/ */
7447: }/* End k1 */
7448: fprintf(fichtm,"</ul>");
1.126 brouard 7449:
1.222 brouard 7450: fprintf(fichtm,"\
1.126 brouard 7451: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7452: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7453: - 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 7454: But because parameters are usually highly correlated (a higher incidence of disability \
7455: and a higher incidence of recovery can give very close observed transition) it might \
7456: be very useful to look not only at linear confidence intervals estimated from the \
7457: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7458: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7459: covariance matrix of the one-step probabilities. \
7460: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7461:
1.222 brouard 7462: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7463: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7464: fprintf(fichtm,"\
1.126 brouard 7465: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7466: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7467:
1.222 brouard 7468: fprintf(fichtm,"\
1.126 brouard 7469: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7470: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7471: fprintf(fichtm,"\
1.126 brouard 7472: - 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): \
7473: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7474: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7475: fprintf(fichtm,"\
1.126 brouard 7476: - (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): \
7477: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7478: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7479: fprintf(fichtm,"\
1.288 brouard 7480: - 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 7481: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7482: fprintf(fichtm,"\
1.128 brouard 7483: - 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 7484: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7485: fprintf(fichtm,"\
1.288 brouard 7486: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7487: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7488:
7489: /* if(popforecast==1) fprintf(fichtm,"\n */
7490: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7491: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7492: /* <br>",fileres,fileres,fileres,fileres); */
7493: /* else */
7494: /* 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 7495: fflush(fichtm);
1.126 brouard 7496:
1.225 brouard 7497: m=pow(2,cptcoveff);
1.222 brouard 7498: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7499:
1.317 brouard 7500: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7501:
7502: jj1=0;
7503:
7504: fprintf(fichtm," \n<ul>");
7505: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7506: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7507: if(m != 1 && TKresult[nres]!= k1)
7508: continue;
7509: jj1++;
7510: if (cptcovn > 0) {
7511: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
7512: for (cpt=1; cpt<=cptcoveff;cpt++){
7513: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7514: }
7515: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7516: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7517: }
7518: fprintf(fichtm,"\">");
7519:
7520: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7521: fprintf(fichtm,"************ Results for covariates");
7522: for (cpt=1; cpt<=cptcoveff;cpt++){
7523: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7524: }
7525: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7526: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7527: }
7528: if(invalidvarcomb[k1]){
7529: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7530: continue;
7531: }
7532: fprintf(fichtm,"</a></li>");
7533: } /* cptcovn >0 */
7534: }
7535: fprintf(fichtm," \n</ul>");
7536:
1.222 brouard 7537: jj1=0;
1.237 brouard 7538:
1.241 brouard 7539: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7540: for(k1=1; k1<=m;k1++){
1.253 brouard 7541: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7542: continue;
1.222 brouard 7543: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7544: jj1++;
1.126 brouard 7545: if (cptcovn > 0) {
1.317 brouard 7546: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
7547: for (cpt=1; cpt<=cptcoveff;cpt++){
7548: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7549: }
7550: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7551: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7552: }
7553: fprintf(fichtm,"\"</a>");
7554:
1.126 brouard 7555: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.317 brouard 7556: for (cpt=1; cpt<=cptcoveff;cpt++){ /**< cptcoveff number of variables */
1.237 brouard 7557: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
1.317 brouard 7558: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
1.237 brouard 7559: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 7560: }
1.237 brouard 7561: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7562: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7563: }
7564:
1.321 brouard 7565: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 7566:
1.222 brouard 7567: if(invalidvarcomb[k1]){
7568: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7569: continue;
7570: }
1.126 brouard 7571: }
7572: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7573: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7574: 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);
7575: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7576: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7577: }
7578: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7579: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7580: true period expectancies (those weighted with period prevalences are also\
7581: drawn in addition to the population based expectancies computed using\
1.314 brouard 7582: 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);
7583: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7584: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7585: /* } /\* end i1 *\/ */
7586: }/* End k1 */
1.241 brouard 7587: }/* End nres */
1.222 brouard 7588: fprintf(fichtm,"</ul>");
7589: fflush(fichtm);
1.126 brouard 7590: }
7591:
7592: /******************* Gnuplot file **************/
1.296 brouard 7593: 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 7594:
7595: char dirfileres[132],optfileres[132];
1.264 brouard 7596: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7597: 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 7598: int lv=0, vlv=0, kl=0;
1.130 brouard 7599: int ng=0;
1.201 brouard 7600: int vpopbased;
1.223 brouard 7601: int ioffset; /* variable offset for columns */
1.270 brouard 7602: int iyearc=1; /* variable column for year of projection */
7603: int iagec=1; /* variable column for age of projection */
1.235 brouard 7604: int nres=0; /* Index of resultline */
1.266 brouard 7605: int istart=1; /* For starting graphs in projections */
1.219 brouard 7606:
1.126 brouard 7607: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7608: /* printf("Problem with file %s",optionfilegnuplot); */
7609: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7610: /* } */
7611:
7612: /*#ifdef windows */
7613: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7614: /*#endif */
1.225 brouard 7615: m=pow(2,cptcoveff);
1.126 brouard 7616:
1.274 brouard 7617: /* diagram of the model */
7618: fprintf(ficgp,"\n#Diagram of the model \n");
7619: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7620: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7621: 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);
7622:
7623: 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);
7624: fprintf(ficgp,"\n#show arrow\nunset label\n");
7625: 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);
7626: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7627: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7628: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7629: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7630:
1.202 brouard 7631: /* Contribution to likelihood */
7632: /* Plot the probability implied in the likelihood */
1.223 brouard 7633: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7634: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7635: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7636: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7637: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7638: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7639: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7640: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7641: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7642: 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));
7643: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7644: 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));
7645: for (i=1; i<= nlstate ; i ++) {
7646: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7647: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7648: 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);
7649: for (j=2; j<= nlstate+ndeath ; j ++) {
7650: 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);
7651: }
7652: fprintf(ficgp,";\nset out; unset ylabel;\n");
7653: }
7654: /* 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 */
7655: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7656: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7657: fprintf(ficgp,"\nset out;unset log\n");
7658: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7659:
1.126 brouard 7660: strcpy(dirfileres,optionfilefiname);
7661: strcpy(optfileres,"vpl");
1.223 brouard 7662: /* 1eme*/
1.238 brouard 7663: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7664: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7665: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7666: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7667: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7668: continue;
7669: /* We are interested in selected combination by the resultline */
1.246 brouard 7670: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7671: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7672: strcpy(gplotlabel,"(");
1.238 brouard 7673: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7674: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7675: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7676: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7677: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7678: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7679: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7680: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7681: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7682: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7683: }
7684: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7685: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7686: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7687: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7688: }
7689: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7690: /* printf("\n#\n"); */
1.238 brouard 7691: fprintf(ficgp,"\n#\n");
7692: if(invalidvarcomb[k1]){
1.260 brouard 7693: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7694: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7695: continue;
7696: }
1.235 brouard 7697:
1.241 brouard 7698: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7699: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7700: /* 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 7701: fprintf(ficgp,"set title \"Alive state %d %s model=%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 7702: 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);
7703: /* 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); */
7704: /* k1-1 error should be nres-1*/
1.238 brouard 7705: for (i=1; i<= nlstate ; i ++) {
7706: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7707: else fprintf(ficgp," %%*lf (%%*lf)");
7708: }
1.288 brouard 7709: 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 7710: for (i=1; i<= nlstate ; i ++) {
7711: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7712: else fprintf(ficgp," %%*lf (%%*lf)");
7713: }
1.260 brouard 7714: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7715: for (i=1; i<= nlstate ; i ++) {
7716: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7717: else fprintf(ficgp," %%*lf (%%*lf)");
7718: }
1.265 brouard 7719: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7720:
7721: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7722: if(cptcoveff ==0){
1.271 brouard 7723: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7724: }else{
7725: kl=0;
7726: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7727: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7728: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7729: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7730: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7731: vlv= nbcode[Tvaraff[k]][lv];
7732: kl++;
7733: /* 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 *\/ */
7734: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7735: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7736: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7737: if(k==cptcoveff){
7738: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7739: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7740: }else{
7741: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7742: kl++;
7743: }
7744: } /* end covariate */
7745: } /* end if no covariate */
7746:
1.296 brouard 7747: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7748: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7749: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7750: if(cptcoveff ==0){
1.245 brouard 7751: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7752: }else{
7753: kl=0;
7754: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7755: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7756: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7757: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7758: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7759: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7760: kl++;
1.238 brouard 7761: /* 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 *\/ */
7762: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7763: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7764: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7765: if(k==cptcoveff){
1.245 brouard 7766: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7767: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7768: }else{
7769: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7770: kl++;
7771: }
7772: } /* end covariate */
7773: } /* end if no covariate */
1.296 brouard 7774: if(prevbcast == 1){
1.268 brouard 7775: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7776: /* k1-1 error should be nres-1*/
7777: for (i=1; i<= nlstate ; i ++) {
7778: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7779: else fprintf(ficgp," %%*lf (%%*lf)");
7780: }
1.271 brouard 7781: 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 7782: for (i=1; i<= nlstate ; i ++) {
7783: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7784: else fprintf(ficgp," %%*lf (%%*lf)");
7785: }
1.276 brouard 7786: 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 7787: for (i=1; i<= nlstate ; i ++) {
7788: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7789: else fprintf(ficgp," %%*lf (%%*lf)");
7790: }
1.274 brouard 7791: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7792: } /* end if backprojcast */
1.296 brouard 7793: } /* end if prevbcast */
1.276 brouard 7794: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7795: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7796: } /* nres */
1.201 brouard 7797: } /* k1 */
7798: } /* cpt */
1.235 brouard 7799:
7800:
1.126 brouard 7801: /*2 eme*/
1.238 brouard 7802: for (k1=1; k1<= m ; k1 ++){
7803: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7804: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7805: continue;
7806: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7807: strcpy(gplotlabel,"(");
1.238 brouard 7808: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7809: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7810: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7811: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7812: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7813: vlv= nbcode[Tvaraff[k]][lv];
7814: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7815: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7816: }
1.237 brouard 7817: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7818: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7819: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7820: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7821: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7822: }
1.264 brouard 7823: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7824: fprintf(ficgp,"\n#\n");
1.223 brouard 7825: if(invalidvarcomb[k1]){
7826: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7827: continue;
7828: }
1.219 brouard 7829:
1.241 brouard 7830: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7831: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7832: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7833: if(vpopbased==0){
1.238 brouard 7834: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7835: }else
1.238 brouard 7836: fprintf(ficgp,"\nreplot ");
7837: for (i=1; i<= nlstate+1 ; i ++) {
7838: k=2*i;
1.261 brouard 7839: 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 7840: for (j=1; j<= nlstate+1 ; j ++) {
7841: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7842: else fprintf(ficgp," %%*lf (%%*lf)");
7843: }
7844: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7845: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7846: 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 7847: for (j=1; j<= nlstate+1 ; j ++) {
7848: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7849: else fprintf(ficgp," %%*lf (%%*lf)");
7850: }
7851: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7852: 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 7853: for (j=1; j<= nlstate+1 ; j ++) {
7854: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7855: else fprintf(ficgp," %%*lf (%%*lf)");
7856: }
7857: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7858: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7859: } /* state */
7860: } /* vpopbased */
1.264 brouard 7861: 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 7862: } /* end nres */
7863: } /* k1 end 2 eme*/
7864:
7865:
7866: /*3eme*/
7867: for (k1=1; k1<= m ; k1 ++){
7868: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7869: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7870: continue;
7871:
7872: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7873: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7874: strcpy(gplotlabel,"(");
1.238 brouard 7875: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7876: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7877: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7878: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7879: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7880: vlv= nbcode[Tvaraff[k]][lv];
7881: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7882: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7883: }
7884: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7885: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7886: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7887: }
1.264 brouard 7888: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7889: fprintf(ficgp,"\n#\n");
7890: if(invalidvarcomb[k1]){
7891: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7892: continue;
7893: }
7894:
7895: /* k=2+nlstate*(2*cpt-2); */
7896: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7897: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7898: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7899: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7900: 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 7901: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7902: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7903: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7904: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7905: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7906: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7907:
1.238 brouard 7908: */
7909: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7910: 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 7911: /* 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 7912:
1.238 brouard 7913: }
1.261 brouard 7914: 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 7915: }
1.264 brouard 7916: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7917: } /* end nres */
7918: } /* end kl 3eme */
1.126 brouard 7919:
1.223 brouard 7920: /* 4eme */
1.201 brouard 7921: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7922: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7923: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7924: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7925: continue;
1.238 brouard 7926: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7927: strcpy(gplotlabel,"(");
1.238 brouard 7928: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7929: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7930: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7931: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7932: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7933: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7934: vlv= nbcode[Tvaraff[k]][lv];
7935: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7936: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7937: }
7938: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7939: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7940: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7941: }
1.264 brouard 7942: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7943: fprintf(ficgp,"\n#\n");
7944: if(invalidvarcomb[k1]){
7945: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7946: continue;
1.223 brouard 7947: }
1.238 brouard 7948:
1.241 brouard 7949: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7950: 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 7951: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7952: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7953: k=3;
7954: for (i=1; i<= nlstate ; i ++){
7955: if(i==1){
7956: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7957: }else{
7958: fprintf(ficgp,", '' ");
7959: }
7960: l=(nlstate+ndeath)*(i-1)+1;
7961: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7962: for (j=2; j<= nlstate+ndeath ; j ++)
7963: fprintf(ficgp,"+$%d",k+l+j-1);
7964: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7965: } /* nlstate */
1.264 brouard 7966: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7967: } /* end cpt state*/
7968: } /* end nres */
7969: } /* end covariate k1 */
7970:
1.220 brouard 7971: /* 5eme */
1.201 brouard 7972: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7973: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7974: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7975: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7976: continue;
1.238 brouard 7977: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7978: strcpy(gplotlabel,"(");
1.238 brouard 7979: 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);
7980: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7981: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7982: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7983: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7984: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7985: vlv= nbcode[Tvaraff[k]][lv];
7986: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7987: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7988: }
7989: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7990: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7991: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7992: }
1.264 brouard 7993: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7994: fprintf(ficgp,"\n#\n");
7995: if(invalidvarcomb[k1]){
7996: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7997: continue;
7998: }
1.227 brouard 7999:
1.241 brouard 8000: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8001: 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 8002: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8003: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8004: k=3;
8005: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8006: if(j==1)
8007: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8008: else
8009: fprintf(ficgp,", '' ");
8010: l=(nlstate+ndeath)*(cpt-1) +j;
8011: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8012: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8013: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8014: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8015: } /* nlstate */
8016: fprintf(ficgp,", '' ");
8017: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8018: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8019: l=(nlstate+ndeath)*(cpt-1) +j;
8020: if(j < nlstate)
8021: fprintf(ficgp,"$%d +",k+l);
8022: else
8023: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8024: }
1.264 brouard 8025: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8026: } /* end cpt state*/
8027: } /* end covariate */
8028: } /* end nres */
1.227 brouard 8029:
1.220 brouard 8030: /* 6eme */
1.202 brouard 8031: /* CV preval stable (period) for each covariate */
1.237 brouard 8032: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8033: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8034: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8035: continue;
1.255 brouard 8036: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8037: strcpy(gplotlabel,"(");
1.288 brouard 8038: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 8039: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 8040: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
8041: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8042: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8043: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8044: vlv= nbcode[Tvaraff[k]][lv];
8045: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8046: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 8047: }
1.237 brouard 8048: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8049: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8050: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8051: }
1.264 brouard 8052: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8053: fprintf(ficgp,"\n#\n");
1.223 brouard 8054: if(invalidvarcomb[k1]){
1.227 brouard 8055: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8056: continue;
1.223 brouard 8057: }
1.227 brouard 8058:
1.241 brouard 8059: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8060: 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 8061: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8062: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8063: k=3; /* Offset */
1.255 brouard 8064: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8065: if(i==1)
8066: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8067: else
8068: fprintf(ficgp,", '' ");
1.255 brouard 8069: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8070: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8071: for (j=2; j<= nlstate ; j ++)
8072: fprintf(ficgp,"+$%d",k+l+j-1);
8073: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8074: } /* nlstate */
1.264 brouard 8075: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8076: } /* end cpt state*/
8077: } /* end covariate */
1.227 brouard 8078:
8079:
1.220 brouard 8080: /* 7eme */
1.296 brouard 8081: if(prevbcast == 1){
1.288 brouard 8082: /* CV backward prevalence for each covariate */
1.237 brouard 8083: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8084: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8085: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8086: continue;
1.268 brouard 8087: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8088: strcpy(gplotlabel,"(");
1.288 brouard 8089: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8090: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
8091: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
8092: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8093: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 8094: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 8095: vlv= nbcode[Tvaraff[k]][lv];
8096: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8097: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8098: }
1.237 brouard 8099: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8100: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8101: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8102: }
1.264 brouard 8103: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8104: fprintf(ficgp,"\n#\n");
8105: if(invalidvarcomb[k1]){
8106: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8107: continue;
8108: }
8109:
1.241 brouard 8110: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8111: 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 8112: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8113: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8114: k=3; /* Offset */
1.268 brouard 8115: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8116: if(i==1)
8117: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8118: else
8119: fprintf(ficgp,", '' ");
8120: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8121: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8122: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8123: /* 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 8124: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8125: /* for (j=2; j<= nlstate ; j ++) */
8126: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8127: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8128: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8129: } /* nlstate */
1.264 brouard 8130: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8131: } /* end cpt state*/
8132: } /* end covariate */
1.296 brouard 8133: } /* End if prevbcast */
1.218 brouard 8134:
1.223 brouard 8135: /* 8eme */
1.218 brouard 8136: if(prevfcast==1){
1.288 brouard 8137: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8138:
1.237 brouard 8139: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8140: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8141: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8142: continue;
1.211 brouard 8143: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8144: strcpy(gplotlabel,"(");
1.288 brouard 8145: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8146: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8147: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8148: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8149: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8150: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8151: vlv= nbcode[Tvaraff[k]][lv];
8152: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8153: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8154: }
1.237 brouard 8155: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8156: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8157: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8158: }
1.264 brouard 8159: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8160: fprintf(ficgp,"\n#\n");
8161: if(invalidvarcomb[k1]){
8162: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8163: continue;
8164: }
8165:
8166: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8167: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8168: 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 8169: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8170: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8171:
8172: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8173: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8174: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8175: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8176: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8177: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8178: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8179: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8180: if(i==istart){
1.227 brouard 8181: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8182: }else{
8183: fprintf(ficgp,",\\\n '' ");
8184: }
8185: if(cptcoveff ==0){ /* No covariate */
8186: ioffset=2; /* Age is in 2 */
8187: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8188: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8189: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8190: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8191: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8192: if(i==nlstate+1){
1.270 brouard 8193: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8194: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8195: fprintf(ficgp,",\\\n '' ");
8196: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8197: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8198: offyear, \
1.268 brouard 8199: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8200: }else
1.227 brouard 8201: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8202: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8203: }else{ /* more than 2 covariates */
1.270 brouard 8204: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8205: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8206: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8207: iyearc=ioffset-1;
8208: iagec=ioffset;
1.227 brouard 8209: fprintf(ficgp," u %d:(",ioffset);
8210: kl=0;
8211: strcpy(gplotcondition,"(");
8212: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8213: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8214: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8215: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8216: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8217: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8218: kl++;
8219: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8220: kl++;
8221: if(k <cptcoveff && cptcoveff>1)
8222: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8223: }
8224: strcpy(gplotcondition+strlen(gplotcondition),")");
8225: /* 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 *\/ */
8226: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8227: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8228: /* '' 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*/
8229: if(i==nlstate+1){
1.270 brouard 8230: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8231: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8232: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8233: fprintf(ficgp," u %d:(",iagec);
8234: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8235: iyearc, iagec, offyear, \
8236: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8237: /* '' 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 8238: }else{
8239: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8240: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8241: }
8242: } /* end if covariate */
8243: } /* nlstate */
1.264 brouard 8244: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8245: } /* end cpt state*/
8246: } /* end covariate */
8247: } /* End if prevfcast */
1.227 brouard 8248:
1.296 brouard 8249: if(prevbcast==1){
1.268 brouard 8250: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8251:
8252: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8253: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8254: if(m != 1 && TKresult[nres]!= k1)
8255: continue;
8256: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8257: strcpy(gplotlabel,"(");
8258: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
8259: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8260: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8261: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8262: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8263: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8264: vlv= nbcode[Tvaraff[k]][lv];
8265: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8266: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8267: }
8268: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8269: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8270: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8271: }
8272: strcpy(gplotlabel+strlen(gplotlabel),")");
8273: fprintf(ficgp,"\n#\n");
8274: if(invalidvarcomb[k1]){
8275: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8276: continue;
8277: }
8278:
8279: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8280: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8281: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8282: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8283: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8284:
8285: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8286: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8287: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8288: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8289: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8290: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8291: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8292: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8293: if(i==istart){
8294: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8295: }else{
8296: fprintf(ficgp,",\\\n '' ");
8297: }
8298: if(cptcoveff ==0){ /* No covariate */
8299: ioffset=2; /* Age is in 2 */
8300: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8301: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8302: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8303: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8304: fprintf(ficgp," u %d:(", ioffset);
8305: if(i==nlstate+1){
1.270 brouard 8306: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8307: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8308: fprintf(ficgp,",\\\n '' ");
8309: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8310: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8311: offbyear, \
8312: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8313: }else
8314: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8315: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8316: }else{ /* more than 2 covariates */
1.270 brouard 8317: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8318: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8319: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8320: iyearc=ioffset-1;
8321: iagec=ioffset;
1.268 brouard 8322: fprintf(ficgp," u %d:(",ioffset);
8323: kl=0;
8324: strcpy(gplotcondition,"(");
8325: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8326: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8327: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8328: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8329: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8330: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8331: kl++;
8332: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8333: kl++;
8334: if(k <cptcoveff && cptcoveff>1)
8335: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8336: }
8337: strcpy(gplotcondition+strlen(gplotcondition),")");
8338: /* 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 *\/ */
8339: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8340: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8341: /* '' 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*/
8342: if(i==nlstate+1){
1.270 brouard 8343: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8344: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8345: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8346: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8347: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8348: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8349: iyearc,iagec,offbyear, \
8350: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8351: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8352: }else{
8353: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8354: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8355: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8356: }
8357: } /* end if covariate */
8358: } /* nlstate */
8359: fprintf(ficgp,"\nset out; unset label;\n");
8360: } /* end cpt state*/
8361: } /* end covariate */
1.296 brouard 8362: } /* End if prevbcast */
1.268 brouard 8363:
1.227 brouard 8364:
1.238 brouard 8365: /* 9eme writing MLE parameters */
8366: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8367: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8368: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8369: for(k=1; k <=(nlstate+ndeath); k++){
8370: if (k != i) {
1.227 brouard 8371: fprintf(ficgp,"# current state %d\n",k);
8372: for(j=1; j <=ncovmodel; j++){
8373: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8374: jk++;
8375: }
8376: fprintf(ficgp,"\n");
1.126 brouard 8377: }
8378: }
1.223 brouard 8379: }
1.187 brouard 8380: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8381:
1.145 brouard 8382: /*goto avoid;*/
1.238 brouard 8383: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8384: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8385: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8386: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8387: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8388: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8389: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8390: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8391: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8392: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8393: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8394: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8395: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8396: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8397: fprintf(ficgp,"#\n");
1.223 brouard 8398: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8399: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8400: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8401: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8402: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8403: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8404: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8405: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8406: continue;
1.264 brouard 8407: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8408: strcpy(gplotlabel,"(");
1.276 brouard 8409: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8410: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8411: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8412: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8413: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8414: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8415: vlv= nbcode[Tvaraff[k]][lv];
8416: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8417: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8418: }
1.237 brouard 8419: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8420: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8421: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8422: }
1.264 brouard 8423: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8424: fprintf(ficgp,"\n#\n");
1.264 brouard 8425: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8426: fprintf(ficgp,"\nset key outside ");
8427: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8428: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8429: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8430: if (ng==1){
8431: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8432: fprintf(ficgp,"\nunset log y");
8433: }else if (ng==2){
8434: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8435: fprintf(ficgp,"\nset log y");
8436: }else if (ng==3){
8437: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8438: fprintf(ficgp,"\nset log y");
8439: }else
8440: fprintf(ficgp,"\nunset title ");
8441: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8442: i=1;
8443: for(k2=1; k2<=nlstate; k2++) {
8444: k3=i;
8445: for(k=1; k<=(nlstate+ndeath); k++) {
8446: if (k != k2){
8447: switch( ng) {
8448: case 1:
8449: if(nagesqr==0)
8450: fprintf(ficgp," p%d+p%d*x",i,i+1);
8451: else /* nagesqr =1 */
8452: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8453: break;
8454: case 2: /* ng=2 */
8455: if(nagesqr==0)
8456: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8457: else /* nagesqr =1 */
8458: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8459: break;
8460: case 3:
8461: if(nagesqr==0)
8462: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8463: else /* nagesqr =1 */
8464: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8465: break;
8466: }
8467: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8468: ijp=1; /* product no age */
8469: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8470: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8471: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 8472: switch(Typevar[j]){
8473: case 1:
8474: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8475: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
8476: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8477: if(DummyV[j]==0){/* Bug valgrind */
8478: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8479: }else{ /* quantitative */
8480: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8481: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8482: }
8483: ij++;
1.268 brouard 8484: }
1.237 brouard 8485: }
1.329 brouard 8486: }
8487: break;
8488: case 2:
8489: if(cptcovprod >0){
8490: if(j==Tprod[ijp]) { /* */
8491: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8492: if(ijp <=cptcovprod) { /* Product */
8493: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8494: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8495: /* 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)]); */
8496: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8497: }else{ /* Vn is dummy and Vm is quanti */
8498: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8499: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8500: }
8501: }else{ /* Vn*Vm Vn is quanti */
8502: if(DummyV[Tvard[ijp][2]]==0){
8503: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8504: }else{ /* Both quanti */
8505: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8506: }
1.268 brouard 8507: }
1.329 brouard 8508: ijp++;
1.237 brouard 8509: }
1.329 brouard 8510: } /* end Tprod */
8511: }
8512: break;
8513: case 0:
8514: /* simple covariate */
1.264 brouard 8515: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8516: if(Dummy[j]==0){
8517: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8518: }else{ /* quantitative */
8519: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8520: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8521: }
1.329 brouard 8522: /* end simple */
8523: break;
8524: default:
8525: break;
8526: } /* end switch */
1.237 brouard 8527: } /* end j */
1.329 brouard 8528: }else{ /* k=k2 */
8529: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
8530: fprintf(ficgp," (1.");i=i-ncovmodel;
8531: }else
8532: i=i-ncovmodel;
1.223 brouard 8533: }
1.227 brouard 8534:
1.223 brouard 8535: if(ng != 1){
8536: fprintf(ficgp,")/(1");
1.227 brouard 8537:
1.264 brouard 8538: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8539: if(nagesqr==0)
1.264 brouard 8540: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8541: else /* nagesqr =1 */
1.264 brouard 8542: 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 8543:
1.223 brouard 8544: ij=1;
1.329 brouard 8545: ijp=1;
8546: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
8547: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
8548: switch(Typevar[j]){
8549: case 1:
8550: if(cptcovage >0){
8551: if(j==Tage[ij]) { /* Bug valgrind */
8552: if(ij <=cptcovage) { /* Bug valgrind */
8553: if(DummyV[j]==0){/* Bug valgrind */
8554: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
8555: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
8556: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
8557: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
8558: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8559: }else{ /* quantitative */
8560: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8561: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8562: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8563: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8564: }
8565: ij++;
8566: }
8567: }
8568: }
8569: break;
8570: case 2:
8571: if(cptcovprod >0){
8572: if(j==Tprod[ijp]) { /* */
8573: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8574: if(ijp <=cptcovprod) { /* Product */
8575: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8576: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8577: /* 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)]); */
8578: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8579: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
8580: }else{ /* Vn is dummy and Vm is quanti */
8581: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8582: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8583: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8584: }
8585: }else{ /* Vn*Vm Vn is quanti */
8586: if(DummyV[Tvard[ijp][2]]==0){
8587: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8588: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
8589: }else{ /* Both quanti */
8590: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8591: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8592: }
8593: }
8594: ijp++;
8595: }
8596: } /* end Tprod */
8597: } /* end if */
8598: break;
8599: case 0:
8600: /* simple covariate */
8601: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
8602: if(Dummy[j]==0){
8603: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8604: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
8605: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8606: }else{ /* quantitative */
8607: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
8608: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
8609: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8610: }
8611: /* end simple */
8612: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
8613: break;
8614: default:
8615: break;
8616: } /* end switch */
1.223 brouard 8617: }
8618: fprintf(ficgp,")");
8619: }
8620: fprintf(ficgp,")");
8621: if(ng ==2)
1.276 brouard 8622: 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 8623: else /* ng= 3 */
1.276 brouard 8624: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"i%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.329 brouard 8625: }else{ /* end ng <> 1 */
1.223 brouard 8626: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8627: 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 8628: }
8629: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8630: fprintf(ficgp,",");
8631: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8632: fprintf(ficgp,",");
8633: i=i+ncovmodel;
8634: } /* end k */
8635: } /* end k2 */
1.276 brouard 8636: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8637: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8638: } /* end k1 */
1.223 brouard 8639: } /* end ng */
8640: /* avoid: */
8641: fflush(ficgp);
1.126 brouard 8642: } /* end gnuplot */
8643:
8644:
8645: /*************** Moving average **************/
1.219 brouard 8646: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8647: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8648:
1.222 brouard 8649: int i, cpt, cptcod;
8650: int modcovmax =1;
8651: int mobilavrange, mob;
8652: int iage=0;
1.288 brouard 8653: int firstA1=0, firstA2=0;
1.222 brouard 8654:
1.266 brouard 8655: double sum=0., sumr=0.;
1.222 brouard 8656: double age;
1.266 brouard 8657: double *sumnewp, *sumnewm, *sumnewmr;
8658: double *agemingood, *agemaxgood;
8659: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8660:
8661:
1.278 brouard 8662: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8663: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8664:
8665: sumnewp = vector(1,ncovcombmax);
8666: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8667: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8668: agemingood = vector(1,ncovcombmax);
1.266 brouard 8669: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8670: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8671: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8672:
8673: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8674: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8675: sumnewp[cptcod]=0.;
1.266 brouard 8676: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8677: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8678: }
8679: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8680:
1.266 brouard 8681: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8682: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8683: else mobilavrange=mobilav;
8684: for (age=bage; age<=fage; age++)
8685: for (i=1; i<=nlstate;i++)
8686: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8687: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8688: /* We keep the original values on the extreme ages bage, fage and for
8689: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8690: we use a 5 terms etc. until the borders are no more concerned.
8691: */
8692: for (mob=3;mob <=mobilavrange;mob=mob+2){
8693: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8694: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8695: sumnewm[cptcod]=0.;
8696: for (i=1; i<=nlstate;i++){
1.222 brouard 8697: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8698: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8699: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8700: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8701: }
8702: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8703: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8704: } /* end i */
8705: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8706: } /* end cptcod */
1.222 brouard 8707: }/* end age */
8708: }/* end mob */
1.266 brouard 8709: }else{
8710: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8711: return -1;
1.266 brouard 8712: }
8713:
8714: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8715: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8716: if(invalidvarcomb[cptcod]){
8717: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8718: continue;
8719: }
1.219 brouard 8720:
1.266 brouard 8721: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8722: sumnewm[cptcod]=0.;
8723: sumnewmr[cptcod]=0.;
8724: for (i=1; i<=nlstate;i++){
8725: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8726: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8727: }
8728: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8729: agemingoodr[cptcod]=age;
8730: }
8731: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8732: agemingood[cptcod]=age;
8733: }
8734: } /* age */
8735: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8736: sumnewm[cptcod]=0.;
1.266 brouard 8737: sumnewmr[cptcod]=0.;
1.222 brouard 8738: for (i=1; i<=nlstate;i++){
8739: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8740: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8741: }
8742: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8743: agemaxgoodr[cptcod]=age;
1.222 brouard 8744: }
8745: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8746: agemaxgood[cptcod]=age;
8747: }
8748: } /* age */
8749: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8750: /* but they will change */
1.288 brouard 8751: firstA1=0;firstA2=0;
1.266 brouard 8752: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8753: sumnewm[cptcod]=0.;
8754: sumnewmr[cptcod]=0.;
8755: for (i=1; i<=nlstate;i++){
8756: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8757: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8758: }
8759: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8760: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8761: agemaxgoodr[cptcod]=age; /* age min */
8762: for (i=1; i<=nlstate;i++)
8763: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8764: }else{ /* bad we change the value with the values of good ages */
8765: for (i=1; i<=nlstate;i++){
8766: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8767: } /* i */
8768: } /* end bad */
8769: }else{
8770: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8771: agemaxgood[cptcod]=age;
8772: }else{ /* bad we change the value with the values of good ages */
8773: for (i=1; i<=nlstate;i++){
8774: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8775: } /* i */
8776: } /* end bad */
8777: }/* end else */
8778: sum=0.;sumr=0.;
8779: for (i=1; i<=nlstate;i++){
8780: sum+=mobaverage[(int)age][i][cptcod];
8781: sumr+=probs[(int)age][i][cptcod];
8782: }
8783: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8784: if(!firstA1){
8785: firstA1=1;
8786: 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);
8787: }
8788: 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 8789: } /* end bad */
8790: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8791: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8792: if(!firstA2){
8793: firstA2=1;
8794: 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);
8795: }
8796: 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 8797: } /* end bad */
8798: }/* age */
1.266 brouard 8799:
8800: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8801: sumnewm[cptcod]=0.;
1.266 brouard 8802: sumnewmr[cptcod]=0.;
1.222 brouard 8803: for (i=1; i<=nlstate;i++){
8804: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8805: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8806: }
8807: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8808: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8809: agemingoodr[cptcod]=age;
8810: for (i=1; i<=nlstate;i++)
8811: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8812: }else{ /* bad we change the value with the values of good ages */
8813: for (i=1; i<=nlstate;i++){
8814: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8815: } /* i */
8816: } /* end bad */
8817: }else{
8818: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8819: agemingood[cptcod]=age;
8820: }else{ /* bad */
8821: for (i=1; i<=nlstate;i++){
8822: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8823: } /* i */
8824: } /* end bad */
8825: }/* end else */
8826: sum=0.;sumr=0.;
8827: for (i=1; i<=nlstate;i++){
8828: sum+=mobaverage[(int)age][i][cptcod];
8829: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8830: }
1.266 brouard 8831: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8832: 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 8833: } /* end bad */
8834: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8835: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8836: 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 8837: } /* end bad */
8838: }/* age */
1.266 brouard 8839:
1.222 brouard 8840:
8841: for (age=bage; age<=fage; age++){
1.235 brouard 8842: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8843: sumnewp[cptcod]=0.;
8844: sumnewm[cptcod]=0.;
8845: for (i=1; i<=nlstate;i++){
8846: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8847: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8848: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8849: }
8850: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8851: }
8852: /* printf("\n"); */
8853: /* } */
1.266 brouard 8854:
1.222 brouard 8855: /* brutal averaging */
1.266 brouard 8856: /* for (i=1; i<=nlstate;i++){ */
8857: /* for (age=1; age<=bage; age++){ */
8858: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8859: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8860: /* } */
8861: /* for (age=fage; age<=AGESUP; age++){ */
8862: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8863: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8864: /* } */
8865: /* } /\* end i status *\/ */
8866: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8867: /* for (age=1; age<=AGESUP; age++){ */
8868: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8869: /* mobaverage[(int)age][i][cptcod]=0.; */
8870: /* } */
8871: /* } */
1.222 brouard 8872: }/* end cptcod */
1.266 brouard 8873: free_vector(agemaxgoodr,1, ncovcombmax);
8874: free_vector(agemaxgood,1, ncovcombmax);
8875: free_vector(agemingood,1, ncovcombmax);
8876: free_vector(agemingoodr,1, ncovcombmax);
8877: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8878: free_vector(sumnewm,1, ncovcombmax);
8879: free_vector(sumnewp,1, ncovcombmax);
8880: return 0;
8881: }/* End movingaverage */
1.218 brouard 8882:
1.126 brouard 8883:
1.296 brouard 8884:
1.126 brouard 8885: /************** Forecasting ******************/
1.296 brouard 8886: /* 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)*/
8887: 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){
8888: /* dateintemean, mean date of interviews
8889: dateprojd, year, month, day of starting projection
8890: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8891: agemin, agemax range of age
8892: dateprev1 dateprev2 range of dates during which prevalence is computed
8893: */
1.296 brouard 8894: /* double anprojd, mprojd, jprojd; */
8895: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8896: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8897: double agec; /* generic age */
1.296 brouard 8898: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8899: double *popeffectif,*popcount;
8900: double ***p3mat;
1.218 brouard 8901: /* double ***mobaverage; */
1.126 brouard 8902: char fileresf[FILENAMELENGTH];
8903:
8904: agelim=AGESUP;
1.211 brouard 8905: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8906: in each health status at the date of interview (if between dateprev1 and dateprev2).
8907: We still use firstpass and lastpass as another selection.
8908: */
1.214 brouard 8909: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8910: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8911:
1.201 brouard 8912: strcpy(fileresf,"F_");
8913: strcat(fileresf,fileresu);
1.126 brouard 8914: if((ficresf=fopen(fileresf,"w"))==NULL) {
8915: printf("Problem with forecast resultfile: %s\n", fileresf);
8916: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8917: }
1.235 brouard 8918: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8919: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8920:
1.225 brouard 8921: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8922:
8923:
8924: stepsize=(int) (stepm+YEARM-1)/YEARM;
8925: if (stepm<=12) stepsize=1;
8926: if(estepm < stepm){
8927: printf ("Problem %d lower than %d\n",estepm, stepm);
8928: }
1.270 brouard 8929: else{
8930: hstepm=estepm;
8931: }
8932: if(estepm > stepm){ /* Yes every two year */
8933: stepsize=2;
8934: }
1.296 brouard 8935: hstepm=hstepm/stepm;
1.126 brouard 8936:
1.296 brouard 8937:
8938: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8939: /* fractional in yp1 *\/ */
8940: /* aintmean=yp; */
8941: /* yp2=modf((yp1*12),&yp); */
8942: /* mintmean=yp; */
8943: /* yp1=modf((yp2*30.5),&yp); */
8944: /* jintmean=yp; */
8945: /* if(jintmean==0) jintmean=1; */
8946: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8947:
1.296 brouard 8948:
8949: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8950: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8951: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8952: i1=pow(2,cptcoveff);
1.126 brouard 8953: if (cptcovn < 1){i1=1;}
8954:
1.296 brouard 8955: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8956:
8957: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8958:
1.126 brouard 8959: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8960: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8961: for(k=1; k<=i1;k++){
1.253 brouard 8962: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8963: continue;
1.227 brouard 8964: if(invalidvarcomb[k]){
8965: printf("\nCombination (%d) projection ignored because no cases \n",k);
8966: continue;
8967: }
8968: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8969: for(j=1;j<=cptcoveff;j++) {
1.330 ! brouard 8970: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.227 brouard 8971: }
1.235 brouard 8972: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8973: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8974: }
1.227 brouard 8975: fprintf(ficresf," yearproj age");
8976: for(j=1; j<=nlstate+ndeath;j++){
8977: for(i=1; i<=nlstate;i++)
8978: fprintf(ficresf," p%d%d",i,j);
8979: fprintf(ficresf," wp.%d",j);
8980: }
1.296 brouard 8981: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8982: fprintf(ficresf,"\n");
1.296 brouard 8983: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8984: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8985: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8986: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8987: nhstepm = nhstepm/hstepm;
8988: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8989: oldm=oldms;savm=savms;
1.268 brouard 8990: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8991: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8992: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8993: for (h=0; h<=nhstepm; h++){
8994: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8995: break;
8996: }
8997: }
8998: fprintf(ficresf,"\n");
8999: for(j=1;j<=cptcoveff;j++)
1.330 ! brouard 9000: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.296 brouard 9001: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9002:
9003: for(j=1; j<=nlstate+ndeath;j++) {
9004: ppij=0.;
9005: for(i=1; i<=nlstate;i++) {
1.278 brouard 9006: if (mobilav>=1)
9007: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9008: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9009: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9010: }
1.268 brouard 9011: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9012: } /* end i */
9013: fprintf(ficresf," %.3f", ppij);
9014: }/* end j */
1.227 brouard 9015: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9016: } /* end agec */
1.266 brouard 9017: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9018: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9019: } /* end yearp */
9020: } /* end k */
1.219 brouard 9021:
1.126 brouard 9022: fclose(ficresf);
1.215 brouard 9023: printf("End of Computing forecasting \n");
9024: fprintf(ficlog,"End of Computing forecasting\n");
9025:
1.126 brouard 9026: }
9027:
1.269 brouard 9028: /************** Back Forecasting ******************/
1.296 brouard 9029: /* 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){ */
9030: 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){
9031: /* back1, year, month, day of starting backprojection
1.267 brouard 9032: agemin, agemax range of age
9033: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9034: anback2 year of end of backprojection (same day and month as back1).
9035: prevacurrent and prev are prevalences.
1.267 brouard 9036: */
9037: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9038: double agec; /* generic age */
1.302 brouard 9039: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9040: double *popeffectif,*popcount;
9041: double ***p3mat;
9042: /* double ***mobaverage; */
9043: char fileresfb[FILENAMELENGTH];
9044:
1.268 brouard 9045: agelim=AGEINF;
1.267 brouard 9046: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9047: in each health status at the date of interview (if between dateprev1 and dateprev2).
9048: We still use firstpass and lastpass as another selection.
9049: */
9050: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9051: /* firstpass, lastpass, stepm, weightopt, model); */
9052:
9053: /*Do we need to compute prevalence again?*/
9054:
9055: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9056:
9057: strcpy(fileresfb,"FB_");
9058: strcat(fileresfb,fileresu);
9059: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9060: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9061: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9062: }
9063: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9064: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9065:
9066: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9067:
9068:
9069: stepsize=(int) (stepm+YEARM-1)/YEARM;
9070: if (stepm<=12) stepsize=1;
9071: if(estepm < stepm){
9072: printf ("Problem %d lower than %d\n",estepm, stepm);
9073: }
1.270 brouard 9074: else{
9075: hstepm=estepm;
9076: }
9077: if(estepm >= stepm){ /* Yes every two year */
9078: stepsize=2;
9079: }
1.267 brouard 9080:
9081: hstepm=hstepm/stepm;
1.296 brouard 9082: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9083: /* fractional in yp1 *\/ */
9084: /* aintmean=yp; */
9085: /* yp2=modf((yp1*12),&yp); */
9086: /* mintmean=yp; */
9087: /* yp1=modf((yp2*30.5),&yp); */
9088: /* jintmean=yp; */
9089: /* if(jintmean==0) jintmean=1; */
9090: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9091:
9092: i1=pow(2,cptcoveff);
9093: if (cptcovn < 1){i1=1;}
9094:
1.296 brouard 9095: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9096: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9097:
9098: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9099:
9100: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9101: for(k=1; k<=i1;k++){
9102: if(i1 != 1 && TKresult[nres]!= k)
9103: continue;
9104: if(invalidvarcomb[k]){
9105: printf("\nCombination (%d) projection ignored because no cases \n",k);
9106: continue;
9107: }
1.268 brouard 9108: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9109: for(j=1;j<=cptcoveff;j++) {
1.330 ! brouard 9110: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.267 brouard 9111: }
9112: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9113: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9114: }
9115: fprintf(ficresfb," yearbproj age");
9116: for(j=1; j<=nlstate+ndeath;j++){
9117: for(i=1; i<=nlstate;i++)
1.268 brouard 9118: fprintf(ficresfb," b%d%d",i,j);
9119: fprintf(ficresfb," b.%d",j);
1.267 brouard 9120: }
1.296 brouard 9121: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9122: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9123: fprintf(ficresfb,"\n");
1.296 brouard 9124: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9125: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9126: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9127: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9128: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9129: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9130: nhstepm = nhstepm/hstepm;
9131: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9132: oldm=oldms;savm=savms;
1.268 brouard 9133: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9134: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9135: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9136: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9137: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9138: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9139: for (h=0; h<=nhstepm; h++){
1.268 brouard 9140: if (h*hstepm/YEARM*stepm ==-yearp) {
9141: break;
9142: }
9143: }
9144: fprintf(ficresfb,"\n");
9145: for(j=1;j<=cptcoveff;j++)
1.330 ! brouard 9146: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.296 brouard 9147: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9148: for(i=1; i<=nlstate+ndeath;i++) {
9149: ppij=0.;ppi=0.;
9150: for(j=1; j<=nlstate;j++) {
9151: /* if (mobilav==1) */
1.269 brouard 9152: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9153: ppi=ppi+prevacurrent[(int)agec][j][k];
9154: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9155: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9156: /* else { */
9157: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9158: /* } */
1.268 brouard 9159: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9160: } /* end j */
9161: if(ppi <0.99){
9162: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9163: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9164: }
9165: fprintf(ficresfb," %.3f", ppij);
9166: }/* end j */
1.267 brouard 9167: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9168: } /* end agec */
9169: } /* end yearp */
9170: } /* end k */
1.217 brouard 9171:
1.267 brouard 9172: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9173:
1.267 brouard 9174: fclose(ficresfb);
9175: printf("End of Computing Back forecasting \n");
9176: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9177:
1.267 brouard 9178: }
1.217 brouard 9179:
1.269 brouard 9180: /* Variance of prevalence limit: varprlim */
9181: 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 9182: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9183:
9184: char fileresvpl[FILENAMELENGTH];
9185: FILE *ficresvpl;
9186: double **oldm, **savm;
9187: double **varpl; /* Variances of prevalence limits by age */
9188: int i1, k, nres, j ;
9189:
9190: strcpy(fileresvpl,"VPL_");
9191: strcat(fileresvpl,fileresu);
9192: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9193: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9194: exit(0);
9195: }
1.288 brouard 9196: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9197: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9198:
9199: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9200: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9201:
9202: i1=pow(2,cptcoveff);
9203: if (cptcovn < 1){i1=1;}
9204:
9205: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9206: for(k=1; k<=i1;k++){
9207: if(i1 != 1 && TKresult[nres]!= k)
9208: continue;
9209: fprintf(ficresvpl,"\n#****** ");
9210: printf("\n#****** ");
9211: fprintf(ficlog,"\n#****** ");
9212: for(j=1;j<=cptcoveff;j++) {
1.330 ! brouard 9213: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 9214: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 9215: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.269 brouard 9216: }
9217: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9218: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9219: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9220: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9221: }
9222: fprintf(ficresvpl,"******\n");
9223: printf("******\n");
9224: fprintf(ficlog,"******\n");
9225:
9226: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9227: oldm=oldms;savm=savms;
9228: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9229: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9230: /*}*/
9231: }
9232:
9233: fclose(ficresvpl);
1.288 brouard 9234: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9235: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9236:
9237: }
9238: /* Variance of back prevalence: varbprlim */
9239: 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){
9240: /*------- Variance of back (stable) prevalence------*/
9241:
9242: char fileresvbl[FILENAMELENGTH];
9243: FILE *ficresvbl;
9244:
9245: double **oldm, **savm;
9246: double **varbpl; /* Variances of back prevalence limits by age */
9247: int i1, k, nres, j ;
9248:
9249: strcpy(fileresvbl,"VBL_");
9250: strcat(fileresvbl,fileresu);
9251: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9252: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9253: exit(0);
9254: }
9255: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9256: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9257:
9258:
9259: i1=pow(2,cptcoveff);
9260: if (cptcovn < 1){i1=1;}
9261:
9262: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9263: for(k=1; k<=i1;k++){
9264: if(i1 != 1 && TKresult[nres]!= k)
9265: continue;
9266: fprintf(ficresvbl,"\n#****** ");
9267: printf("\n#****** ");
9268: fprintf(ficlog,"\n#****** ");
9269: for(j=1;j<=cptcoveff;j++) {
1.330 ! brouard 9270: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 9271: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 9272: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.269 brouard 9273: }
9274: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9275: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9276: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9277: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9278: }
9279: fprintf(ficresvbl,"******\n");
9280: printf("******\n");
9281: fprintf(ficlog,"******\n");
9282:
9283: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9284: oldm=oldms;savm=savms;
9285:
9286: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9287: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9288: /*}*/
9289: }
9290:
9291: fclose(ficresvbl);
9292: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9293: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9294:
9295: } /* End of varbprlim */
9296:
1.126 brouard 9297: /************** Forecasting *****not tested NB*************/
1.227 brouard 9298: /* 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 9299:
1.227 brouard 9300: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9301: /* int *popage; */
9302: /* double calagedatem, agelim, kk1, kk2; */
9303: /* double *popeffectif,*popcount; */
9304: /* double ***p3mat,***tabpop,***tabpopprev; */
9305: /* /\* double ***mobaverage; *\/ */
9306: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9307:
1.227 brouard 9308: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9309: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9310: /* agelim=AGESUP; */
9311: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9312:
1.227 brouard 9313: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9314:
9315:
1.227 brouard 9316: /* strcpy(filerespop,"POP_"); */
9317: /* strcat(filerespop,fileresu); */
9318: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9319: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9320: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9321: /* } */
9322: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9323: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9324:
1.227 brouard 9325: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9326:
1.227 brouard 9327: /* /\* if (mobilav!=0) { *\/ */
9328: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9329: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9330: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9331: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9332: /* /\* } *\/ */
9333: /* /\* } *\/ */
1.126 brouard 9334:
1.227 brouard 9335: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9336: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9337:
1.227 brouard 9338: /* agelim=AGESUP; */
1.126 brouard 9339:
1.227 brouard 9340: /* hstepm=1; */
9341: /* hstepm=hstepm/stepm; */
1.218 brouard 9342:
1.227 brouard 9343: /* if (popforecast==1) { */
9344: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9345: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9346: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9347: /* } */
9348: /* popage=ivector(0,AGESUP); */
9349: /* popeffectif=vector(0,AGESUP); */
9350: /* popcount=vector(0,AGESUP); */
1.126 brouard 9351:
1.227 brouard 9352: /* i=1; */
9353: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9354:
1.227 brouard 9355: /* imx=i; */
9356: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9357: /* } */
1.218 brouard 9358:
1.227 brouard 9359: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9360: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9361: /* k=k+1; */
9362: /* fprintf(ficrespop,"\n#******"); */
9363: /* for(j=1;j<=cptcoveff;j++) { */
9364: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9365: /* } */
9366: /* fprintf(ficrespop,"******\n"); */
9367: /* fprintf(ficrespop,"# Age"); */
9368: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9369: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9370:
1.227 brouard 9371: /* for (cpt=0; cpt<=0;cpt++) { */
9372: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9373:
1.227 brouard 9374: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9375: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9376: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9377:
1.227 brouard 9378: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9379: /* oldm=oldms;savm=savms; */
9380: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9381:
1.227 brouard 9382: /* for (h=0; h<=nhstepm; h++){ */
9383: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9384: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9385: /* } */
9386: /* for(j=1; j<=nlstate+ndeath;j++) { */
9387: /* kk1=0.;kk2=0; */
9388: /* for(i=1; i<=nlstate;i++) { */
9389: /* if (mobilav==1) */
9390: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9391: /* else { */
9392: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9393: /* } */
9394: /* } */
9395: /* if (h==(int)(calagedatem+12*cpt)){ */
9396: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9397: /* /\*fprintf(ficrespop," %.3f", kk1); */
9398: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9399: /* } */
9400: /* } */
9401: /* for(i=1; i<=nlstate;i++){ */
9402: /* kk1=0.; */
9403: /* for(j=1; j<=nlstate;j++){ */
9404: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9405: /* } */
9406: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9407: /* } */
1.218 brouard 9408:
1.227 brouard 9409: /* if (h==(int)(calagedatem+12*cpt)) */
9410: /* for(j=1; j<=nlstate;j++) */
9411: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9412: /* } */
9413: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9414: /* } */
9415: /* } */
1.218 brouard 9416:
1.227 brouard 9417: /* /\******\/ */
1.218 brouard 9418:
1.227 brouard 9419: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9420: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9421: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9422: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9423: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9424:
1.227 brouard 9425: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9426: /* oldm=oldms;savm=savms; */
9427: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9428: /* for (h=0; h<=nhstepm; h++){ */
9429: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9430: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9431: /* } */
9432: /* for(j=1; j<=nlstate+ndeath;j++) { */
9433: /* kk1=0.;kk2=0; */
9434: /* for(i=1; i<=nlstate;i++) { */
9435: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9436: /* } */
9437: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9438: /* } */
9439: /* } */
9440: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9441: /* } */
9442: /* } */
9443: /* } */
9444: /* } */
1.218 brouard 9445:
1.227 brouard 9446: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9447:
1.227 brouard 9448: /* if (popforecast==1) { */
9449: /* free_ivector(popage,0,AGESUP); */
9450: /* free_vector(popeffectif,0,AGESUP); */
9451: /* free_vector(popcount,0,AGESUP); */
9452: /* } */
9453: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9454: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9455: /* fclose(ficrespop); */
9456: /* } /\* End of popforecast *\/ */
1.218 brouard 9457:
1.126 brouard 9458: int fileappend(FILE *fichier, char *optionfich)
9459: {
9460: if((fichier=fopen(optionfich,"a"))==NULL) {
9461: printf("Problem with file: %s\n", optionfich);
9462: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9463: return (0);
9464: }
9465: fflush(fichier);
9466: return (1);
9467: }
9468:
9469:
9470: /**************** function prwizard **********************/
9471: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9472: {
9473:
9474: /* Wizard to print covariance matrix template */
9475:
1.164 brouard 9476: char ca[32], cb[32];
9477: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9478: int numlinepar;
9479:
9480: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9481: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9482: for(i=1; i <=nlstate; i++){
9483: jj=0;
9484: for(j=1; j <=nlstate+ndeath; j++){
9485: if(j==i) continue;
9486: jj++;
9487: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9488: printf("%1d%1d",i,j);
9489: fprintf(ficparo,"%1d%1d",i,j);
9490: for(k=1; k<=ncovmodel;k++){
9491: /* printf(" %lf",param[i][j][k]); */
9492: /* fprintf(ficparo," %lf",param[i][j][k]); */
9493: printf(" 0.");
9494: fprintf(ficparo," 0.");
9495: }
9496: printf("\n");
9497: fprintf(ficparo,"\n");
9498: }
9499: }
9500: printf("# Scales (for hessian or gradient estimation)\n");
9501: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9502: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9503: for(i=1; i <=nlstate; i++){
9504: jj=0;
9505: for(j=1; j <=nlstate+ndeath; j++){
9506: if(j==i) continue;
9507: jj++;
9508: fprintf(ficparo,"%1d%1d",i,j);
9509: printf("%1d%1d",i,j);
9510: fflush(stdout);
9511: for(k=1; k<=ncovmodel;k++){
9512: /* printf(" %le",delti3[i][j][k]); */
9513: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9514: printf(" 0.");
9515: fprintf(ficparo," 0.");
9516: }
9517: numlinepar++;
9518: printf("\n");
9519: fprintf(ficparo,"\n");
9520: }
9521: }
9522: printf("# Covariance matrix\n");
9523: /* # 121 Var(a12)\n\ */
9524: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9525: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9526: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9527: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9528: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9529: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9530: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9531: fflush(stdout);
9532: fprintf(ficparo,"# Covariance matrix\n");
9533: /* # 121 Var(a12)\n\ */
9534: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9535: /* # ...\n\ */
9536: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9537:
9538: for(itimes=1;itimes<=2;itimes++){
9539: jj=0;
9540: for(i=1; i <=nlstate; i++){
9541: for(j=1; j <=nlstate+ndeath; j++){
9542: if(j==i) continue;
9543: for(k=1; k<=ncovmodel;k++){
9544: jj++;
9545: ca[0]= k+'a'-1;ca[1]='\0';
9546: if(itimes==1){
9547: printf("#%1d%1d%d",i,j,k);
9548: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9549: }else{
9550: printf("%1d%1d%d",i,j,k);
9551: fprintf(ficparo,"%1d%1d%d",i,j,k);
9552: /* printf(" %.5le",matcov[i][j]); */
9553: }
9554: ll=0;
9555: for(li=1;li <=nlstate; li++){
9556: for(lj=1;lj <=nlstate+ndeath; lj++){
9557: if(lj==li) continue;
9558: for(lk=1;lk<=ncovmodel;lk++){
9559: ll++;
9560: if(ll<=jj){
9561: cb[0]= lk +'a'-1;cb[1]='\0';
9562: if(ll<jj){
9563: if(itimes==1){
9564: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9565: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9566: }else{
9567: printf(" 0.");
9568: fprintf(ficparo," 0.");
9569: }
9570: }else{
9571: if(itimes==1){
9572: printf(" Var(%s%1d%1d)",ca,i,j);
9573: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9574: }else{
9575: printf(" 0.");
9576: fprintf(ficparo," 0.");
9577: }
9578: }
9579: }
9580: } /* end lk */
9581: } /* end lj */
9582: } /* end li */
9583: printf("\n");
9584: fprintf(ficparo,"\n");
9585: numlinepar++;
9586: } /* end k*/
9587: } /*end j */
9588: } /* end i */
9589: } /* end itimes */
9590:
9591: } /* end of prwizard */
9592: /******************* Gompertz Likelihood ******************************/
9593: double gompertz(double x[])
9594: {
1.302 brouard 9595: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9596: int i,n=0; /* n is the size of the sample */
9597:
1.220 brouard 9598: for (i=1;i<=imx ; i++) {
1.126 brouard 9599: sump=sump+weight[i];
9600: /* sump=sump+1;*/
9601: num=num+1;
9602: }
1.302 brouard 9603: L=0.0;
9604: /* agegomp=AGEGOMP; */
1.126 brouard 9605: /* for (i=0; i<=imx; i++)
9606: 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]);*/
9607:
1.302 brouard 9608: for (i=1;i<=imx ; i++) {
9609: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9610: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9611: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9612: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9613: * +
9614: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9615: */
9616: if (wav[i] > 1 || agedc[i] < AGESUP) {
9617: if (cens[i] == 1){
9618: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9619: } else if (cens[i] == 0){
1.126 brouard 9620: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9621: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9622: } else
9623: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9624: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9625: L=L+A*weight[i];
1.126 brouard 9626: /* 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 9627: }
9628: }
1.126 brouard 9629:
1.302 brouard 9630: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9631:
9632: return -2*L*num/sump;
9633: }
9634:
1.136 brouard 9635: #ifdef GSL
9636: /******************* Gompertz_f Likelihood ******************************/
9637: double gompertz_f(const gsl_vector *v, void *params)
9638: {
1.302 brouard 9639: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9640: double *x= (double *) v->data;
9641: int i,n=0; /* n is the size of the sample */
9642:
9643: for (i=0;i<=imx-1 ; i++) {
9644: sump=sump+weight[i];
9645: /* sump=sump+1;*/
9646: num=num+1;
9647: }
9648:
9649:
9650: /* for (i=0; i<=imx; i++)
9651: 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]);*/
9652: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9653: for (i=1;i<=imx ; i++)
9654: {
9655: if (cens[i] == 1 && wav[i]>1)
9656: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9657:
9658: if (cens[i] == 0 && wav[i]>1)
9659: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9660: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9661:
9662: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9663: if (wav[i] > 1 ) { /* ??? */
9664: LL=LL+A*weight[i];
9665: /* 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]);*/
9666: }
9667: }
9668:
9669: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9670: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9671:
9672: return -2*LL*num/sump;
9673: }
9674: #endif
9675:
1.126 brouard 9676: /******************* Printing html file ***********/
1.201 brouard 9677: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9678: int lastpass, int stepm, int weightopt, char model[],\
9679: int imx, double p[],double **matcov,double agemortsup){
9680: int i,k;
9681:
9682: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9683: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9684: for (i=1;i<=2;i++)
9685: 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 9686: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9687: fprintf(fichtm,"</ul>");
9688:
9689: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9690:
9691: 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>");
9692:
9693: for (k=agegomp;k<(agemortsup-2);k++)
9694: 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]);
9695:
9696:
9697: fflush(fichtm);
9698: }
9699:
9700: /******************* Gnuplot file **************/
1.201 brouard 9701: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9702:
9703: char dirfileres[132],optfileres[132];
1.164 brouard 9704:
1.126 brouard 9705: int ng;
9706:
9707:
9708: /*#ifdef windows */
9709: fprintf(ficgp,"cd \"%s\" \n",pathc);
9710: /*#endif */
9711:
9712:
9713: strcpy(dirfileres,optionfilefiname);
9714: strcpy(optfileres,"vpl");
1.199 brouard 9715: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9716: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9717: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9718: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9719: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9720:
9721: }
9722:
1.136 brouard 9723: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9724: {
1.126 brouard 9725:
1.136 brouard 9726: /*-------- data file ----------*/
9727: FILE *fic;
9728: char dummy[]=" ";
1.240 brouard 9729: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9730: int lstra;
1.136 brouard 9731: int linei, month, year,iout;
1.302 brouard 9732: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9733: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9734: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9735: char *stratrunc;
1.223 brouard 9736:
1.240 brouard 9737: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9738: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 9739: for(v=1;v<NCOVMAX;v++){
9740: DummyV[v]=0;
9741: FixedV[v]=0;
9742: }
1.126 brouard 9743:
1.240 brouard 9744: for(v=1; v <=ncovcol;v++){
9745: DummyV[v]=0;
9746: FixedV[v]=0;
9747: }
9748: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9749: DummyV[v]=1;
9750: FixedV[v]=0;
9751: }
9752: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9753: DummyV[v]=0;
9754: FixedV[v]=1;
9755: }
9756: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9757: DummyV[v]=1;
9758: FixedV[v]=1;
9759: }
9760: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9761: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9762: 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]);
9763: }
1.126 brouard 9764:
1.136 brouard 9765: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9766: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9767: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9768: }
1.126 brouard 9769:
1.302 brouard 9770: /* Is it a BOM UTF-8 Windows file? */
9771: /* First data line */
9772: linei=0;
9773: while(fgets(line, MAXLINE, fic)) {
9774: noffset=0;
9775: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9776: {
9777: noffset=noffset+3;
9778: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9779: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9780: fflush(ficlog); return 1;
9781: }
9782: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9783: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9784: {
9785: noffset=noffset+2;
1.304 brouard 9786: 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);
9787: 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 9788: fflush(ficlog); return 1;
9789: }
9790: else if( line[0] == 0 && line[1] == 0)
9791: {
9792: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9793: noffset=noffset+4;
1.304 brouard 9794: 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);
9795: 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 9796: fflush(ficlog); return 1;
9797: }
9798: } else{
9799: ;/*printf(" Not a BOM file\n");*/
9800: }
9801: /* If line starts with a # it is a comment */
9802: if (line[noffset] == '#') {
9803: linei=linei+1;
9804: break;
9805: }else{
9806: break;
9807: }
9808: }
9809: fclose(fic);
9810: if((fic=fopen(datafile,"r"))==NULL) {
9811: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9812: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9813: }
9814: /* Not a Bom file */
9815:
1.136 brouard 9816: i=1;
9817: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9818: linei=linei+1;
9819: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9820: if(line[j] == '\t')
9821: line[j] = ' ';
9822: }
9823: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9824: ;
9825: };
9826: line[j+1]=0; /* Trims blanks at end of line */
9827: if(line[0]=='#'){
9828: fprintf(ficlog,"Comment line\n%s\n",line);
9829: printf("Comment line\n%s\n",line);
9830: continue;
9831: }
9832: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9833: strcpy(line, linetmp);
1.223 brouard 9834:
9835: /* Loops on waves */
9836: for (j=maxwav;j>=1;j--){
9837: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9838: cutv(stra, strb, line, ' ');
9839: if(strb[0]=='.') { /* Missing value */
9840: lval=-1;
9841: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9842: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9843: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9844: 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);
9845: 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);
9846: return 1;
9847: }
9848: }else{
9849: errno=0;
9850: /* what_kind_of_number(strb); */
9851: dval=strtod(strb,&endptr);
9852: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9853: /* if(strb != endptr && *endptr == '\0') */
9854: /* dval=dlval; */
9855: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9856: if( strb[0]=='\0' || (*endptr != '\0')){
9857: 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);
9858: 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);
9859: return 1;
9860: }
9861: cotqvar[j][iv][i]=dval;
9862: cotvar[j][ntv+iv][i]=dval;
9863: }
9864: strcpy(line,stra);
1.223 brouard 9865: }/* end loop ntqv */
1.225 brouard 9866:
1.223 brouard 9867: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9868: cutv(stra, strb, line, ' ');
9869: if(strb[0]=='.') { /* Missing value */
9870: lval=-1;
9871: }else{
9872: errno=0;
9873: lval=strtol(strb,&endptr,10);
9874: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9875: if( strb[0]=='\0' || (*endptr != '\0')){
9876: 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);
9877: 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);
9878: return 1;
9879: }
9880: }
9881: if(lval <-1 || lval >1){
9882: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9883: 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 9884: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9885: For example, for multinomial values like 1, 2 and 3,\n \
9886: build V1=0 V2=0 for the reference value (1),\n \
9887: V1=1 V2=0 for (2) \n \
1.223 brouard 9888: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9889: output of IMaCh is often meaningless.\n \
1.319 brouard 9890: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 9891: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9892: 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 9893: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9894: For example, for multinomial values like 1, 2 and 3,\n \
9895: build V1=0 V2=0 for the reference value (1),\n \
9896: V1=1 V2=0 for (2) \n \
1.223 brouard 9897: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9898: output of IMaCh is often meaningless.\n \
1.319 brouard 9899: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 9900: return 1;
9901: }
9902: cotvar[j][iv][i]=(double)(lval);
9903: strcpy(line,stra);
1.223 brouard 9904: }/* end loop ntv */
1.225 brouard 9905:
1.223 brouard 9906: /* Statuses at wave */
1.137 brouard 9907: cutv(stra, strb, line, ' ');
1.223 brouard 9908: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9909: lval=-1;
1.136 brouard 9910: }else{
1.238 brouard 9911: errno=0;
9912: lval=strtol(strb,&endptr,10);
9913: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9914: if( strb[0]=='\0' || (*endptr != '\0')){
9915: 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);
9916: 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);
9917: return 1;
9918: }
1.136 brouard 9919: }
1.225 brouard 9920:
1.136 brouard 9921: s[j][i]=lval;
1.225 brouard 9922:
1.223 brouard 9923: /* Date of Interview */
1.136 brouard 9924: strcpy(line,stra);
9925: cutv(stra, strb,line,' ');
1.169 brouard 9926: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9927: }
1.169 brouard 9928: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9929: month=99;
9930: year=9999;
1.136 brouard 9931: }else{
1.225 brouard 9932: 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);
9933: 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);
9934: return 1;
1.136 brouard 9935: }
9936: anint[j][i]= (double) year;
1.302 brouard 9937: mint[j][i]= (double)month;
9938: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
9939: /* 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]); */
9940: /* 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]); */
9941: /* } */
1.136 brouard 9942: strcpy(line,stra);
1.223 brouard 9943: } /* End loop on waves */
1.225 brouard 9944:
1.223 brouard 9945: /* Date of death */
1.136 brouard 9946: cutv(stra, strb,line,' ');
1.169 brouard 9947: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9948: }
1.169 brouard 9949: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9950: month=99;
9951: year=9999;
9952: }else{
1.141 brouard 9953: 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 9954: 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);
9955: return 1;
1.136 brouard 9956: }
9957: andc[i]=(double) year;
9958: moisdc[i]=(double) month;
9959: strcpy(line,stra);
9960:
1.223 brouard 9961: /* Date of birth */
1.136 brouard 9962: cutv(stra, strb,line,' ');
1.169 brouard 9963: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9964: }
1.169 brouard 9965: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9966: month=99;
9967: year=9999;
9968: }else{
1.141 brouard 9969: 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);
9970: 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 9971: return 1;
1.136 brouard 9972: }
9973: if (year==9999) {
1.141 brouard 9974: 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);
9975: 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 9976: return 1;
9977:
1.136 brouard 9978: }
9979: annais[i]=(double)(year);
1.302 brouard 9980: moisnais[i]=(double)(month);
9981: for (j=1;j<=maxwav;j++){
9982: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
9983: 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]);
9984: 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]);
9985: }
9986: }
9987:
1.136 brouard 9988: strcpy(line,stra);
1.225 brouard 9989:
1.223 brouard 9990: /* Sample weight */
1.136 brouard 9991: cutv(stra, strb,line,' ');
9992: errno=0;
9993: dval=strtod(strb,&endptr);
9994: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9995: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9996: 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 9997: fflush(ficlog);
9998: return 1;
9999: }
10000: weight[i]=dval;
10001: strcpy(line,stra);
1.225 brouard 10002:
1.223 brouard 10003: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10004: cutv(stra, strb, line, ' ');
10005: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10006: lval=-1;
1.311 brouard 10007: coqvar[iv][i]=NAN;
10008: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10009: }else{
1.225 brouard 10010: errno=0;
10011: /* what_kind_of_number(strb); */
10012: dval=strtod(strb,&endptr);
10013: /* if(strb != endptr && *endptr == '\0') */
10014: /* dval=dlval; */
10015: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10016: if( strb[0]=='\0' || (*endptr != '\0')){
10017: 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);
10018: 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);
10019: return 1;
10020: }
10021: coqvar[iv][i]=dval;
1.226 brouard 10022: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10023: }
10024: strcpy(line,stra);
10025: }/* end loop nqv */
1.136 brouard 10026:
1.223 brouard 10027: /* Covariate values */
1.136 brouard 10028: for (j=ncovcol;j>=1;j--){
10029: cutv(stra, strb,line,' ');
1.223 brouard 10030: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10031: lval=-1;
1.136 brouard 10032: }else{
1.225 brouard 10033: errno=0;
10034: lval=strtol(strb,&endptr,10);
10035: if( strb[0]=='\0' || (*endptr != '\0')){
10036: 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);
10037: 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);
10038: return 1;
10039: }
1.136 brouard 10040: }
10041: if(lval <-1 || lval >1){
1.225 brouard 10042: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10043: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10044: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10045: For example, for multinomial values like 1, 2 and 3,\n \
10046: build V1=0 V2=0 for the reference value (1),\n \
10047: V1=1 V2=0 for (2) \n \
1.136 brouard 10048: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10049: output of IMaCh is often meaningless.\n \
1.136 brouard 10050: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10051: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10052: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10053: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10054: For example, for multinomial values like 1, 2 and 3,\n \
10055: build V1=0 V2=0 for the reference value (1),\n \
10056: V1=1 V2=0 for (2) \n \
1.136 brouard 10057: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10058: output of IMaCh is often meaningless.\n \
1.136 brouard 10059: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10060: return 1;
1.136 brouard 10061: }
10062: covar[j][i]=(double)(lval);
10063: strcpy(line,stra);
10064: }
10065: lstra=strlen(stra);
1.225 brouard 10066:
1.136 brouard 10067: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10068: stratrunc = &(stra[lstra-9]);
10069: num[i]=atol(stratrunc);
10070: }
10071: else
10072: num[i]=atol(stra);
10073: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10074: 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;}*/
10075:
10076: i=i+1;
10077: } /* End loop reading data */
1.225 brouard 10078:
1.136 brouard 10079: *imax=i-1; /* Number of individuals */
10080: fclose(fic);
1.225 brouard 10081:
1.136 brouard 10082: return (0);
1.164 brouard 10083: /* endread: */
1.225 brouard 10084: printf("Exiting readdata: ");
10085: fclose(fic);
10086: return (1);
1.223 brouard 10087: }
1.126 brouard 10088:
1.234 brouard 10089: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10090: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10091: while (*p2 == ' ')
1.234 brouard 10092: p2++;
10093: /* while ((*p1++ = *p2++) !=0) */
10094: /* ; */
10095: /* do */
10096: /* while (*p2 == ' ') */
10097: /* p2++; */
10098: /* while (*p1++ == *p2++); */
10099: *stri=p2;
1.145 brouard 10100: }
10101:
1.330 ! brouard 10102: int decoderesult( char resultline[], int nres)
1.230 brouard 10103: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10104: {
1.235 brouard 10105: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10106: char resultsav[MAXLINE];
1.330 ! brouard 10107: /* int resultmodel[MAXLINE]; */
1.234 brouard 10108: int modelresult[MAXLINE];
1.230 brouard 10109: char stra[80], strb[80], strc[80], strd[80],stre[80];
10110:
1.234 brouard 10111: removefirstspace(&resultline);
1.230 brouard 10112:
10113: if (strstr(resultline,"v") !=0){
10114: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
10115: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
10116: return 1;
10117: }
10118: trimbb(resultsav, resultline);
10119: if (strlen(resultsav) >1){
10120: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
10121: }
1.253 brouard 10122: if(j == 0){ /* Resultline but no = */
10123: TKresult[nres]=0; /* Combination for the nresult and the model */
10124: return (0);
10125: }
1.234 brouard 10126: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.318 brouard 10127: 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 10128: 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 10129: }
10130: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
10131: if(nbocc(resultsav,'=') >1){
1.318 brouard 10132: 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" */
10133: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.234 brouard 10134: }else
10135: cutl(strc,strd,resultsav,'=');
1.318 brouard 10136: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10137:
1.230 brouard 10138: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10139: 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 10140: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10141: /* cptcovsel++; */
10142: if (nbocc(stra,'=') >0)
10143: strcpy(resultsav,stra); /* and analyzes it */
10144: }
1.235 brouard 10145: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 10146: for(k1=1; k1<= cptcovt ;k1++){ /* Loop on model. model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10147: 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 10148: match=0;
1.318 brouard 10149: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10150: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 10151: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10152: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10153: break;
10154: }
10155: }
10156: if(match == 0){
1.310 brouard 10157: printf("Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
10158: fprintf(ficlog,"Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
10159: return 1;
1.234 brouard 10160: }
10161: }
10162: }
1.235 brouard 10163: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 10164: 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 10165: match=0;
1.318 brouard 10166: 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 10167: if(Typevar[k1]==0){ /* Single */
1.237 brouard 10168: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 ! brouard 10169: resultmodel[nres][k1]=k2; /* k1th position in the model equation corresponds to k2th position in the result line. resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 10170: ++match;
10171: }
10172: }
10173: }
10174: if(match == 0){
10175: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
1.310 brouard 10176: fprintf(ficlog,"Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
10177: return 1;
1.234 brouard 10178: }else if(match > 1){
10179: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 10180: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
10181: return 1;
1.234 brouard 10182: }
10183: }
1.235 brouard 10184:
1.234 brouard 10185: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10186: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 ! brouard 10187: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
! 10188: /* should correspond to the combination 6 of dummy: V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 1*1 + 0*2 + 1*4 = 5 + (1offset) = 6*/
! 10189: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 10190: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10191: /* 1 0 0 0 */
10192: /* 2 1 0 0 */
10193: /* 3 0 1 0 */
1.330 ! brouard 10194: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 10195: /* 5 0 0 1 */
1.330 ! brouard 10196: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 10197: /* 7 0 1 1 */
10198: /* 8 1 1 1 */
1.237 brouard 10199: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10200: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10201: /* V5*age V5 known which value for nres? */
10202: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.330 ! brouard 10203: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* loop k1 on position in the model line (excluding product) */
1.235 brouard 10204: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.330 ! brouard 10205: /* k4+1= position in the resultline V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) */
! 10206: /* modelresult[k3]=k1: k3th position in the result line correspond to the k1 position in the model line */
! 10207: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
! 10208: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
! 10209: /* k3 is the position in the nres result line of the k1th variable of the model equation */
! 10210: /* Tvarsel: Name of the variable at the k3th position in the result line Tvarsel[k3]. */
! 10211: /* Tvalsel: Value of the variable at the k3th position in the result line Tvarsel[k3]. */
! 10212: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
! 10213: /* Tvresult[nres][result_position]= id of the dummy variable at the result_position in the nres resultline */
! 10214: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
! 10215: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
! 10216: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
! 10217: k2=(int)Tvarsel[k3]; /* nres=1 k1=2=>k3=1 Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 (V4); k1=3=>k3=2 Tvarsel[2]=3 (V3)*/
! 10218: k+=Tvalsel[k3]*pow(2,k4); /* nres=1 k1=2 Tvalsel[1]=1 (V4=1); k1=3 k3=2 Tvalsel[2]=0 (V3=0) */
! 10219: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
1.237 brouard 10220: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10221: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.330 ! brouard 10222: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 10223: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
10224: k4++;;
10225: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
1.330 ! brouard 10226: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
! 10227: /* Tqvresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
! 10228: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
! 10229: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.318 brouard 10230: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.237 brouard 10231: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10232: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
10233: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 ! brouard 10234: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 10235: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
10236: k4q++;;
1.330 ! brouard 10237: }else{
! 10238: printf("Decodemodel probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
! 10239: fprintf(ficlog,"Decodemodel probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 10240: }
10241: }
1.234 brouard 10242:
1.235 brouard 10243: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 10244: return (0);
10245: }
1.235 brouard 10246:
1.230 brouard 10247: int decodemodel( char model[], int lastobs)
10248: /**< This routine decodes the model and returns:
1.224 brouard 10249: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10250: * - nagesqr = 1 if age*age in the model, otherwise 0.
10251: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10252: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10253: * - cptcovage number of covariates with age*products =2
10254: * - cptcovs number of simple covariates
10255: * - 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
10256: * which is a new column after the 9 (ncovcol) variables.
1.319 brouard 10257: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10258: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10259: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10260: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10261: */
1.319 brouard 10262: /* 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 10263: {
1.238 brouard 10264: int i, j, k, ks, v;
1.227 brouard 10265: int j1, k1, k2, k3, k4;
1.136 brouard 10266: char modelsav[80];
1.145 brouard 10267: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10268: char *strpt;
1.136 brouard 10269:
1.145 brouard 10270: /*removespace(model);*/
1.136 brouard 10271: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10272: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10273: if (strstr(model,"AGE") !=0){
1.192 brouard 10274: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10275: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10276: return 1;
10277: }
1.141 brouard 10278: if (strstr(model,"v") !=0){
10279: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
10280: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
10281: return 1;
10282: }
1.187 brouard 10283: strcpy(modelsav,model);
10284: if ((strpt=strstr(model,"age*age")) !=0){
10285: printf(" strpt=%s, model=%s\n",strpt, model);
10286: if(strpt != model){
1.234 brouard 10287: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10288: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10289: corresponding column of parameters.\n",model);
1.234 brouard 10290: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10291: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10292: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10293: return 1;
1.225 brouard 10294: }
1.187 brouard 10295: nagesqr=1;
10296: if (strstr(model,"+age*age") !=0)
1.234 brouard 10297: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10298: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10299: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10300: else
1.234 brouard 10301: substrchaine(modelsav, model, "age*age");
1.187 brouard 10302: }else
10303: nagesqr=0;
10304: if (strlen(modelsav) >1){
10305: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10306: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10307: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10308: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10309: * cst, age and age*age
10310: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10311: /* including age products which are counted in cptcovage.
10312: * but the covariates which are products must be treated
10313: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10314: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10315: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10316:
10317:
1.187 brouard 10318: /* Design
10319: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10320: * < ncovcol=8 >
10321: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10322: * k= 1 2 3 4 5 6 7 8
10323: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10324: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10325: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10326: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10327: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10328: * Tage[++cptcovage]=k
10329: * if products, new covar are created after ncovcol with k1
10330: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10331: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10332: * 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
10333: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10334: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10335: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10336: * < ncovcol=8 >
10337: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10338: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10339: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10340: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10341: * p Tprod[1]@2={ 6, 5}
10342: *p Tvard[1][1]@4= {7, 8, 5, 6}
10343: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10344: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10345: *How to reorganize? Tvars(orted)
1.187 brouard 10346: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10347: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
10348: * {2, 1, 4, 8, 5, 6, 3, 7}
10349: * Struct []
10350: */
1.225 brouard 10351:
1.187 brouard 10352: /* This loop fills the array Tvar from the string 'model'.*/
10353: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
10354: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
10355: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
10356: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
10357: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
10358: /* k=1 Tvar[1]=2 (from V2) */
10359: /* k=5 Tvar[5] */
10360: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 10361: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 10362: /* } */
1.198 brouard 10363: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 10364: /*
10365: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 10366: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
10367: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
10368: }
1.187 brouard 10369: cptcovage=0;
1.319 brouard 10370: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
10371: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
10372: 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" */
10373: if (nbocc(modelsav,'+')==0)
10374: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 10375: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
10376: /*scanf("%d",i);*/
1.319 brouard 10377: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
10378: 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 10379: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
10380: /* covar is not filled and then is empty */
10381: cptcovprod--;
10382: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 10383: 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 10384: Typevar[k]=1; /* 1 for age product */
1.319 brouard 10385: cptcovage++; /* Counts the number of covariates which include age as a product */
10386: 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 10387: /*printf("stre=%s ", stre);*/
10388: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
10389: cptcovprod--;
10390: cutl(stre,strb,strc,'V');
10391: Tvar[k]=atoi(stre);
10392: Typevar[k]=1; /* 1 for age product */
10393: cptcovage++;
10394: Tage[cptcovage]=k;
10395: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
10396: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
10397: cptcovn++;
10398: cptcovprodnoage++;k1++;
10399: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10400: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10401: because this model-covariate is a construction we invent a new column
10402: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.319 brouard 10403: If already ncovcol=4 and model=V2 + V1 +V1*V4 +age*V3 +V3*V2
10404: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
10405: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=4 etc */
1.234 brouard 10406: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10407: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10408: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 10409: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 10410: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 ! brouard 10411: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 10412: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 ! brouard 10413: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 10414: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10415: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10416: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10417: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10418: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10419: for (i=1; i<=lastobs;i++){
10420: /* Computes the new covariate which is a product of
10421: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10422: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10423: }
10424: } /* End age is not in the model */
10425: } /* End if model includes a product */
1.319 brouard 10426: else { /* not a product */
1.234 brouard 10427: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10428: /* scanf("%d",i);*/
10429: cutl(strd,strc,strb,'V');
10430: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10431: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10432: Tvar[k]=atoi(strd);
10433: Typevar[k]=0; /* 0 for simple covariates */
10434: }
10435: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10436: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10437: scanf("%d",i);*/
1.187 brouard 10438: } /* end of loop + on total covariates */
10439: } /* end if strlen(modelsave == 0) age*age might exist */
10440: } /* end if strlen(model == 0) */
1.136 brouard 10441:
10442: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10443: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10444:
1.136 brouard 10445: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10446: printf("cptcovprod=%d ", cptcovprod);
10447: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10448: scanf("%d ",i);*/
10449:
10450:
1.230 brouard 10451: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10452: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10453: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10454: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10455: k = 1 2 3 4 5 6 7 8 9
10456: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 10457: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 10458: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10459: Dummy[k] 1 0 0 0 3 1 1 2 3
10460: Tmodelind[combination of covar]=k;
1.225 brouard 10461: */
10462: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10463: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10464: /* 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 10465: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 10466: printf("Model=1+age+%s\n\
1.227 brouard 10467: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10468: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10469: 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 10470: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 10471: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10472: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10473: 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 10474: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10475: 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 */
10476: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10477: Fixed[k]= 0;
10478: Dummy[k]= 0;
1.225 brouard 10479: ncoveff++;
1.232 brouard 10480: ncovf++;
1.234 brouard 10481: nsd++;
10482: modell[k].maintype= FTYPE;
10483: TvarsD[nsd]=Tvar[k];
10484: TvarsDind[nsd]=k;
1.330 ! brouard 10485: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 10486: TvarF[ncovf]=Tvar[k];
10487: TvarFind[ncovf]=k;
10488: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10489: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10490: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10491: Fixed[k]= 0;
10492: Dummy[k]= 0;
10493: ncoveff++;
10494: ncovf++;
10495: modell[k].maintype= FTYPE;
10496: TvarF[ncovf]=Tvar[k];
1.330 ! brouard 10497: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 10498: TvarFind[ncovf]=k;
1.230 brouard 10499: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10500: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10501: }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 10502: Fixed[k]= 0;
10503: Dummy[k]= 1;
1.230 brouard 10504: nqfveff++;
1.234 brouard 10505: modell[k].maintype= FTYPE;
10506: modell[k].subtype= FQ;
10507: nsq++;
10508: TvarsQ[nsq]=Tvar[k];
10509: TvarsQind[nsq]=k;
1.232 brouard 10510: ncovf++;
1.234 brouard 10511: TvarF[ncovf]=Tvar[k];
10512: TvarFind[ncovf]=k;
1.231 brouard 10513: 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 10514: 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 10515: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10516: Fixed[k]= 1;
10517: Dummy[k]= 0;
1.225 brouard 10518: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10519: modell[k].maintype= VTYPE;
10520: modell[k].subtype= VD;
10521: nsd++;
10522: TvarsD[nsd]=Tvar[k];
10523: TvarsDind[nsd]=k;
1.330 ! brouard 10524: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 10525: ncovv++; /* Only simple time varying variables */
10526: TvarV[ncovv]=Tvar[k];
1.242 brouard 10527: 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 10528: 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 */
10529: 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 10530: 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);
10531: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10532: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10533: Fixed[k]= 1;
10534: Dummy[k]= 1;
10535: nqtveff++;
10536: modell[k].maintype= VTYPE;
10537: modell[k].subtype= VQ;
10538: ncovv++; /* Only simple time varying variables */
10539: nsq++;
1.319 brouard 10540: TvarsQ[nsq]=Tvar[k]; /* k=1 Tvar=5 nsq=1 TvarsQ[1]=5 */
1.234 brouard 10541: TvarsQind[nsq]=k;
10542: TvarV[ncovv]=Tvar[k];
1.242 brouard 10543: 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 10544: 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 */
10545: 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 10546: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10547: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10548: 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 10549: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10550: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10551: ncova++;
10552: TvarA[ncova]=Tvar[k];
10553: TvarAind[ncova]=k;
1.231 brouard 10554: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10555: Fixed[k]= 2;
10556: Dummy[k]= 2;
10557: modell[k].maintype= ATYPE;
10558: modell[k].subtype= APFD;
10559: /* ncoveff++; */
1.227 brouard 10560: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10561: Fixed[k]= 2;
10562: Dummy[k]= 3;
10563: modell[k].maintype= ATYPE;
10564: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10565: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10566: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10567: Fixed[k]= 3;
10568: Dummy[k]= 2;
10569: modell[k].maintype= ATYPE;
10570: modell[k].subtype= APVD; /* Product age * varying dummy */
10571: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10572: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10573: Fixed[k]= 3;
10574: Dummy[k]= 3;
10575: modell[k].maintype= ATYPE;
10576: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10577: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10578: }
10579: }else if (Typevar[k] == 2) { /* product without age */
10580: k1=Tposprod[k];
10581: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10582: if(Tvard[k1][2] <=ncovcol){
10583: Fixed[k]= 1;
10584: Dummy[k]= 0;
10585: modell[k].maintype= FTYPE;
10586: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10587: ncovf++; /* Fixed variables without age */
10588: TvarF[ncovf]=Tvar[k];
10589: TvarFind[ncovf]=k;
10590: }else if(Tvard[k1][2] <=ncovcol+nqv){
10591: Fixed[k]= 0; /* or 2 ?*/
10592: Dummy[k]= 1;
10593: modell[k].maintype= FTYPE;
10594: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10595: ncovf++; /* Varying variables without age */
10596: TvarF[ncovf]=Tvar[k];
10597: TvarFind[ncovf]=k;
10598: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10599: Fixed[k]= 1;
10600: Dummy[k]= 0;
10601: modell[k].maintype= VTYPE;
10602: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10603: ncovv++; /* Varying variables without age */
10604: TvarV[ncovv]=Tvar[k];
10605: TvarVind[ncovv]=k;
10606: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10607: Fixed[k]= 1;
10608: Dummy[k]= 1;
10609: modell[k].maintype= VTYPE;
10610: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10611: ncovv++; /* Varying variables without age */
10612: TvarV[ncovv]=Tvar[k];
10613: TvarVind[ncovv]=k;
10614: }
1.227 brouard 10615: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10616: if(Tvard[k1][2] <=ncovcol){
10617: Fixed[k]= 0; /* or 2 ?*/
10618: Dummy[k]= 1;
10619: modell[k].maintype= FTYPE;
10620: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10621: ncovf++; /* Fixed variables without age */
10622: TvarF[ncovf]=Tvar[k];
10623: TvarFind[ncovf]=k;
10624: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10625: Fixed[k]= 1;
10626: Dummy[k]= 1;
10627: modell[k].maintype= VTYPE;
10628: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10629: ncovv++; /* Varying variables without age */
10630: TvarV[ncovv]=Tvar[k];
10631: TvarVind[ncovv]=k;
10632: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10633: Fixed[k]= 1;
10634: Dummy[k]= 1;
10635: modell[k].maintype= VTYPE;
10636: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10637: ncovv++; /* Varying variables without age */
10638: TvarV[ncovv]=Tvar[k];
10639: TvarVind[ncovv]=k;
10640: ncovv++; /* Varying variables without age */
10641: TvarV[ncovv]=Tvar[k];
10642: TvarVind[ncovv]=k;
10643: }
1.227 brouard 10644: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10645: if(Tvard[k1][2] <=ncovcol){
10646: Fixed[k]= 1;
10647: Dummy[k]= 1;
10648: modell[k].maintype= VTYPE;
10649: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10650: ncovv++; /* Varying variables without age */
10651: TvarV[ncovv]=Tvar[k];
10652: TvarVind[ncovv]=k;
10653: }else if(Tvard[k1][2] <=ncovcol+nqv){
10654: Fixed[k]= 1;
10655: Dummy[k]= 1;
10656: modell[k].maintype= VTYPE;
10657: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10658: ncovv++; /* Varying variables without age */
10659: TvarV[ncovv]=Tvar[k];
10660: TvarVind[ncovv]=k;
10661: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10662: Fixed[k]= 1;
10663: Dummy[k]= 0;
10664: modell[k].maintype= VTYPE;
10665: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10666: ncovv++; /* Varying variables without age */
10667: TvarV[ncovv]=Tvar[k];
10668: TvarVind[ncovv]=k;
10669: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10670: Fixed[k]= 1;
10671: Dummy[k]= 1;
10672: modell[k].maintype= VTYPE;
10673: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10674: ncovv++; /* Varying variables without age */
10675: TvarV[ncovv]=Tvar[k];
10676: TvarVind[ncovv]=k;
10677: }
1.227 brouard 10678: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10679: if(Tvard[k1][2] <=ncovcol){
10680: Fixed[k]= 1;
10681: Dummy[k]= 1;
10682: modell[k].maintype= VTYPE;
10683: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10684: ncovv++; /* Varying variables without age */
10685: TvarV[ncovv]=Tvar[k];
10686: TvarVind[ncovv]=k;
10687: }else if(Tvard[k1][2] <=ncovcol+nqv){
10688: Fixed[k]= 1;
10689: Dummy[k]= 1;
10690: modell[k].maintype= VTYPE;
10691: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10692: ncovv++; /* Varying variables without age */
10693: TvarV[ncovv]=Tvar[k];
10694: TvarVind[ncovv]=k;
10695: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10696: Fixed[k]= 1;
10697: Dummy[k]= 1;
10698: modell[k].maintype= VTYPE;
10699: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10700: ncovv++; /* Varying variables without age */
10701: TvarV[ncovv]=Tvar[k];
10702: TvarVind[ncovv]=k;
10703: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10704: Fixed[k]= 1;
10705: Dummy[k]= 1;
10706: modell[k].maintype= VTYPE;
10707: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10708: ncovv++; /* Varying variables without age */
10709: TvarV[ncovv]=Tvar[k];
10710: TvarVind[ncovv]=k;
10711: }
1.227 brouard 10712: }else{
1.240 brouard 10713: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10714: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10715: } /*end k1*/
1.225 brouard 10716: }else{
1.226 brouard 10717: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10718: 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 10719: }
1.227 brouard 10720: 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 10721: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10722: 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]);
10723: }
10724: /* Searching for doublons in the model */
10725: for(k1=1; k1<= cptcovt;k1++){
10726: for(k2=1; k2 <k1;k2++){
1.285 brouard 10727: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10728: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10729: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10730: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10731: 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]);
10732: 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 10733: return(1);
10734: }
10735: }else if (Typevar[k1] ==2){
10736: k3=Tposprod[k1];
10737: k4=Tposprod[k2];
10738: 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])) ){
10739: 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]]);
10740: 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);
10741: return(1);
10742: }
10743: }
1.227 brouard 10744: }
10745: }
1.225 brouard 10746: }
10747: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10748: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10749: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10750: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10751: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10752: /*endread:*/
1.225 brouard 10753: printf("Exiting decodemodel: ");
10754: return (1);
1.136 brouard 10755: }
10756:
1.169 brouard 10757: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10758: {/* Check ages at death */
1.136 brouard 10759: int i, m;
1.218 brouard 10760: int firstone=0;
10761:
1.136 brouard 10762: for (i=1; i<=imx; i++) {
10763: for(m=2; (m<= maxwav); m++) {
10764: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10765: anint[m][i]=9999;
1.216 brouard 10766: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10767: s[m][i]=-1;
1.136 brouard 10768: }
10769: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10770: *nberr = *nberr + 1;
1.218 brouard 10771: if(firstone == 0){
10772: firstone=1;
1.260 brouard 10773: 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 10774: }
1.262 brouard 10775: 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 10776: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10777: }
10778: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10779: (*nberr)++;
1.259 brouard 10780: 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 10781: 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 10782: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10783: }
10784: }
10785: }
10786:
10787: for (i=1; i<=imx; i++) {
10788: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10789: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10790: 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 10791: if (s[m][i] >= nlstate+1) {
1.169 brouard 10792: if(agedc[i]>0){
10793: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10794: agev[m][i]=agedc[i];
1.214 brouard 10795: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10796: }else {
1.136 brouard 10797: if ((int)andc[i]!=9999){
10798: nbwarn++;
10799: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10800: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10801: agev[m][i]=-1;
10802: }
10803: }
1.169 brouard 10804: } /* agedc > 0 */
1.214 brouard 10805: } /* end if */
1.136 brouard 10806: else if(s[m][i] !=9){ /* Standard case, age in fractional
10807: years but with the precision of a month */
10808: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10809: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10810: agev[m][i]=1;
10811: else if(agev[m][i] < *agemin){
10812: *agemin=agev[m][i];
10813: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10814: }
10815: else if(agev[m][i] >*agemax){
10816: *agemax=agev[m][i];
1.156 brouard 10817: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10818: }
10819: /*agev[m][i]=anint[m][i]-annais[i];*/
10820: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10821: } /* en if 9*/
1.136 brouard 10822: else { /* =9 */
1.214 brouard 10823: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10824: agev[m][i]=1;
10825: s[m][i]=-1;
10826: }
10827: }
1.214 brouard 10828: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10829: agev[m][i]=1;
1.214 brouard 10830: else{
10831: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10832: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10833: agev[m][i]=0;
10834: }
10835: } /* End for lastpass */
10836: }
1.136 brouard 10837:
10838: for (i=1; i<=imx; i++) {
10839: for(m=firstpass; (m<=lastpass); m++){
10840: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10841: (*nberr)++;
1.136 brouard 10842: 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);
10843: 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);
10844: return 1;
10845: }
10846: }
10847: }
10848:
10849: /*for (i=1; i<=imx; i++){
10850: for (m=firstpass; (m<lastpass); m++){
10851: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10852: }
10853:
10854: }*/
10855:
10856:
1.139 brouard 10857: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10858: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10859:
10860: return (0);
1.164 brouard 10861: /* endread:*/
1.136 brouard 10862: printf("Exiting calandcheckages: ");
10863: return (1);
10864: }
10865:
1.172 brouard 10866: #if defined(_MSC_VER)
10867: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10868: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10869: //#include "stdafx.h"
10870: //#include <stdio.h>
10871: //#include <tchar.h>
10872: //#include <windows.h>
10873: //#include <iostream>
10874: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10875:
10876: LPFN_ISWOW64PROCESS fnIsWow64Process;
10877:
10878: BOOL IsWow64()
10879: {
10880: BOOL bIsWow64 = FALSE;
10881:
10882: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10883: // (HANDLE, PBOOL);
10884:
10885: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10886:
10887: HMODULE module = GetModuleHandle(_T("kernel32"));
10888: const char funcName[] = "IsWow64Process";
10889: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10890: GetProcAddress(module, funcName);
10891:
10892: if (NULL != fnIsWow64Process)
10893: {
10894: if (!fnIsWow64Process(GetCurrentProcess(),
10895: &bIsWow64))
10896: //throw std::exception("Unknown error");
10897: printf("Unknown error\n");
10898: }
10899: return bIsWow64 != FALSE;
10900: }
10901: #endif
1.177 brouard 10902:
1.191 brouard 10903: void syscompilerinfo(int logged)
1.292 brouard 10904: {
10905: #include <stdint.h>
10906:
10907: /* #include "syscompilerinfo.h"*/
1.185 brouard 10908: /* command line Intel compiler 32bit windows, XP compatible:*/
10909: /* /GS /W3 /Gy
10910: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10911: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10912: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10913: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10914: */
10915: /* 64 bits */
1.185 brouard 10916: /*
10917: /GS /W3 /Gy
10918: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10919: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10920: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10921: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10922: /* Optimization are useless and O3 is slower than O2 */
10923: /*
10924: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10925: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10926: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10927: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10928: */
1.186 brouard 10929: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10930: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10931: /PDB:"visual studio
10932: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10933: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10934: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10935: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10936: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10937: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10938: uiAccess='false'"
10939: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10940: /NOLOGO /TLBID:1
10941: */
1.292 brouard 10942:
10943:
1.177 brouard 10944: #if defined __INTEL_COMPILER
1.178 brouard 10945: #if defined(__GNUC__)
10946: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10947: #endif
1.177 brouard 10948: #elif defined(__GNUC__)
1.179 brouard 10949: #ifndef __APPLE__
1.174 brouard 10950: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10951: #endif
1.177 brouard 10952: struct utsname sysInfo;
1.178 brouard 10953: int cross = CROSS;
10954: if (cross){
10955: printf("Cross-");
1.191 brouard 10956: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10957: }
1.174 brouard 10958: #endif
10959:
1.191 brouard 10960: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10961: #if defined(__clang__)
1.191 brouard 10962: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10963: #endif
10964: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10965: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10966: #endif
10967: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10968: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10969: #endif
10970: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10971: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10972: #endif
10973: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10974: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10975: #endif
10976: #if defined(_MSC_VER)
1.191 brouard 10977: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10978: #endif
10979: #if defined(__PGI)
1.191 brouard 10980: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10981: #endif
10982: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10983: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10984: #endif
1.191 brouard 10985: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10986:
1.167 brouard 10987: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10988: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10989: // Windows (x64 and x86)
1.191 brouard 10990: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10991: #elif __unix__ // all unices, not all compilers
10992: // Unix
1.191 brouard 10993: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10994: #elif __linux__
10995: // linux
1.191 brouard 10996: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10997: #elif __APPLE__
1.174 brouard 10998: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10999: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11000: #endif
11001:
11002: /* __MINGW32__ */
11003: /* __CYGWIN__ */
11004: /* __MINGW64__ */
11005: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11006: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11007: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11008: /* _WIN64 // Defined for applications for Win64. */
11009: /* _M_X64 // Defined for compilations that target x64 processors. */
11010: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11011:
1.167 brouard 11012: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11013: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11014: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11015: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11016: #else
1.191 brouard 11017: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11018: #endif
11019:
1.169 brouard 11020: #if defined(__GNUC__)
11021: # if defined(__GNUC_PATCHLEVEL__)
11022: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11023: + __GNUC_MINOR__ * 100 \
11024: + __GNUC_PATCHLEVEL__)
11025: # else
11026: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11027: + __GNUC_MINOR__ * 100)
11028: # endif
1.174 brouard 11029: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11030: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11031:
11032: if (uname(&sysInfo) != -1) {
11033: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11034: 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 11035: }
11036: else
11037: perror("uname() error");
1.179 brouard 11038: //#ifndef __INTEL_COMPILER
11039: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11040: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11041: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11042: #endif
1.169 brouard 11043: #endif
1.172 brouard 11044:
1.286 brouard 11045: // void main ()
1.172 brouard 11046: // {
1.169 brouard 11047: #if defined(_MSC_VER)
1.174 brouard 11048: if (IsWow64()){
1.191 brouard 11049: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11050: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11051: }
11052: else{
1.191 brouard 11053: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11054: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11055: }
1.172 brouard 11056: // printf("\nPress Enter to continue...");
11057: // getchar();
11058: // }
11059:
1.169 brouard 11060: #endif
11061:
1.167 brouard 11062:
1.219 brouard 11063: }
1.136 brouard 11064:
1.219 brouard 11065: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11066: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 11067: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11068: /* double ftolpl = 1.e-10; */
1.180 brouard 11069: double age, agebase, agelim;
1.203 brouard 11070: double tot;
1.180 brouard 11071:
1.202 brouard 11072: strcpy(filerespl,"PL_");
11073: strcat(filerespl,fileresu);
11074: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11075: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11076: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11077: }
1.288 brouard 11078: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11079: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11080: pstamp(ficrespl);
1.288 brouard 11081: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11082: fprintf(ficrespl,"#Age ");
11083: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11084: fprintf(ficrespl,"\n");
1.180 brouard 11085:
1.219 brouard 11086: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11087:
1.219 brouard 11088: agebase=ageminpar;
11089: agelim=agemaxpar;
1.180 brouard 11090:
1.227 brouard 11091: /* i1=pow(2,ncoveff); */
1.234 brouard 11092: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11093: if (cptcovn < 1){i1=1;}
1.180 brouard 11094:
1.238 brouard 11095: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
11096: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 11097: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11098: continue;
1.235 brouard 11099:
1.238 brouard 11100: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11101: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11102: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11103: /* k=k+1; */
11104: /* to clean */
11105: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
11106: fprintf(ficrespl,"#******");
11107: printf("#******");
11108: fprintf(ficlog,"#******");
11109: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
1.330 ! brouard 11110: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /* Here problem for varying dummy*/
! 11111: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 11112: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.238 brouard 11113: }
11114: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11115: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11116: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11117: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11118: }
11119: fprintf(ficrespl,"******\n");
11120: printf("******\n");
11121: fprintf(ficlog,"******\n");
11122: if(invalidvarcomb[k]){
11123: printf("\nCombination (%d) ignored because no case \n",k);
11124: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
11125: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
11126: continue;
11127: }
1.219 brouard 11128:
1.238 brouard 11129: fprintf(ficrespl,"#Age ");
11130: for(j=1;j<=cptcoveff;j++) {
1.330 ! brouard 11131: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.238 brouard 11132: }
11133: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
11134: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 11135:
1.238 brouard 11136: for (age=agebase; age<=agelim; age++){
11137: /* for (age=agebase; age<=agebase; age++){ */
11138: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
11139: fprintf(ficrespl,"%.0f ",age );
11140: for(j=1;j<=cptcoveff;j++)
1.330 ! brouard 11141: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.238 brouard 11142: tot=0.;
11143: for(i=1; i<=nlstate;i++){
11144: tot += prlim[i][i];
11145: fprintf(ficrespl," %.5f", prlim[i][i]);
11146: }
11147: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
11148: } /* Age */
11149: /* was end of cptcod */
11150: } /* cptcov */
11151: } /* nres */
1.219 brouard 11152: return 0;
1.180 brouard 11153: }
11154:
1.218 brouard 11155: 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 11156: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 11157:
11158: /* Computes the back prevalence limit for any combination of covariate values
11159: * at any age between ageminpar and agemaxpar
11160: */
1.235 brouard 11161: int i, j, k, i1, nres=0 ;
1.217 brouard 11162: /* double ftolpl = 1.e-10; */
11163: double age, agebase, agelim;
11164: double tot;
1.218 brouard 11165: /* double ***mobaverage; */
11166: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 11167:
11168: strcpy(fileresplb,"PLB_");
11169: strcat(fileresplb,fileresu);
11170: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 11171: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
11172: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 11173: }
1.288 brouard 11174: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
11175: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 11176: pstamp(ficresplb);
1.288 brouard 11177: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 11178: fprintf(ficresplb,"#Age ");
11179: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
11180: fprintf(ficresplb,"\n");
11181:
1.218 brouard 11182:
11183: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
11184:
11185: agebase=ageminpar;
11186: agelim=agemaxpar;
11187:
11188:
1.227 brouard 11189: i1=pow(2,cptcoveff);
1.218 brouard 11190: if (cptcovn < 1){i1=1;}
1.227 brouard 11191:
1.238 brouard 11192: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11193: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11194: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11195: continue;
11196: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
11197: fprintf(ficresplb,"#******");
11198: printf("#******");
11199: fprintf(ficlog,"#******");
11200: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
1.330 ! brouard 11201: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 11202: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 11203: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.238 brouard 11204: }
11205: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11206: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11207: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11208: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11209: }
11210: fprintf(ficresplb,"******\n");
11211: printf("******\n");
11212: fprintf(ficlog,"******\n");
11213: if(invalidvarcomb[k]){
11214: printf("\nCombination (%d) ignored because no cases \n",k);
11215: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
11216: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
11217: continue;
11218: }
1.218 brouard 11219:
1.238 brouard 11220: fprintf(ficresplb,"#Age ");
11221: for(j=1;j<=cptcoveff;j++) {
1.330 ! brouard 11222: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.238 brouard 11223: }
11224: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
11225: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 11226:
11227:
1.238 brouard 11228: for (age=agebase; age<=agelim; age++){
11229: /* for (age=agebase; age<=agebase; age++){ */
11230: if(mobilavproj > 0){
11231: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11232: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11233: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11234: }else if (mobilavproj == 0){
11235: 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);
11236: 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);
11237: exit(1);
11238: }else{
11239: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11240: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11241: /* printf("TOTOT\n"); */
11242: /* exit(1); */
1.238 brouard 11243: }
11244: fprintf(ficresplb,"%.0f ",age );
11245: for(j=1;j<=cptcoveff;j++)
1.330 ! brouard 11246: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.238 brouard 11247: tot=0.;
11248: for(i=1; i<=nlstate;i++){
11249: tot += bprlim[i][i];
11250: fprintf(ficresplb," %.5f", bprlim[i][i]);
11251: }
11252: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11253: } /* Age */
11254: /* was end of cptcod */
1.255 brouard 11255: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 11256: } /* end of any combination */
11257: } /* end of nres */
1.218 brouard 11258: /* hBijx(p, bage, fage); */
11259: /* fclose(ficrespijb); */
11260:
11261: return 0;
1.217 brouard 11262: }
1.218 brouard 11263:
1.180 brouard 11264: int hPijx(double *p, int bage, int fage){
11265: /*------------- h Pij x at various ages ------------*/
11266:
11267: int stepsize;
11268: int agelim;
11269: int hstepm;
11270: int nhstepm;
1.235 brouard 11271: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11272:
11273: double agedeb;
11274: double ***p3mat;
11275:
1.201 brouard 11276: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 11277: if((ficrespij=fopen(filerespij,"w"))==NULL) {
11278: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
11279: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
11280: }
11281: printf("Computing pij: result on file '%s' \n", filerespij);
11282: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
11283:
11284: stepsize=(int) (stepm+YEARM-1)/YEARM;
11285: /*if (stepm<=24) stepsize=2;*/
11286:
11287: agelim=AGESUP;
11288: hstepm=stepsize*YEARM; /* Every year of age */
11289: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 11290:
1.180 brouard 11291: /* hstepm=1; aff par mois*/
11292: pstamp(ficrespij);
11293: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 11294: i1= pow(2,cptcoveff);
1.218 brouard 11295: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11296: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11297: /* k=k+1; */
1.235 brouard 11298: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11299: for(k=1; k<=i1;k++){
1.253 brouard 11300: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11301: continue;
1.183 brouard 11302: fprintf(ficrespij,"\n#****** ");
1.227 brouard 11303: for(j=1;j<=cptcoveff;j++)
1.330 ! brouard 11304: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.235 brouard 11305: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11306: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11307: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11308: }
1.183 brouard 11309: fprintf(ficrespij,"******\n");
11310:
11311: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
11312: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
11313: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
11314:
11315: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 11316:
1.183 brouard 11317: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11318: oldm=oldms;savm=savms;
1.235 brouard 11319: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 11320: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
11321: for(i=1; i<=nlstate;i++)
11322: for(j=1; j<=nlstate+ndeath;j++)
11323: fprintf(ficrespij," %1d-%1d",i,j);
11324: fprintf(ficrespij,"\n");
11325: for (h=0; h<=nhstepm; h++){
11326: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11327: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 11328: for(i=1; i<=nlstate;i++)
11329: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 11330: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 11331: fprintf(ficrespij,"\n");
11332: }
1.183 brouard 11333: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11334: fprintf(ficrespij,"\n");
11335: }
1.180 brouard 11336: /*}*/
11337: }
1.218 brouard 11338: return 0;
1.180 brouard 11339: }
1.218 brouard 11340:
11341: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 11342: /*------------- h Bij x at various ages ------------*/
11343:
11344: int stepsize;
1.218 brouard 11345: /* int agelim; */
11346: int ageminl;
1.217 brouard 11347: int hstepm;
11348: int nhstepm;
1.238 brouard 11349: int h, i, i1, j, k, nres;
1.218 brouard 11350:
1.217 brouard 11351: double agedeb;
11352: double ***p3mat;
1.218 brouard 11353:
11354: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
11355: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
11356: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11357: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11358: }
11359: printf("Computing pij back: result on file '%s' \n", filerespijb);
11360: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
11361:
11362: stepsize=(int) (stepm+YEARM-1)/YEARM;
11363: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 11364:
1.218 brouard 11365: /* agelim=AGESUP; */
1.289 brouard 11366: ageminl=AGEINF; /* was 30 */
1.218 brouard 11367: hstepm=stepsize*YEARM; /* Every year of age */
11368: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11369:
11370: /* hstepm=1; aff par mois*/
11371: pstamp(ficrespijb);
1.255 brouard 11372: 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 11373: i1= pow(2,cptcoveff);
1.218 brouard 11374: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11375: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11376: /* k=k+1; */
1.238 brouard 11377: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11378: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11379: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11380: continue;
11381: fprintf(ficrespijb,"\n#****** ");
11382: for(j=1;j<=cptcoveff;j++)
1.330 ! brouard 11383: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.238 brouard 11384: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11385: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11386: }
11387: fprintf(ficrespijb,"******\n");
1.264 brouard 11388: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 11389: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
11390: continue;
11391: }
11392:
11393: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
11394: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
11395: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 11396: 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 */
11397: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 11398:
11399: /* nhstepm=nhstepm*YEARM; aff par mois*/
11400:
1.266 brouard 11401: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
11402: /* and memory limitations if stepm is small */
11403:
1.238 brouard 11404: /* oldm=oldms;savm=savms; */
11405: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.325 brouard 11406: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
1.238 brouard 11407: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 11408: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 11409: for(i=1; i<=nlstate;i++)
11410: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 11411: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 11412: fprintf(ficrespijb,"\n");
1.238 brouard 11413: for (h=0; h<=nhstepm; h++){
11414: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11415: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
11416: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
11417: for(i=1; i<=nlstate;i++)
11418: for(j=1; j<=nlstate+ndeath;j++)
1.325 brouard 11419: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.238 brouard 11420: fprintf(ficrespijb,"\n");
11421: }
11422: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11423: fprintf(ficrespijb,"\n");
11424: } /* end age deb */
11425: } /* end combination */
11426: } /* end nres */
1.218 brouard 11427: return 0;
11428: } /* hBijx */
1.217 brouard 11429:
1.180 brouard 11430:
1.136 brouard 11431: /***********************************************/
11432: /**************** Main Program *****************/
11433: /***********************************************/
11434:
11435: int main(int argc, char *argv[])
11436: {
11437: #ifdef GSL
11438: const gsl_multimin_fminimizer_type *T;
11439: size_t iteri = 0, it;
11440: int rval = GSL_CONTINUE;
11441: int status = GSL_SUCCESS;
11442: double ssval;
11443: #endif
11444: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11445: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11446: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11447: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11448: int jj, ll, li, lj, lk;
1.136 brouard 11449: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11450: int num_filled;
1.136 brouard 11451: int itimes;
11452: int NDIM=2;
11453: int vpopbased=0;
1.235 brouard 11454: int nres=0;
1.258 brouard 11455: int endishere=0;
1.277 brouard 11456: int noffset=0;
1.274 brouard 11457: int ncurrv=0; /* Temporary variable */
11458:
1.164 brouard 11459: char ca[32], cb[32];
1.136 brouard 11460: /* FILE *fichtm; *//* Html File */
11461: /* FILE *ficgp;*/ /*Gnuplot File */
11462: struct stat info;
1.191 brouard 11463: double agedeb=0.;
1.194 brouard 11464:
11465: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11466: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11467:
1.165 brouard 11468: double fret;
1.191 brouard 11469: double dum=0.; /* Dummy variable */
1.136 brouard 11470: double ***p3mat;
1.218 brouard 11471: /* double ***mobaverage; */
1.319 brouard 11472: double wald;
1.164 brouard 11473:
11474: char line[MAXLINE];
1.197 brouard 11475: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11476:
1.234 brouard 11477: char modeltemp[MAXLINE];
1.230 brouard 11478: char resultline[MAXLINE];
11479:
1.136 brouard 11480: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11481: char *tok, *val; /* pathtot */
1.290 brouard 11482: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 11483: int c, h , cpt, c2;
1.191 brouard 11484: int jl=0;
11485: int i1, j1, jk, stepsize=0;
1.194 brouard 11486: int count=0;
11487:
1.164 brouard 11488: int *tab;
1.136 brouard 11489: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11490: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11491: /* double anprojf, mprojf, jprojf; */
11492: /* double jintmean,mintmean,aintmean; */
11493: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11494: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11495: double yrfproj= 10.0; /* Number of years of forward projections */
11496: double yrbproj= 10.0; /* Number of years of backward projections */
11497: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11498: int mobilav=0,popforecast=0;
1.191 brouard 11499: int hstepm=0, nhstepm=0;
1.136 brouard 11500: int agemortsup;
11501: float sumlpop=0.;
11502: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11503: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11504:
1.191 brouard 11505: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11506: double ftolpl=FTOL;
11507: double **prlim;
1.217 brouard 11508: double **bprlim;
1.317 brouard 11509: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
11510: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11511: double ***paramstart; /* Matrix of starting parameter values */
11512: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11513: double **matcov; /* Matrix of covariance */
1.203 brouard 11514: double **hess; /* Hessian matrix */
1.136 brouard 11515: double ***delti3; /* Scale */
11516: double *delti; /* Scale */
11517: double ***eij, ***vareij;
11518: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11519:
1.136 brouard 11520: double *epj, vepp;
1.164 brouard 11521:
1.273 brouard 11522: double dateprev1, dateprev2;
1.296 brouard 11523: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11524: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11525:
1.217 brouard 11526:
1.136 brouard 11527: double **ximort;
1.145 brouard 11528: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11529: int *dcwave;
11530:
1.164 brouard 11531: char z[1]="c";
1.136 brouard 11532:
11533: /*char *strt;*/
11534: char strtend[80];
1.126 brouard 11535:
1.164 brouard 11536:
1.126 brouard 11537: /* setlocale (LC_ALL, ""); */
11538: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11539: /* textdomain (PACKAGE); */
11540: /* setlocale (LC_CTYPE, ""); */
11541: /* setlocale (LC_MESSAGES, ""); */
11542:
11543: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11544: rstart_time = time(NULL);
11545: /* (void) gettimeofday(&start_time,&tzp);*/
11546: start_time = *localtime(&rstart_time);
1.126 brouard 11547: curr_time=start_time;
1.157 brouard 11548: /*tml = *localtime(&start_time.tm_sec);*/
11549: /* strcpy(strstart,asctime(&tml)); */
11550: strcpy(strstart,asctime(&start_time));
1.126 brouard 11551:
11552: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11553: /* tp.tm_sec = tp.tm_sec +86400; */
11554: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11555: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11556: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11557: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11558: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11559: /* strt=asctime(&tmg); */
11560: /* printf("Time(after) =%s",strstart); */
11561: /* (void) time (&time_value);
11562: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11563: * tm = *localtime(&time_value);
11564: * strstart=asctime(&tm);
11565: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11566: */
11567:
11568: nberr=0; /* Number of errors and warnings */
11569: nbwarn=0;
1.184 brouard 11570: #ifdef WIN32
11571: _getcwd(pathcd, size);
11572: #else
1.126 brouard 11573: getcwd(pathcd, size);
1.184 brouard 11574: #endif
1.191 brouard 11575: syscompilerinfo(0);
1.196 brouard 11576: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11577: if(argc <=1){
11578: printf("\nEnter the parameter file name: ");
1.205 brouard 11579: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11580: printf("ERROR Empty parameter file name\n");
11581: goto end;
11582: }
1.126 brouard 11583: i=strlen(pathr);
11584: if(pathr[i-1]=='\n')
11585: pathr[i-1]='\0';
1.156 brouard 11586: i=strlen(pathr);
1.205 brouard 11587: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11588: pathr[i-1]='\0';
1.205 brouard 11589: }
11590: i=strlen(pathr);
11591: if( i==0 ){
11592: printf("ERROR Empty parameter file name\n");
11593: goto end;
11594: }
11595: for (tok = pathr; tok != NULL; ){
1.126 brouard 11596: printf("Pathr |%s|\n",pathr);
11597: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11598: printf("val= |%s| pathr=%s\n",val,pathr);
11599: strcpy (pathtot, val);
11600: if(pathr[0] == '\0') break; /* Dirty */
11601: }
11602: }
1.281 brouard 11603: else if (argc<=2){
11604: strcpy(pathtot,argv[1]);
11605: }
1.126 brouard 11606: else{
11607: strcpy(pathtot,argv[1]);
1.281 brouard 11608: strcpy(z,argv[2]);
11609: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11610: }
11611: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11612: /*cygwin_split_path(pathtot,path,optionfile);
11613: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11614: /* cutv(path,optionfile,pathtot,'\\');*/
11615:
11616: /* Split argv[0], imach program to get pathimach */
11617: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11618: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11619: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11620: /* strcpy(pathimach,argv[0]); */
11621: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11622: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11623: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11624: #ifdef WIN32
11625: _chdir(path); /* Can be a relative path */
11626: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11627: #else
1.126 brouard 11628: chdir(path); /* Can be a relative path */
1.184 brouard 11629: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11630: #endif
11631: printf("Current directory %s!\n",pathcd);
1.126 brouard 11632: strcpy(command,"mkdir ");
11633: strcat(command,optionfilefiname);
11634: if((outcmd=system(command)) != 0){
1.169 brouard 11635: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11636: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11637: /* fclose(ficlog); */
11638: /* exit(1); */
11639: }
11640: /* if((imk=mkdir(optionfilefiname))<0){ */
11641: /* perror("mkdir"); */
11642: /* } */
11643:
11644: /*-------- arguments in the command line --------*/
11645:
1.186 brouard 11646: /* Main Log file */
1.126 brouard 11647: strcat(filelog, optionfilefiname);
11648: strcat(filelog,".log"); /* */
11649: if((ficlog=fopen(filelog,"w"))==NULL) {
11650: printf("Problem with logfile %s\n",filelog);
11651: goto end;
11652: }
11653: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11654: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11655: fprintf(ficlog,"\nEnter the parameter file name: \n");
11656: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11657: path=%s \n\
11658: optionfile=%s\n\
11659: optionfilext=%s\n\
1.156 brouard 11660: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11661:
1.197 brouard 11662: syscompilerinfo(1);
1.167 brouard 11663:
1.126 brouard 11664: printf("Local time (at start):%s",strstart);
11665: fprintf(ficlog,"Local time (at start): %s",strstart);
11666: fflush(ficlog);
11667: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11668: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11669:
11670: /* */
11671: strcpy(fileres,"r");
11672: strcat(fileres, optionfilefiname);
1.201 brouard 11673: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11674: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11675: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11676:
1.186 brouard 11677: /* Main ---------arguments file --------*/
1.126 brouard 11678:
11679: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11680: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11681: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11682: fflush(ficlog);
1.149 brouard 11683: /* goto end; */
11684: exit(70);
1.126 brouard 11685: }
11686:
11687: strcpy(filereso,"o");
1.201 brouard 11688: strcat(filereso,fileresu);
1.126 brouard 11689: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11690: printf("Problem with Output resultfile: %s\n", filereso);
11691: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11692: fflush(ficlog);
11693: goto end;
11694: }
1.278 brouard 11695: /*-------- Rewriting parameter file ----------*/
11696: strcpy(rfileres,"r"); /* "Rparameterfile */
11697: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11698: strcat(rfileres,"."); /* */
11699: strcat(rfileres,optionfilext); /* Other files have txt extension */
11700: if((ficres =fopen(rfileres,"w"))==NULL) {
11701: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11702: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11703: fflush(ficlog);
11704: goto end;
11705: }
11706: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11707:
1.278 brouard 11708:
1.126 brouard 11709: /* Reads comments: lines beginning with '#' */
11710: numlinepar=0;
1.277 brouard 11711: /* Is it a BOM UTF-8 Windows file? */
11712: /* First parameter line */
1.197 brouard 11713: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11714: noffset=0;
11715: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11716: {
11717: noffset=noffset+3;
11718: printf("# File is an UTF8 Bom.\n"); // 0xBF
11719: }
1.302 brouard 11720: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11721: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11722: {
11723: noffset=noffset+2;
11724: printf("# File is an UTF16BE BOM file\n");
11725: }
11726: else if( line[0] == 0 && line[1] == 0)
11727: {
11728: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11729: noffset=noffset+4;
11730: printf("# File is an UTF16BE BOM file\n");
11731: }
11732: } else{
11733: ;/*printf(" Not a BOM file\n");*/
11734: }
11735:
1.197 brouard 11736: /* If line starts with a # it is a comment */
1.277 brouard 11737: if (line[noffset] == '#') {
1.197 brouard 11738: numlinepar++;
11739: fputs(line,stdout);
11740: fputs(line,ficparo);
1.278 brouard 11741: fputs(line,ficres);
1.197 brouard 11742: fputs(line,ficlog);
11743: continue;
11744: }else
11745: break;
11746: }
11747: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11748: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11749: if (num_filled != 5) {
11750: printf("Should be 5 parameters\n");
1.283 brouard 11751: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11752: }
1.126 brouard 11753: numlinepar++;
1.197 brouard 11754: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11755: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11756: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11757: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11758: }
11759: /* Second parameter line */
11760: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11761: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11762: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11763: if (line[0] == '#') {
11764: numlinepar++;
1.283 brouard 11765: printf("%s",line);
11766: fprintf(ficres,"%s",line);
11767: fprintf(ficparo,"%s",line);
11768: fprintf(ficlog,"%s",line);
1.197 brouard 11769: continue;
11770: }else
11771: break;
11772: }
1.223 brouard 11773: 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", \
11774: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11775: if (num_filled != 11) {
11776: 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 11777: printf("but line=%s\n",line);
1.283 brouard 11778: 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");
11779: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11780: }
1.286 brouard 11781: if( lastpass > maxwav){
11782: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11783: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11784: fflush(ficlog);
11785: goto end;
11786: }
11787: 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 11788: 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 11789: 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 11790: 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 11791: }
1.203 brouard 11792: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11793: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11794: /* Third parameter line */
11795: while(fgets(line, MAXLINE, ficpar)) {
11796: /* If line starts with a # it is a comment */
11797: if (line[0] == '#') {
11798: numlinepar++;
1.283 brouard 11799: printf("%s",line);
11800: fprintf(ficres,"%s",line);
11801: fprintf(ficparo,"%s",line);
11802: fprintf(ficlog,"%s",line);
1.197 brouard 11803: continue;
11804: }else
11805: break;
11806: }
1.201 brouard 11807: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11808: if (num_filled != 1){
1.302 brouard 11809: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
11810: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11811: model[0]='\0';
11812: goto end;
11813: }
11814: else{
11815: if (model[0]=='+'){
11816: for(i=1; i<=strlen(model);i++)
11817: modeltemp[i-1]=model[i];
1.201 brouard 11818: strcpy(model,modeltemp);
1.197 brouard 11819: }
11820: }
1.199 brouard 11821: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11822: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11823: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11824: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11825: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11826: }
11827: /* 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); */
11828: /* numlinepar=numlinepar+3; /\* In general *\/ */
11829: /* 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 11830: /* 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); */
11831: /* 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 11832: fflush(ficlog);
1.190 brouard 11833: /* if(model[0]=='#'|| model[0]== '\0'){ */
11834: if(model[0]=='#'){
1.279 brouard 11835: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11836: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11837: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11838: if(mle != -1){
1.279 brouard 11839: 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 11840: exit(1);
11841: }
11842: }
1.126 brouard 11843: while((c=getc(ficpar))=='#' && c!= EOF){
11844: ungetc(c,ficpar);
11845: fgets(line, MAXLINE, ficpar);
11846: numlinepar++;
1.195 brouard 11847: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11848: z[0]=line[1];
11849: }
11850: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11851: fputs(line, stdout);
11852: //puts(line);
1.126 brouard 11853: fputs(line,ficparo);
11854: fputs(line,ficlog);
11855: }
11856: ungetc(c,ficpar);
11857:
11858:
1.290 brouard 11859: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11860: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11861: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11862: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11863: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11864: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11865: v1+v2*age+v2*v3 makes cptcovn = 3
11866: */
11867: if (strlen(model)>1)
1.187 brouard 11868: 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 11869: else
1.187 brouard 11870: ncovmodel=2; /* Constant and age */
1.133 brouard 11871: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11872: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11873: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11874: 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);
11875: 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);
11876: fflush(stdout);
11877: fclose (ficlog);
11878: goto end;
11879: }
1.126 brouard 11880: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11881: delti=delti3[1][1];
11882: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11883: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11884: /* We could also provide initial parameters values giving by simple logistic regression
11885: * only one way, that is without matrix product. We will have nlstate maximizations */
11886: /* for(i=1;i<nlstate;i++){ */
11887: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11888: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11889: /* } */
1.126 brouard 11890: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11891: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11892: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11893: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11894: fclose (ficparo);
11895: fclose (ficlog);
11896: goto end;
11897: exit(0);
1.220 brouard 11898: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11899: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11900: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11901: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11902: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11903: matcov=matrix(1,npar,1,npar);
1.203 brouard 11904: hess=matrix(1,npar,1,npar);
1.220 brouard 11905: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11906: /* Read guessed parameters */
1.126 brouard 11907: /* Reads comments: lines beginning with '#' */
11908: while((c=getc(ficpar))=='#' && c!= EOF){
11909: ungetc(c,ficpar);
11910: fgets(line, MAXLINE, ficpar);
11911: numlinepar++;
1.141 brouard 11912: fputs(line,stdout);
1.126 brouard 11913: fputs(line,ficparo);
11914: fputs(line,ficlog);
11915: }
11916: ungetc(c,ficpar);
11917:
11918: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11919: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11920: for(i=1; i <=nlstate; i++){
1.234 brouard 11921: j=0;
1.126 brouard 11922: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11923: if(jj==i) continue;
11924: j++;
1.292 brouard 11925: while((c=getc(ficpar))=='#' && c!= EOF){
11926: ungetc(c,ficpar);
11927: fgets(line, MAXLINE, ficpar);
11928: numlinepar++;
11929: fputs(line,stdout);
11930: fputs(line,ficparo);
11931: fputs(line,ficlog);
11932: }
11933: ungetc(c,ficpar);
1.234 brouard 11934: fscanf(ficpar,"%1d%1d",&i1,&j1);
11935: if ((i1 != i) || (j1 != jj)){
11936: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11937: It might be a problem of design; if ncovcol and the model are correct\n \
11938: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11939: exit(1);
11940: }
11941: fprintf(ficparo,"%1d%1d",i1,j1);
11942: if(mle==1)
11943: printf("%1d%1d",i,jj);
11944: fprintf(ficlog,"%1d%1d",i,jj);
11945: for(k=1; k<=ncovmodel;k++){
11946: fscanf(ficpar," %lf",¶m[i][j][k]);
11947: if(mle==1){
11948: printf(" %lf",param[i][j][k]);
11949: fprintf(ficlog," %lf",param[i][j][k]);
11950: }
11951: else
11952: fprintf(ficlog," %lf",param[i][j][k]);
11953: fprintf(ficparo," %lf",param[i][j][k]);
11954: }
11955: fscanf(ficpar,"\n");
11956: numlinepar++;
11957: if(mle==1)
11958: printf("\n");
11959: fprintf(ficlog,"\n");
11960: fprintf(ficparo,"\n");
1.126 brouard 11961: }
11962: }
11963: fflush(ficlog);
1.234 brouard 11964:
1.251 brouard 11965: /* Reads parameters values */
1.126 brouard 11966: p=param[1][1];
1.251 brouard 11967: pstart=paramstart[1][1];
1.126 brouard 11968:
11969: /* Reads comments: lines beginning with '#' */
11970: while((c=getc(ficpar))=='#' && c!= EOF){
11971: ungetc(c,ficpar);
11972: fgets(line, MAXLINE, ficpar);
11973: numlinepar++;
1.141 brouard 11974: fputs(line,stdout);
1.126 brouard 11975: fputs(line,ficparo);
11976: fputs(line,ficlog);
11977: }
11978: ungetc(c,ficpar);
11979:
11980: for(i=1; i <=nlstate; i++){
11981: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11982: fscanf(ficpar,"%1d%1d",&i1,&j1);
11983: if ( (i1-i) * (j1-j) != 0){
11984: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11985: exit(1);
11986: }
11987: printf("%1d%1d",i,j);
11988: fprintf(ficparo,"%1d%1d",i1,j1);
11989: fprintf(ficlog,"%1d%1d",i1,j1);
11990: for(k=1; k<=ncovmodel;k++){
11991: fscanf(ficpar,"%le",&delti3[i][j][k]);
11992: printf(" %le",delti3[i][j][k]);
11993: fprintf(ficparo," %le",delti3[i][j][k]);
11994: fprintf(ficlog," %le",delti3[i][j][k]);
11995: }
11996: fscanf(ficpar,"\n");
11997: numlinepar++;
11998: printf("\n");
11999: fprintf(ficparo,"\n");
12000: fprintf(ficlog,"\n");
1.126 brouard 12001: }
12002: }
12003: fflush(ficlog);
1.234 brouard 12004:
1.145 brouard 12005: /* Reads covariance matrix */
1.126 brouard 12006: delti=delti3[1][1];
1.220 brouard 12007:
12008:
1.126 brouard 12009: /* 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 12010:
1.126 brouard 12011: /* Reads comments: lines beginning with '#' */
12012: while((c=getc(ficpar))=='#' && c!= EOF){
12013: ungetc(c,ficpar);
12014: fgets(line, MAXLINE, ficpar);
12015: numlinepar++;
1.141 brouard 12016: fputs(line,stdout);
1.126 brouard 12017: fputs(line,ficparo);
12018: fputs(line,ficlog);
12019: }
12020: ungetc(c,ficpar);
1.220 brouard 12021:
1.126 brouard 12022: matcov=matrix(1,npar,1,npar);
1.203 brouard 12023: hess=matrix(1,npar,1,npar);
1.131 brouard 12024: for(i=1; i <=npar; i++)
12025: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12026:
1.194 brouard 12027: /* Scans npar lines */
1.126 brouard 12028: for(i=1; i <=npar; i++){
1.226 brouard 12029: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12030: if(count != 3){
1.226 brouard 12031: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12032: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12033: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12034: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12035: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12036: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12037: exit(1);
1.220 brouard 12038: }else{
1.226 brouard 12039: if(mle==1)
12040: printf("%1d%1d%d",i1,j1,jk);
12041: }
12042: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12043: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12044: for(j=1; j <=i; j++){
1.226 brouard 12045: fscanf(ficpar," %le",&matcov[i][j]);
12046: if(mle==1){
12047: printf(" %.5le",matcov[i][j]);
12048: }
12049: fprintf(ficlog," %.5le",matcov[i][j]);
12050: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12051: }
12052: fscanf(ficpar,"\n");
12053: numlinepar++;
12054: if(mle==1)
1.220 brouard 12055: printf("\n");
1.126 brouard 12056: fprintf(ficlog,"\n");
12057: fprintf(ficparo,"\n");
12058: }
1.194 brouard 12059: /* End of read covariance matrix npar lines */
1.126 brouard 12060: for(i=1; i <=npar; i++)
12061: for(j=i+1;j<=npar;j++)
1.226 brouard 12062: matcov[i][j]=matcov[j][i];
1.126 brouard 12063:
12064: if(mle==1)
12065: printf("\n");
12066: fprintf(ficlog,"\n");
12067:
12068: fflush(ficlog);
12069:
12070: } /* End of mle != -3 */
1.218 brouard 12071:
1.186 brouard 12072: /* Main data
12073: */
1.290 brouard 12074: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12075: /* num=lvector(1,n); */
12076: /* moisnais=vector(1,n); */
12077: /* annais=vector(1,n); */
12078: /* moisdc=vector(1,n); */
12079: /* andc=vector(1,n); */
12080: /* weight=vector(1,n); */
12081: /* agedc=vector(1,n); */
12082: /* cod=ivector(1,n); */
12083: /* for(i=1;i<=n;i++){ */
12084: num=lvector(firstobs,lastobs);
12085: moisnais=vector(firstobs,lastobs);
12086: annais=vector(firstobs,lastobs);
12087: moisdc=vector(firstobs,lastobs);
12088: andc=vector(firstobs,lastobs);
12089: weight=vector(firstobs,lastobs);
12090: agedc=vector(firstobs,lastobs);
12091: cod=ivector(firstobs,lastobs);
12092: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 12093: num[i]=0;
12094: moisnais[i]=0;
12095: annais[i]=0;
12096: moisdc[i]=0;
12097: andc[i]=0;
12098: agedc[i]=0;
12099: cod[i]=0;
12100: weight[i]=1.0; /* Equal weights, 1 by default */
12101: }
1.290 brouard 12102: mint=matrix(1,maxwav,firstobs,lastobs);
12103: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 12104: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
12105: printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel));
1.126 brouard 12106: tab=ivector(1,NCOVMAX);
1.144 brouard 12107: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 12108: 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 12109:
1.136 brouard 12110: /* Reads data from file datafile */
12111: if (readdata(datafile, firstobs, lastobs, &imx)==1)
12112: goto end;
12113:
12114: /* Calculation of the number of parameters from char model */
1.234 brouard 12115: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 12116: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
12117: k=3 V4 Tvar[k=3]= 4 (from V4)
12118: k=2 V1 Tvar[k=2]= 1 (from V1)
12119: k=1 Tvar[1]=2 (from V2)
1.234 brouard 12120: */
12121:
12122: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
12123: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 ! brouard 12124: TnsdVar=ivector(1,NCOVMAX); /* */
1.234 brouard 12125: TvarsD=ivector(1,NCOVMAX); /* */
12126: TvarsQind=ivector(1,NCOVMAX); /* */
12127: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 12128: TvarF=ivector(1,NCOVMAX); /* */
12129: TvarFind=ivector(1,NCOVMAX); /* */
12130: TvarV=ivector(1,NCOVMAX); /* */
12131: TvarVind=ivector(1,NCOVMAX); /* */
12132: TvarA=ivector(1,NCOVMAX); /* */
12133: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 12134: TvarFD=ivector(1,NCOVMAX); /* */
12135: TvarFDind=ivector(1,NCOVMAX); /* */
12136: TvarFQ=ivector(1,NCOVMAX); /* */
12137: TvarFQind=ivector(1,NCOVMAX); /* */
12138: TvarVD=ivector(1,NCOVMAX); /* */
12139: TvarVDind=ivector(1,NCOVMAX); /* */
12140: TvarVQ=ivector(1,NCOVMAX); /* */
12141: TvarVQind=ivector(1,NCOVMAX); /* */
12142:
1.230 brouard 12143: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 12144: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 12145: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
12146: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
12147: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 12148: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
12149: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
12150: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
12151: */
12152: /* For model-covariate k tells which data-covariate to use but
12153: because this model-covariate is a construction we invent a new column
12154: ncovcol + k1
12155: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
12156: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 12157: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
12158: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 12159: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
12160: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 12161: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 12162: */
1.145 brouard 12163: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
12164: 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 12165: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
12166: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 ! brouard 12167: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 12168: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 12169: 4 covariates (3 plus signs)
12170: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 12171: */
12172: for(i=1;i<NCOVMAX;i++)
12173: Tage[i]=0;
1.230 brouard 12174: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 12175: * individual dummy, fixed or varying:
12176: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
12177: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 12178: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
12179: * V1 df, V2 qf, V3 & V4 dv, V5 qv
12180: * Tmodelind[1]@9={9,0,3,2,}*/
12181: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
12182: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 12183: * individual quantitative, fixed or varying:
12184: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
12185: * 3, 1, 0, 0, 0, 0, 0, 0},
12186: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 12187: /* Main decodemodel */
12188:
1.187 brouard 12189:
1.223 brouard 12190: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 12191: goto end;
12192:
1.137 brouard 12193: if((double)(lastobs-imx)/(double)imx > 1.10){
12194: nbwarn++;
12195: 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);
12196: 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);
12197: }
1.136 brouard 12198: /* if(mle==1){*/
1.137 brouard 12199: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
12200: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 12201: }
12202:
12203: /*-calculation of age at interview from date of interview and age at death -*/
12204: agev=matrix(1,maxwav,1,imx);
12205:
12206: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
12207: goto end;
12208:
1.126 brouard 12209:
1.136 brouard 12210: agegomp=(int)agemin;
1.290 brouard 12211: free_vector(moisnais,firstobs,lastobs);
12212: free_vector(annais,firstobs,lastobs);
1.126 brouard 12213: /* free_matrix(mint,1,maxwav,1,n);
12214: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 12215: /* free_vector(moisdc,1,n); */
12216: /* free_vector(andc,1,n); */
1.145 brouard 12217: /* */
12218:
1.126 brouard 12219: wav=ivector(1,imx);
1.214 brouard 12220: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
12221: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
12222: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
12223: 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.*/
12224: bh=imatrix(1,lastpass-firstpass+2,1,imx);
12225: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 12226:
12227: /* Concatenates waves */
1.214 brouard 12228: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
12229: Death is a valid wave (if date is known).
12230: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
12231: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
12232: and mw[mi+1][i]. dh depends on stepm.
12233: */
12234:
1.126 brouard 12235: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 12236: /* Concatenates waves */
1.145 brouard 12237:
1.290 brouard 12238: free_vector(moisdc,firstobs,lastobs);
12239: free_vector(andc,firstobs,lastobs);
1.215 brouard 12240:
1.126 brouard 12241: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12242: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12243: ncodemax[1]=1;
1.145 brouard 12244: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12245: cptcoveff=0;
1.220 brouard 12246: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
12247: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 12248: }
12249:
12250: ncovcombmax=pow(2,cptcoveff);
12251: invalidvarcomb=ivector(1, ncovcombmax);
12252: for(i=1;i<ncovcombmax;i++)
12253: invalidvarcomb[i]=0;
12254:
1.211 brouard 12255: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12256: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12257: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12258:
1.200 brouard 12259: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12260: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12261: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12262: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12263: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12264: * (currently 0 or 1) in the data.
12265: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12266: * corresponding modality (h,j).
12267: */
12268:
1.145 brouard 12269: h=0;
12270: /*if (cptcovn > 0) */
1.126 brouard 12271: m=pow(2,cptcoveff);
12272:
1.144 brouard 12273: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12274: * For k=4 covariates, h goes from 1 to m=2**k
12275: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12276: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 12277: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
12278: *______________________________ *______________________
12279: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
12280: * 2 2 1 1 1 * 1 0 0 0 1
12281: * 3 i=2 1 2 1 1 * 2 0 0 1 0
12282: * 4 2 2 1 1 * 3 0 0 1 1
12283: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
12284: * 6 2 1 2 1 * 5 0 1 0 1
12285: * 7 i=4 1 2 2 1 * 6 0 1 1 0
12286: * 8 2 2 2 1 * 7 0 1 1 1
12287: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
12288: * 10 2 1 1 2 * 9 1 0 0 1
12289: * 11 i=6 1 2 1 2 * 10 1 0 1 0
12290: * 12 2 2 1 2 * 11 1 0 1 1
12291: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
12292: * 14 2 1 2 2 * 13 1 1 0 1
12293: * 15 i=8 1 2 2 2 * 14 1 1 1 0
12294: * 16 2 2 2 2 * 15 1 1 1 1
12295: */
1.212 brouard 12296: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 12297: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
12298: * and the value of each covariate?
12299: * V1=1, V2=1, V3=2, V4=1 ?
12300: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
12301: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
12302: * In order to get the real value in the data, we use nbcode
12303: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
12304: * We are keeping this crazy system in order to be able (in the future?)
12305: * to have more than 2 values (0 or 1) for a covariate.
12306: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
12307: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
12308: * bbbbbbbb
12309: * 76543210
12310: * h-1 00000101 (6-1=5)
1.219 brouard 12311: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 12312: * &
12313: * 1 00000001 (1)
1.219 brouard 12314: * 00000000 = 1 & ((h-1) >> (k-1))
12315: * +1= 00000001 =1
1.211 brouard 12316: *
12317: * h=14, k=3 => h'=h-1=13, k'=k-1=2
12318: * h' 1101 =2^3+2^2+0x2^1+2^0
12319: * >>k' 11
12320: * & 00000001
12321: * = 00000001
12322: * +1 = 00000010=2 = codtabm(14,3)
12323: * Reverse h=6 and m=16?
12324: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
12325: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
12326: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
12327: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
12328: * V3=decodtabm(14,3,2**4)=2
12329: * h'=13 1101 =2^3+2^2+0x2^1+2^0
12330: *(h-1) >> (j-1) 0011 =13 >> 2
12331: * &1 000000001
12332: * = 000000001
12333: * +1= 000000010 =2
12334: * 2211
12335: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
12336: * V3=2
1.220 brouard 12337: * codtabm and decodtabm are identical
1.211 brouard 12338: */
12339:
1.145 brouard 12340:
12341: free_ivector(Ndum,-1,NCOVMAX);
12342:
12343:
1.126 brouard 12344:
1.186 brouard 12345: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 12346: strcpy(optionfilegnuplot,optionfilefiname);
12347: if(mle==-3)
1.201 brouard 12348: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 12349: strcat(optionfilegnuplot,".gp");
12350:
12351: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
12352: printf("Problem with file %s",optionfilegnuplot);
12353: }
12354: else{
1.204 brouard 12355: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 12356: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 12357: //fprintf(ficgp,"set missing 'NaNq'\n");
12358: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 12359: }
12360: /* fclose(ficgp);*/
1.186 brouard 12361:
12362:
12363: /* Initialisation of --------- index.htm --------*/
1.126 brouard 12364:
12365: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
12366: if(mle==-3)
1.201 brouard 12367: strcat(optionfilehtm,"-MORT_");
1.126 brouard 12368: strcat(optionfilehtm,".htm");
12369: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 12370: printf("Problem with %s \n",optionfilehtm);
12371: exit(0);
1.126 brouard 12372: }
12373:
12374: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
12375: strcat(optionfilehtmcov,"-cov.htm");
12376: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
12377: printf("Problem with %s \n",optionfilehtmcov), exit(0);
12378: }
12379: else{
12380: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
12381: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12382: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 12383: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
12384: }
12385:
1.324 brouard 12386: 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 12387: <hr size=\"2\" color=\"#EC5E5E\"> \n\
12388: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 12389: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12390: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 12391: \n\
12392: <hr size=\"2\" color=\"#EC5E5E\">\
12393: <ul><li><h4>Parameter files</h4>\n\
12394: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
12395: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
12396: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
12397: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
12398: - Date and time at start: %s</ul>\n",\
12399: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
12400: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
12401: fileres,fileres,\
12402: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
12403: fflush(fichtm);
12404:
12405: strcpy(pathr,path);
12406: strcat(pathr,optionfilefiname);
1.184 brouard 12407: #ifdef WIN32
12408: _chdir(optionfilefiname); /* Move to directory named optionfile */
12409: #else
1.126 brouard 12410: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12411: #endif
12412:
1.126 brouard 12413:
1.220 brouard 12414: /* Calculates basic frequencies. Computes observed prevalence at single age
12415: and for any valid combination of covariates
1.126 brouard 12416: and prints on file fileres'p'. */
1.251 brouard 12417: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12418: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12419:
12420: fprintf(fichtm,"\n");
1.286 brouard 12421: 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 12422: ftol, stepm);
12423: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12424: ncurrv=1;
12425: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12426: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12427: ncurrv=i;
12428: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12429: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12430: ncurrv=i;
12431: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12432: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12433: ncurrv=i;
12434: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12435: 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", \
12436: nlstate, ndeath, maxwav, mle, weightopt);
12437:
12438: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12439: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12440:
12441:
1.317 brouard 12442: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12443: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12444: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12445: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12446: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12447: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12448: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12449: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12450: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12451:
1.126 brouard 12452: /* For Powell, parameters are in a vector p[] starting at p[1]
12453: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12454: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12455:
12456: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12457: /* For mortality only */
1.126 brouard 12458: if (mle==-3){
1.136 brouard 12459: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12460: for(i=1;i<=NDIM;i++)
12461: for(j=1;j<=NDIM;j++)
12462: ximort[i][j]=0.;
1.186 brouard 12463: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12464: cens=ivector(firstobs,lastobs);
12465: ageexmed=vector(firstobs,lastobs);
12466: agecens=vector(firstobs,lastobs);
12467: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12468:
1.126 brouard 12469: for (i=1; i<=imx; i++){
12470: dcwave[i]=-1;
12471: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12472: if (s[m][i]>nlstate) {
12473: dcwave[i]=m;
12474: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12475: break;
12476: }
1.126 brouard 12477: }
1.226 brouard 12478:
1.126 brouard 12479: for (i=1; i<=imx; i++) {
12480: if (wav[i]>0){
1.226 brouard 12481: ageexmed[i]=agev[mw[1][i]][i];
12482: j=wav[i];
12483: agecens[i]=1.;
12484:
12485: if (ageexmed[i]> 1 && wav[i] > 0){
12486: agecens[i]=agev[mw[j][i]][i];
12487: cens[i]= 1;
12488: }else if (ageexmed[i]< 1)
12489: cens[i]= -1;
12490: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12491: cens[i]=0 ;
1.126 brouard 12492: }
12493: else cens[i]=-1;
12494: }
12495:
12496: for (i=1;i<=NDIM;i++) {
12497: for (j=1;j<=NDIM;j++)
1.226 brouard 12498: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12499: }
12500:
1.302 brouard 12501: p[1]=0.0268; p[NDIM]=0.083;
12502: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12503:
12504:
1.136 brouard 12505: #ifdef GSL
12506: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12507: #else
1.126 brouard 12508: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12509: #endif
1.201 brouard 12510: strcpy(filerespow,"POW-MORT_");
12511: strcat(filerespow,fileresu);
1.126 brouard 12512: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12513: printf("Problem with resultfile: %s\n", filerespow);
12514: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12515: }
1.136 brouard 12516: #ifdef GSL
12517: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12518: #else
1.126 brouard 12519: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12520: #endif
1.126 brouard 12521: /* for (i=1;i<=nlstate;i++)
12522: for(j=1;j<=nlstate+ndeath;j++)
12523: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12524: */
12525: fprintf(ficrespow,"\n");
1.136 brouard 12526: #ifdef GSL
12527: /* gsl starts here */
12528: T = gsl_multimin_fminimizer_nmsimplex;
12529: gsl_multimin_fminimizer *sfm = NULL;
12530: gsl_vector *ss, *x;
12531: gsl_multimin_function minex_func;
12532:
12533: /* Initial vertex size vector */
12534: ss = gsl_vector_alloc (NDIM);
12535:
12536: if (ss == NULL){
12537: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12538: }
12539: /* Set all step sizes to 1 */
12540: gsl_vector_set_all (ss, 0.001);
12541:
12542: /* Starting point */
1.126 brouard 12543:
1.136 brouard 12544: x = gsl_vector_alloc (NDIM);
12545:
12546: if (x == NULL){
12547: gsl_vector_free(ss);
12548: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12549: }
12550:
12551: /* Initialize method and iterate */
12552: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12553: /* gsl_vector_set(x, 0, 0.0268); */
12554: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12555: gsl_vector_set(x, 0, p[1]);
12556: gsl_vector_set(x, 1, p[2]);
12557:
12558: minex_func.f = &gompertz_f;
12559: minex_func.n = NDIM;
12560: minex_func.params = (void *)&p; /* ??? */
12561:
12562: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12563: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12564:
12565: printf("Iterations beginning .....\n\n");
12566: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12567:
12568: iteri=0;
12569: while (rval == GSL_CONTINUE){
12570: iteri++;
12571: status = gsl_multimin_fminimizer_iterate(sfm);
12572:
12573: if (status) printf("error: %s\n", gsl_strerror (status));
12574: fflush(0);
12575:
12576: if (status)
12577: break;
12578:
12579: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12580: ssval = gsl_multimin_fminimizer_size (sfm);
12581:
12582: if (rval == GSL_SUCCESS)
12583: printf ("converged to a local maximum at\n");
12584:
12585: printf("%5d ", iteri);
12586: for (it = 0; it < NDIM; it++){
12587: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12588: }
12589: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12590: }
12591:
12592: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12593:
12594: gsl_vector_free(x); /* initial values */
12595: gsl_vector_free(ss); /* inital step size */
12596: for (it=0; it<NDIM; it++){
12597: p[it+1]=gsl_vector_get(sfm->x,it);
12598: fprintf(ficrespow," %.12lf", p[it]);
12599: }
12600: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12601: #endif
12602: #ifdef POWELL
12603: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12604: #endif
1.126 brouard 12605: fclose(ficrespow);
12606:
1.203 brouard 12607: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12608:
12609: for(i=1; i <=NDIM; i++)
12610: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12611: matcov[i][j]=matcov[j][i];
1.126 brouard 12612:
12613: printf("\nCovariance matrix\n ");
1.203 brouard 12614: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12615: for(i=1; i <=NDIM; i++) {
12616: for(j=1;j<=NDIM;j++){
1.220 brouard 12617: printf("%f ",matcov[i][j]);
12618: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12619: }
1.203 brouard 12620: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12621: }
12622:
12623: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12624: for (i=1;i<=NDIM;i++) {
1.126 brouard 12625: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12626: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12627: }
1.302 brouard 12628: lsurv=vector(agegomp,AGESUP);
12629: lpop=vector(agegomp,AGESUP);
12630: tpop=vector(agegomp,AGESUP);
1.126 brouard 12631: lsurv[agegomp]=100000;
12632:
12633: for (k=agegomp;k<=AGESUP;k++) {
12634: agemortsup=k;
12635: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12636: }
12637:
12638: for (k=agegomp;k<agemortsup;k++)
12639: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12640:
12641: for (k=agegomp;k<agemortsup;k++){
12642: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12643: sumlpop=sumlpop+lpop[k];
12644: }
12645:
12646: tpop[agegomp]=sumlpop;
12647: for (k=agegomp;k<(agemortsup-3);k++){
12648: /* tpop[k+1]=2;*/
12649: tpop[k+1]=tpop[k]-lpop[k];
12650: }
12651:
12652:
12653: printf("\nAge lx qx dx Lx Tx e(x)\n");
12654: for (k=agegomp;k<(agemortsup-2);k++)
12655: 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]);
12656:
12657:
12658: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12659: ageminpar=50;
12660: agemaxpar=100;
1.194 brouard 12661: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12662: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12663: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12664: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12665: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12666: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12667: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12668: }else{
12669: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12670: 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 12671: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12672: }
1.201 brouard 12673: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12674: stepm, weightopt,\
12675: model,imx,p,matcov,agemortsup);
12676:
1.302 brouard 12677: free_vector(lsurv,agegomp,AGESUP);
12678: free_vector(lpop,agegomp,AGESUP);
12679: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12680: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12681: free_ivector(dcwave,firstobs,lastobs);
12682: free_vector(agecens,firstobs,lastobs);
12683: free_vector(ageexmed,firstobs,lastobs);
12684: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12685: #ifdef GSL
1.136 brouard 12686: #endif
1.186 brouard 12687: } /* Endof if mle==-3 mortality only */
1.205 brouard 12688: /* Standard */
12689: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12690: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12691: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12692: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12693: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12694: for (k=1; k<=npar;k++)
12695: printf(" %d %8.5f",k,p[k]);
12696: printf("\n");
1.205 brouard 12697: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12698: /* mlikeli uses func not funcone */
1.247 brouard 12699: /* for(i=1;i<nlstate;i++){ */
12700: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12701: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12702: /* } */
1.205 brouard 12703: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12704: }
12705: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12706: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12707: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12708: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12709: }
12710: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12711: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12712: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12713: for (k=1; k<=npar;k++)
12714: printf(" %d %8.5f",k,p[k]);
12715: printf("\n");
12716:
12717: /*--------- results files --------------*/
1.283 brouard 12718: /* 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 12719:
12720:
12721: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12722: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 12723: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12724:
12725: printf("#model= 1 + age ");
12726: fprintf(ficres,"#model= 1 + age ");
12727: fprintf(ficlog,"#model= 1 + age ");
12728: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
12729: </ul>", model);
12730:
12731: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
12732: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
12733: if(nagesqr==1){
12734: printf(" + age*age ");
12735: fprintf(ficres," + age*age ");
12736: fprintf(ficlog," + age*age ");
12737: fprintf(fichtm, "<th>+ age*age</th>");
12738: }
12739: for(j=1;j <=ncovmodel-2;j++){
12740: if(Typevar[j]==0) {
12741: printf(" + V%d ",Tvar[j]);
12742: fprintf(ficres," + V%d ",Tvar[j]);
12743: fprintf(ficlog," + V%d ",Tvar[j]);
12744: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12745: }else if(Typevar[j]==1) {
12746: printf(" + V%d*age ",Tvar[j]);
12747: fprintf(ficres," + V%d*age ",Tvar[j]);
12748: fprintf(ficlog," + V%d*age ",Tvar[j]);
12749: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12750: }else if(Typevar[j]==2) {
12751: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12752: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12753: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12754: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12755: }
12756: }
12757: printf("\n");
12758: fprintf(ficres,"\n");
12759: fprintf(ficlog,"\n");
12760: fprintf(fichtm, "</tr>");
12761: fprintf(fichtm, "\n");
12762:
12763:
1.126 brouard 12764: for(i=1,jk=1; i <=nlstate; i++){
12765: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12766: if (k != i) {
1.319 brouard 12767: fprintf(fichtm, "<tr>");
1.225 brouard 12768: printf("%d%d ",i,k);
12769: fprintf(ficlog,"%d%d ",i,k);
12770: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 12771: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12772: for(j=1; j <=ncovmodel; j++){
12773: printf("%12.7f ",p[jk]);
12774: fprintf(ficlog,"%12.7f ",p[jk]);
12775: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 12776: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 12777: jk++;
12778: }
12779: printf("\n");
12780: fprintf(ficlog,"\n");
12781: fprintf(ficres,"\n");
1.319 brouard 12782: fprintf(fichtm, "</tr>\n");
1.225 brouard 12783: }
1.126 brouard 12784: }
12785: }
1.319 brouard 12786: /* fprintf(fichtm,"</tr>\n"); */
12787: fprintf(fichtm,"</table>\n");
12788: fprintf(fichtm, "\n");
12789:
1.203 brouard 12790: if(mle != 0){
12791: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12792: ftolhess=ftol; /* Usually correct */
1.203 brouard 12793: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12794: 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");
12795: 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 12796: 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 12797: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
12798: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
12799: if(nagesqr==1){
12800: printf(" + age*age ");
12801: fprintf(ficres," + age*age ");
12802: fprintf(ficlog," + age*age ");
12803: fprintf(fichtm, "<th>+ age*age</th>");
12804: }
12805: for(j=1;j <=ncovmodel-2;j++){
12806: if(Typevar[j]==0) {
12807: printf(" + V%d ",Tvar[j]);
12808: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12809: }else if(Typevar[j]==1) {
12810: printf(" + V%d*age ",Tvar[j]);
12811: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12812: }else if(Typevar[j]==2) {
12813: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12814: }
12815: }
12816: fprintf(fichtm, "</tr>\n");
12817:
1.203 brouard 12818: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12819: for(k=1; k <=(nlstate+ndeath); k++){
12820: if (k != i) {
1.319 brouard 12821: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 12822: printf("%d%d ",i,k);
12823: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 12824: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12825: for(j=1; j <=ncovmodel; j++){
1.319 brouard 12826: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 12827: 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]));
12828: 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 12829: if(fabs(wald) > 1.96){
1.321 brouard 12830: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 12831: }else{
12832: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
12833: }
1.324 brouard 12834: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 12835: 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 12836: jk++;
12837: }
12838: printf("\n");
12839: fprintf(ficlog,"\n");
1.319 brouard 12840: fprintf(fichtm, "</tr>\n");
1.225 brouard 12841: }
12842: }
1.193 brouard 12843: }
1.203 brouard 12844: } /* end of hesscov and Wald tests */
1.319 brouard 12845: fprintf(fichtm,"</table>\n");
1.225 brouard 12846:
1.203 brouard 12847: /* */
1.126 brouard 12848: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12849: printf("# Scales (for hessian or gradient estimation)\n");
12850: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12851: for(i=1,jk=1; i <=nlstate; i++){
12852: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12853: if (j!=i) {
12854: fprintf(ficres,"%1d%1d",i,j);
12855: printf("%1d%1d",i,j);
12856: fprintf(ficlog,"%1d%1d",i,j);
12857: for(k=1; k<=ncovmodel;k++){
12858: printf(" %.5e",delti[jk]);
12859: fprintf(ficlog," %.5e",delti[jk]);
12860: fprintf(ficres," %.5e",delti[jk]);
12861: jk++;
12862: }
12863: printf("\n");
12864: fprintf(ficlog,"\n");
12865: fprintf(ficres,"\n");
12866: }
1.126 brouard 12867: }
12868: }
12869:
12870: 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 12871: if(mle >= 1) /* To big for the screen */
1.126 brouard 12872: 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");
12873: 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");
12874: /* # 121 Var(a12)\n\ */
12875: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12876: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12877: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12878: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12879: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12880: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12881: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12882:
12883:
12884: /* Just to have a covariance matrix which will be more understandable
12885: even is we still don't want to manage dictionary of variables
12886: */
12887: for(itimes=1;itimes<=2;itimes++){
12888: jj=0;
12889: for(i=1; i <=nlstate; i++){
1.225 brouard 12890: for(j=1; j <=nlstate+ndeath; j++){
12891: if(j==i) continue;
12892: for(k=1; k<=ncovmodel;k++){
12893: jj++;
12894: ca[0]= k+'a'-1;ca[1]='\0';
12895: if(itimes==1){
12896: if(mle>=1)
12897: printf("#%1d%1d%d",i,j,k);
12898: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12899: fprintf(ficres,"#%1d%1d%d",i,j,k);
12900: }else{
12901: if(mle>=1)
12902: printf("%1d%1d%d",i,j,k);
12903: fprintf(ficlog,"%1d%1d%d",i,j,k);
12904: fprintf(ficres,"%1d%1d%d",i,j,k);
12905: }
12906: ll=0;
12907: for(li=1;li <=nlstate; li++){
12908: for(lj=1;lj <=nlstate+ndeath; lj++){
12909: if(lj==li) continue;
12910: for(lk=1;lk<=ncovmodel;lk++){
12911: ll++;
12912: if(ll<=jj){
12913: cb[0]= lk +'a'-1;cb[1]='\0';
12914: if(ll<jj){
12915: if(itimes==1){
12916: if(mle>=1)
12917: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12918: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12919: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12920: }else{
12921: if(mle>=1)
12922: printf(" %.5e",matcov[jj][ll]);
12923: fprintf(ficlog," %.5e",matcov[jj][ll]);
12924: fprintf(ficres," %.5e",matcov[jj][ll]);
12925: }
12926: }else{
12927: if(itimes==1){
12928: if(mle>=1)
12929: printf(" Var(%s%1d%1d)",ca,i,j);
12930: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12931: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12932: }else{
12933: if(mle>=1)
12934: printf(" %.7e",matcov[jj][ll]);
12935: fprintf(ficlog," %.7e",matcov[jj][ll]);
12936: fprintf(ficres," %.7e",matcov[jj][ll]);
12937: }
12938: }
12939: }
12940: } /* end lk */
12941: } /* end lj */
12942: } /* end li */
12943: if(mle>=1)
12944: printf("\n");
12945: fprintf(ficlog,"\n");
12946: fprintf(ficres,"\n");
12947: numlinepar++;
12948: } /* end k*/
12949: } /*end j */
1.126 brouard 12950: } /* end i */
12951: } /* end itimes */
12952:
12953: fflush(ficlog);
12954: fflush(ficres);
1.225 brouard 12955: while(fgets(line, MAXLINE, ficpar)) {
12956: /* If line starts with a # it is a comment */
12957: if (line[0] == '#') {
12958: numlinepar++;
12959: fputs(line,stdout);
12960: fputs(line,ficparo);
12961: fputs(line,ficlog);
1.299 brouard 12962: fputs(line,ficres);
1.225 brouard 12963: continue;
12964: }else
12965: break;
12966: }
12967:
1.209 brouard 12968: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12969: /* ungetc(c,ficpar); */
12970: /* fgets(line, MAXLINE, ficpar); */
12971: /* fputs(line,stdout); */
12972: /* fputs(line,ficparo); */
12973: /* } */
12974: /* ungetc(c,ficpar); */
1.126 brouard 12975:
12976: estepm=0;
1.209 brouard 12977: 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 12978:
12979: if (num_filled != 6) {
12980: 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);
12981: 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);
12982: goto end;
12983: }
12984: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12985: }
12986: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12987: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12988:
1.209 brouard 12989: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12990: if (estepm==0 || estepm < stepm) estepm=stepm;
12991: if (fage <= 2) {
12992: bage = ageminpar;
12993: fage = agemaxpar;
12994: }
12995:
12996: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12997: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12998: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12999:
1.186 brouard 13000: /* Other stuffs, more or less useful */
1.254 brouard 13001: while(fgets(line, MAXLINE, ficpar)) {
13002: /* If line starts with a # it is a comment */
13003: if (line[0] == '#') {
13004: numlinepar++;
13005: fputs(line,stdout);
13006: fputs(line,ficparo);
13007: fputs(line,ficlog);
1.299 brouard 13008: fputs(line,ficres);
1.254 brouard 13009: continue;
13010: }else
13011: break;
13012: }
13013:
13014: 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){
13015:
13016: if (num_filled != 7) {
13017: 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);
13018: 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);
13019: goto end;
13020: }
13021: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13022: 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);
13023: 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);
13024: 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 13025: }
1.254 brouard 13026:
13027: while(fgets(line, MAXLINE, ficpar)) {
13028: /* If line starts with a # it is a comment */
13029: if (line[0] == '#') {
13030: numlinepar++;
13031: fputs(line,stdout);
13032: fputs(line,ficparo);
13033: fputs(line,ficlog);
1.299 brouard 13034: fputs(line,ficres);
1.254 brouard 13035: continue;
13036: }else
13037: break;
1.126 brouard 13038: }
13039:
13040:
13041: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13042: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13043:
1.254 brouard 13044: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13045: if (num_filled != 1) {
13046: 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);
13047: 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);
13048: goto end;
13049: }
13050: printf("pop_based=%d\n",popbased);
13051: fprintf(ficlog,"pop_based=%d\n",popbased);
13052: fprintf(ficparo,"pop_based=%d\n",popbased);
13053: fprintf(ficres,"pop_based=%d\n",popbased);
13054: }
13055:
1.258 brouard 13056: /* Results */
1.307 brouard 13057: endishere=0;
1.258 brouard 13058: nresult=0;
1.308 brouard 13059: parameterline=0;
1.258 brouard 13060: do{
13061: if(!fgets(line, MAXLINE, ficpar)){
13062: endishere=1;
1.308 brouard 13063: parameterline=15;
1.258 brouard 13064: }else if (line[0] == '#') {
13065: /* If line starts with a # it is a comment */
1.254 brouard 13066: numlinepar++;
13067: fputs(line,stdout);
13068: fputs(line,ficparo);
13069: fputs(line,ficlog);
1.299 brouard 13070: fputs(line,ficres);
1.254 brouard 13071: continue;
1.258 brouard 13072: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13073: parameterline=11;
1.296 brouard 13074: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 13075: parameterline=12;
1.307 brouard 13076: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 13077: parameterline=13;
1.307 brouard 13078: }
1.258 brouard 13079: else{
13080: parameterline=14;
1.254 brouard 13081: }
1.308 brouard 13082: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 13083: case 11:
1.296 brouard 13084: 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)){
13085: 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 13086: 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);
13087: 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);
13088: 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);
13089: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 13090: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
13091: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 13092: prvforecast = 1;
13093: }
13094: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 13095: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13096: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13097: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 13098: prvforecast = 2;
13099: }
13100: else {
13101: 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);
13102: 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);
13103: goto end;
1.258 brouard 13104: }
1.254 brouard 13105: break;
1.258 brouard 13106: case 12:
1.296 brouard 13107: 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)){
13108: 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);
13109: 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);
13110: 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);
13111: 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);
13112: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 13113: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
13114: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 13115: prvbackcast = 1;
13116: }
13117: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 13118: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13119: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13120: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 13121: prvbackcast = 2;
13122: }
13123: else {
13124: 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);
13125: 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);
13126: goto end;
1.258 brouard 13127: }
1.230 brouard 13128: break;
1.258 brouard 13129: case 13:
1.307 brouard 13130: num_filled=sscanf(line,"result:%[^\n]\n",resultline);
13131: nresult++; /* Sum of resultlines */
13132: printf("Result %d: result:%s\n",nresult, resultline);
1.318 brouard 13133: if(nresult > MAXRESULTLINESPONE-1){
13134: 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);
13135: 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 13136: goto end;
13137: }
1.310 brouard 13138: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 13139: fprintf(ficparo,"result: %s\n",resultline);
13140: fprintf(ficres,"result: %s\n",resultline);
13141: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 13142: } else
13143: goto end;
1.307 brouard 13144: break;
13145: case 14:
13146: printf("Error: Unknown command '%s'\n",line);
13147: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 13148: if(line[0] == ' ' || line[0] == '\n'){
13149: printf("It should not be an empty line '%s'\n",line);
13150: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
13151: }
1.307 brouard 13152: if(ncovmodel >=2 && nresult==0 ){
13153: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
13154: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 13155: }
1.307 brouard 13156: /* goto end; */
13157: break;
1.308 brouard 13158: case 15:
13159: printf("End of resultlines.\n");
13160: fprintf(ficlog,"End of resultlines.\n");
13161: break;
13162: default: /* parameterline =0 */
1.307 brouard 13163: nresult=1;
13164: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 13165: } /* End switch parameterline */
13166: }while(endishere==0); /* End do */
1.126 brouard 13167:
1.230 brouard 13168: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 13169: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 13170:
13171: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 13172: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 13173: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13174: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13175: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 13176: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13177: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13178: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13179: }else{
1.270 brouard 13180: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 13181: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
13182: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
13183: if(prvforecast==1){
13184: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
13185: jprojd=jproj1;
13186: mprojd=mproj1;
13187: anprojd=anproj1;
13188: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
13189: jprojf=jproj2;
13190: mprojf=mproj2;
13191: anprojf=anproj2;
13192: } else if(prvforecast == 2){
13193: dateprojd=dateintmean;
13194: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
13195: dateprojf=dateintmean+yrfproj;
13196: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
13197: }
13198: if(prvbackcast==1){
13199: datebackd=(jback1+12*mback1+365*anback1)/365;
13200: jbackd=jback1;
13201: mbackd=mback1;
13202: anbackd=anback1;
13203: datebackf=(jback2+12*mback2+365*anback2)/365;
13204: jbackf=jback2;
13205: mbackf=mback2;
13206: anbackf=anback2;
13207: } else if(prvbackcast == 2){
13208: datebackd=dateintmean;
13209: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
13210: datebackf=dateintmean-yrbproj;
13211: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
13212: }
13213:
13214: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 13215: }
13216: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 13217: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
13218: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 13219:
1.225 brouard 13220: /*------------ free_vector -------------*/
13221: /* chdir(path); */
1.220 brouard 13222:
1.215 brouard 13223: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
13224: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
13225: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
13226: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 13227: free_lvector(num,firstobs,lastobs);
13228: free_vector(agedc,firstobs,lastobs);
1.126 brouard 13229: /*free_matrix(covar,0,NCOVMAX,1,n);*/
13230: /*free_matrix(covar,1,NCOVMAX,1,n);*/
13231: fclose(ficparo);
13232: fclose(ficres);
1.220 brouard 13233:
13234:
1.186 brouard 13235: /* Other results (useful)*/
1.220 brouard 13236:
13237:
1.126 brouard 13238: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13239: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13240: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 13241: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13242: fclose(ficrespl);
13243:
13244: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13245: /*#include "hpijx.h"*/
13246: hPijx(p, bage, fage);
1.145 brouard 13247: fclose(ficrespij);
1.227 brouard 13248:
1.220 brouard 13249: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 13250: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 13251: k=1;
1.126 brouard 13252: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13253:
1.269 brouard 13254: /* Prevalence for each covariate combination in probs[age][status][cov] */
13255: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13256: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 13257: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 13258: for(k=1;k<=ncovcombmax;k++)
13259: probs[i][j][k]=0.;
1.269 brouard 13260: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
13261: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 13262: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 13263: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13264: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 13265: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 13266: for(k=1;k<=ncovcombmax;k++)
13267: mobaverages[i][j][k]=0.;
1.219 brouard 13268: mobaverage=mobaverages;
13269: if (mobilav!=0) {
1.235 brouard 13270: printf("Movingaveraging observed prevalence\n");
1.258 brouard 13271: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 13272: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
13273: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
13274: printf(" Error in movingaverage mobilav=%d\n",mobilav);
13275: }
1.269 brouard 13276: } else if (mobilavproj !=0) {
1.235 brouard 13277: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 13278: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 13279: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
13280: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
13281: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
13282: }
1.269 brouard 13283: }else{
13284: printf("Internal error moving average\n");
13285: fflush(stdout);
13286: exit(1);
1.219 brouard 13287: }
13288: }/* end if moving average */
1.227 brouard 13289:
1.126 brouard 13290: /*---------- Forecasting ------------------*/
1.296 brouard 13291: if(prevfcast==1){
13292: /* /\* if(stepm ==1){*\/ */
13293: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13294: /*This done previously after freqsummary.*/
13295: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
13296: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
13297:
13298: /* } else if (prvforecast==2){ */
13299: /* /\* if(stepm ==1){*\/ */
13300: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13301: /* } */
13302: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
13303: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 13304: }
1.269 brouard 13305:
1.296 brouard 13306: /* Prevbcasting */
13307: if(prevbcast==1){
1.219 brouard 13308: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13309: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13310: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13311:
13312: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
13313:
13314: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 13315:
1.219 brouard 13316: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
13317: fclose(ficresplb);
13318:
1.222 brouard 13319: hBijx(p, bage, fage, mobaverage);
13320: fclose(ficrespijb);
1.219 brouard 13321:
1.296 brouard 13322: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
13323: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
13324: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
13325: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
13326: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
13327: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
13328:
13329:
1.269 brouard 13330: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13331:
13332:
1.269 brouard 13333: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 13334: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13335: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13336: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 13337: } /* end Prevbcasting */
1.268 brouard 13338:
1.186 brouard 13339:
13340: /* ------ Other prevalence ratios------------ */
1.126 brouard 13341:
1.215 brouard 13342: free_ivector(wav,1,imx);
13343: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
13344: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
13345: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 13346:
13347:
1.127 brouard 13348: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 13349:
1.201 brouard 13350: strcpy(filerese,"E_");
13351: strcat(filerese,fileresu);
1.126 brouard 13352: if((ficreseij=fopen(filerese,"w"))==NULL) {
13353: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13354: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13355: }
1.208 brouard 13356: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
13357: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 13358:
13359: pstamp(ficreseij);
1.219 brouard 13360:
1.235 brouard 13361: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13362: if (cptcovn < 1){i1=1;}
13363:
13364: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13365: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13366: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13367: continue;
1.219 brouard 13368: fprintf(ficreseij,"\n#****** ");
1.235 brouard 13369: printf("\n#****** ");
1.225 brouard 13370: for(j=1;j<=cptcoveff;j++) {
1.330 ! brouard 13371: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 13372: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.235 brouard 13373: }
13374: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13375: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13376: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 13377: }
13378: fprintf(ficreseij,"******\n");
1.235 brouard 13379: printf("******\n");
1.219 brouard 13380:
13381: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13382: oldm=oldms;savm=savms;
1.330 ! brouard 13383: /* printf("HELLO Entering evsij bage=%d fage=%d k=%d estepm=%d nres=%d\n",(int) bage, (int)fage, k, estepm, nres); */
1.235 brouard 13384: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 13385:
1.219 brouard 13386: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 13387: }
13388: fclose(ficreseij);
1.208 brouard 13389: printf("done evsij\n");fflush(stdout);
13390: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 13391:
1.218 brouard 13392:
1.227 brouard 13393: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 13394:
1.201 brouard 13395: strcpy(filerest,"T_");
13396: strcat(filerest,fileresu);
1.127 brouard 13397: if((ficrest=fopen(filerest,"w"))==NULL) {
13398: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
13399: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
13400: }
1.208 brouard 13401: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
13402: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 13403: strcpy(fileresstde,"STDE_");
13404: strcat(fileresstde,fileresu);
1.126 brouard 13405: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 13406: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
13407: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 13408: }
1.227 brouard 13409: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
13410: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 13411:
1.201 brouard 13412: strcpy(filerescve,"CVE_");
13413: strcat(filerescve,fileresu);
1.126 brouard 13414: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 13415: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
13416: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 13417: }
1.227 brouard 13418: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
13419: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 13420:
1.201 brouard 13421: strcpy(fileresv,"V_");
13422: strcat(fileresv,fileresu);
1.126 brouard 13423: if((ficresvij=fopen(fileresv,"w"))==NULL) {
13424: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
13425: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
13426: }
1.227 brouard 13427: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
13428: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 13429:
1.235 brouard 13430: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13431: if (cptcovn < 1){i1=1;}
13432:
13433: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13434: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13435: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13436: continue;
1.321 brouard 13437: printf("\n# model %s \n#****** Result for:", model);
13438: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
13439: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.227 brouard 13440: for(j=1;j<=cptcoveff;j++){
1.330 ! brouard 13441: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 13442: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 13443: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.227 brouard 13444: }
1.235 brouard 13445: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13446: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13447: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13448: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13449: }
1.208 brouard 13450: fprintf(ficrest,"******\n");
1.227 brouard 13451: fprintf(ficlog,"******\n");
13452: printf("******\n");
1.208 brouard 13453:
13454: fprintf(ficresstdeij,"\n#****** ");
13455: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 13456: for(j=1;j<=cptcoveff;j++) {
1.330 ! brouard 13457: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
! 13458: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.208 brouard 13459: }
1.235 brouard 13460: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13461: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13462: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13463: }
1.208 brouard 13464: fprintf(ficresstdeij,"******\n");
13465: fprintf(ficrescveij,"******\n");
13466:
13467: fprintf(ficresvij,"\n#****** ");
1.238 brouard 13468: /* pstamp(ficresvij); */
1.225 brouard 13469: for(j=1;j<=cptcoveff;j++)
1.330 ! brouard 13470: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]);
1.235 brouard 13471: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13472: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13473: }
1.208 brouard 13474: fprintf(ficresvij,"******\n");
13475:
13476: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13477: oldm=oldms;savm=savms;
1.235 brouard 13478: printf(" cvevsij ");
13479: fprintf(ficlog, " cvevsij ");
13480: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 13481: printf(" end cvevsij \n ");
13482: fprintf(ficlog, " end cvevsij \n ");
13483:
13484: /*
13485: */
13486: /* goto endfree; */
13487:
13488: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13489: pstamp(ficrest);
13490:
1.269 brouard 13491: epj=vector(1,nlstate+1);
1.208 brouard 13492: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 13493: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
13494: cptcod= 0; /* To be deleted */
13495: printf("varevsij vpopbased=%d \n",vpopbased);
13496: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 13497: 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 13498: 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 ");
13499: if(vpopbased==1)
13500: 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);
13501: else
1.288 brouard 13502: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13503: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
13504: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
13505: fprintf(ficrest,"\n");
13506: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 13507: printf("Computing age specific forward period (stable) prevalences in each health state \n");
13508: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13509: for(age=bage; age <=fage ;age++){
1.235 brouard 13510: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 13511: if (vpopbased==1) {
13512: if(mobilav ==0){
13513: for(i=1; i<=nlstate;i++)
13514: prlim[i][i]=probs[(int)age][i][k];
13515: }else{ /* mobilav */
13516: for(i=1; i<=nlstate;i++)
13517: prlim[i][i]=mobaverage[(int)age][i][k];
13518: }
13519: }
1.219 brouard 13520:
1.227 brouard 13521: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
13522: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
13523: /* printf(" age %4.0f ",age); */
13524: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
13525: for(i=1, epj[j]=0.;i <=nlstate;i++) {
13526: epj[j] += prlim[i][i]*eij[i][j][(int)age];
13527: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
13528: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
13529: }
13530: epj[nlstate+1] +=epj[j];
13531: }
13532: /* printf(" age %4.0f \n",age); */
1.219 brouard 13533:
1.227 brouard 13534: for(i=1, vepp=0.;i <=nlstate;i++)
13535: for(j=1;j <=nlstate;j++)
13536: vepp += vareij[i][j][(int)age];
13537: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
13538: for(j=1;j <=nlstate;j++){
13539: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
13540: }
13541: fprintf(ficrest,"\n");
13542: }
1.208 brouard 13543: } /* End vpopbased */
1.269 brouard 13544: free_vector(epj,1,nlstate+1);
1.208 brouard 13545: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
13546: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 13547: printf("done selection\n");fflush(stdout);
13548: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 13549:
1.235 brouard 13550: } /* End k selection */
1.227 brouard 13551:
13552: printf("done State-specific expectancies\n");fflush(stdout);
13553: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
13554:
1.288 brouard 13555: /* variance-covariance of forward period prevalence*/
1.269 brouard 13556: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13557:
1.227 brouard 13558:
1.290 brouard 13559: free_vector(weight,firstobs,lastobs);
1.330 ! brouard 13560: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 13561: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 13562: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
13563: free_matrix(anint,1,maxwav,firstobs,lastobs);
13564: free_matrix(mint,1,maxwav,firstobs,lastobs);
13565: free_ivector(cod,firstobs,lastobs);
1.227 brouard 13566: free_ivector(tab,1,NCOVMAX);
13567: fclose(ficresstdeij);
13568: fclose(ficrescveij);
13569: fclose(ficresvij);
13570: fclose(ficrest);
13571: fclose(ficpar);
13572:
13573:
1.126 brouard 13574: /*---------- End : free ----------------*/
1.219 brouard 13575: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 13576: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
13577: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 13578: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
13579: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 13580: } /* mle==-3 arrives here for freeing */
1.227 brouard 13581: /* endfree:*/
13582: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
13583: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
13584: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 13585: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
13586: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
13587: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
13588: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 13589: free_matrix(matcov,1,npar,1,npar);
13590: free_matrix(hess,1,npar,1,npar);
13591: /*free_vector(delti,1,npar);*/
13592: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13593: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 13594: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 13595: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13596:
13597: free_ivector(ncodemax,1,NCOVMAX);
13598: free_ivector(ncodemaxwundef,1,NCOVMAX);
13599: free_ivector(Dummy,-1,NCOVMAX);
13600: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 13601: free_ivector(DummyV,1,NCOVMAX);
13602: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 13603: free_ivector(Typevar,-1,NCOVMAX);
13604: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 13605: free_ivector(TvarsQ,1,NCOVMAX);
13606: free_ivector(TvarsQind,1,NCOVMAX);
13607: free_ivector(TvarsD,1,NCOVMAX);
1.330 ! brouard 13608: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 13609: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 13610: free_ivector(TvarFD,1,NCOVMAX);
13611: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 13612: free_ivector(TvarF,1,NCOVMAX);
13613: free_ivector(TvarFind,1,NCOVMAX);
13614: free_ivector(TvarV,1,NCOVMAX);
13615: free_ivector(TvarVind,1,NCOVMAX);
13616: free_ivector(TvarA,1,NCOVMAX);
13617: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 13618: free_ivector(TvarFQ,1,NCOVMAX);
13619: free_ivector(TvarFQind,1,NCOVMAX);
13620: free_ivector(TvarVD,1,NCOVMAX);
13621: free_ivector(TvarVDind,1,NCOVMAX);
13622: free_ivector(TvarVQ,1,NCOVMAX);
13623: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 13624: free_ivector(Tvarsel,1,NCOVMAX);
13625: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 13626: free_ivector(Tposprod,1,NCOVMAX);
13627: free_ivector(Tprod,1,NCOVMAX);
13628: free_ivector(Tvaraff,1,NCOVMAX);
13629: free_ivector(invalidvarcomb,1,ncovcombmax);
13630: free_ivector(Tage,1,NCOVMAX);
13631: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 13632: free_ivector(TmodelInvind,1,NCOVMAX);
13633: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 13634:
13635: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
13636: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 13637: fflush(fichtm);
13638: fflush(ficgp);
13639:
1.227 brouard 13640:
1.126 brouard 13641: if((nberr >0) || (nbwarn>0)){
1.216 brouard 13642: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
13643: 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 13644: }else{
13645: printf("End of Imach\n");
13646: fprintf(ficlog,"End of Imach\n");
13647: }
13648: printf("See log file on %s\n",filelog);
13649: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 13650: /*(void) gettimeofday(&end_time,&tzp);*/
13651: rend_time = time(NULL);
13652: end_time = *localtime(&rend_time);
13653: /* tml = *localtime(&end_time.tm_sec); */
13654: strcpy(strtend,asctime(&end_time));
1.126 brouard 13655: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
13656: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 13657: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13658:
1.157 brouard 13659: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13660: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13661: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13662: /* printf("Total time was %d uSec.\n", total_usecs);*/
13663: /* if(fileappend(fichtm,optionfilehtm)){ */
13664: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13665: fclose(fichtm);
13666: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13667: fclose(fichtmcov);
13668: fclose(ficgp);
13669: fclose(ficlog);
13670: /*------ End -----------*/
1.227 brouard 13671:
1.281 brouard 13672:
13673: /* Executes gnuplot */
1.227 brouard 13674:
13675: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13676: #ifdef WIN32
1.227 brouard 13677: if (_chdir(pathcd) != 0)
13678: printf("Can't move to directory %s!\n",path);
13679: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13680: #else
1.227 brouard 13681: if(chdir(pathcd) != 0)
13682: printf("Can't move to directory %s!\n", path);
13683: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13684: #endif
1.126 brouard 13685: printf("Current directory %s!\n",pathcd);
13686: /*strcat(plotcmd,CHARSEPARATOR);*/
13687: sprintf(plotcmd,"gnuplot");
1.157 brouard 13688: #ifdef _WIN32
1.126 brouard 13689: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13690: #endif
13691: if(!stat(plotcmd,&info)){
1.158 brouard 13692: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13693: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13694: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13695: }else
13696: strcpy(pplotcmd,plotcmd);
1.157 brouard 13697: #ifdef __unix
1.126 brouard 13698: strcpy(plotcmd,GNUPLOTPROGRAM);
13699: if(!stat(plotcmd,&info)){
1.158 brouard 13700: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13701: }else
13702: strcpy(pplotcmd,plotcmd);
13703: #endif
13704: }else
13705: strcpy(pplotcmd,plotcmd);
13706:
13707: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13708: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13709: strcpy(pplotcmd,plotcmd);
1.227 brouard 13710:
1.126 brouard 13711: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13712: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13713: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13714: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13715: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13716: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13717: strcpy(plotcmd,pplotcmd);
13718: }
1.126 brouard 13719: }
1.158 brouard 13720: printf(" Successful, please wait...");
1.126 brouard 13721: while (z[0] != 'q') {
13722: /* chdir(path); */
1.154 brouard 13723: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13724: scanf("%s",z);
13725: /* if (z[0] == 'c') system("./imach"); */
13726: if (z[0] == 'e') {
1.158 brouard 13727: #ifdef __APPLE__
1.152 brouard 13728: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13729: #elif __linux
13730: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13731: #else
1.152 brouard 13732: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13733: #endif
13734: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13735: system(pplotcmd);
1.126 brouard 13736: }
13737: else if (z[0] == 'g') system(plotcmd);
13738: else if (z[0] == 'q') exit(0);
13739: }
1.227 brouard 13740: end:
1.126 brouard 13741: while (z[0] != 'q') {
1.195 brouard 13742: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13743: scanf("%s",z);
13744: }
1.283 brouard 13745: printf("End\n");
1.282 brouard 13746: exit(0);
1.126 brouard 13747: }
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