Annotation of imach/src/imach.c, revision 1.328
1.328 ! brouard 1: /* $Id: imach.c,v 1.327 2022/07/27 14:47:35 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.328 ! brouard 4: Revision 1.327 2022/07/27 14:47:35 brouard
! 5: Summary: Still a problem for one-step probabilities in case of quantitative variables
! 6:
1.327 brouard 7: Revision 1.326 2022/07/26 17:33:55 brouard
8: Summary: some test with nres=1
9:
1.326 brouard 10: Revision 1.325 2022/07/25 14:27:23 brouard
11: Summary: r30
12:
13: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
14: coredumped, revealed by Feiuno, thank you.
15:
1.325 brouard 16: Revision 1.324 2022/07/23 17:44:26 brouard
17: *** empty log message ***
18:
1.324 brouard 19: Revision 1.323 2022/07/22 12:30:08 brouard
20: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
21:
1.323 brouard 22: Revision 1.322 2022/07/22 12:27:48 brouard
23: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
24:
1.322 brouard 25: Revision 1.321 2022/07/22 12:04:24 brouard
26: Summary: r28
27:
28: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
29:
1.321 brouard 30: Revision 1.320 2022/06/02 05:10:11 brouard
31: *** empty log message ***
32:
1.320 brouard 33: Revision 1.319 2022/06/02 04:45:11 brouard
34: * imach.c (Module): Adding the Wald tests from the log to the main
35: htm for better display of the maximum likelihood estimators.
36:
1.319 brouard 37: Revision 1.318 2022/05/24 08:10:59 brouard
38: * imach.c (Module): Some attempts to find a bug of wrong estimates
39: of confidencce intervals with product in the equation modelC
40:
1.318 brouard 41: Revision 1.317 2022/05/15 15:06:23 brouard
42: * imach.c (Module): Some minor improvements
43:
1.317 brouard 44: Revision 1.316 2022/05/11 15:11:31 brouard
45: Summary: r27
46:
1.316 brouard 47: Revision 1.315 2022/05/11 15:06:32 brouard
48: *** empty log message ***
49:
1.315 brouard 50: Revision 1.314 2022/04/13 17:43:09 brouard
51: * imach.c (Module): Adding link to text data files
52:
1.314 brouard 53: Revision 1.313 2022/04/11 15:57:42 brouard
54: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
55:
1.313 brouard 56: Revision 1.312 2022/04/05 21:24:39 brouard
57: *** empty log message ***
58:
1.312 brouard 59: Revision 1.311 2022/04/05 21:03:51 brouard
60: Summary: Fixed quantitative covariates
61:
62: Fixed covariates (dummy or quantitative)
63: with missing values have never been allowed but are ERRORS and
64: program quits. Standard deviations of fixed covariates were
65: wrongly computed. Mean and standard deviations of time varying
66: covariates are still not computed.
67:
1.311 brouard 68: Revision 1.310 2022/03/17 08:45:53 brouard
69: Summary: 99r25
70:
71: Improving detection of errors: result lines should be compatible with
72: the model.
73:
1.310 brouard 74: Revision 1.309 2021/05/20 12:39:14 brouard
75: Summary: Version 0.99r24
76:
1.309 brouard 77: Revision 1.308 2021/03/31 13:11:57 brouard
78: Summary: Version 0.99r23
79:
80:
81: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
82:
1.308 brouard 83: Revision 1.307 2021/03/08 18:11:32 brouard
84: Summary: 0.99r22 fixed bug on result:
85:
1.307 brouard 86: Revision 1.306 2021/02/20 15:44:02 brouard
87: Summary: Version 0.99r21
88:
89: * imach.c (Module): Fix bug on quitting after result lines!
90: (Module): Version 0.99r21
91:
1.306 brouard 92: Revision 1.305 2021/02/20 15:28:30 brouard
93: * imach.c (Module): Fix bug on quitting after result lines!
94:
1.305 brouard 95: Revision 1.304 2021/02/12 11:34:20 brouard
96: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
97:
1.304 brouard 98: Revision 1.303 2021/02/11 19:50:15 brouard
99: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
100:
1.303 brouard 101: Revision 1.302 2020/02/22 21:00:05 brouard
102: * (Module): imach.c Update mle=-3 (for computing Life expectancy
103: and life table from the data without any state)
104:
1.302 brouard 105: Revision 1.301 2019/06/04 13:51:20 brouard
106: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
107:
1.301 brouard 108: Revision 1.300 2019/05/22 19:09:45 brouard
109: Summary: version 0.99r19 of May 2019
110:
1.300 brouard 111: Revision 1.299 2019/05/22 18:37:08 brouard
112: Summary: Cleaned 0.99r19
113:
1.299 brouard 114: Revision 1.298 2019/05/22 18:19:56 brouard
115: *** empty log message ***
116:
1.298 brouard 117: Revision 1.297 2019/05/22 17:56:10 brouard
118: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
119:
1.297 brouard 120: Revision 1.296 2019/05/20 13:03:18 brouard
121: Summary: Projection syntax simplified
122:
123:
124: We can now start projections, forward or backward, from the mean date
125: of inteviews up to or down to a number of years of projection:
126: prevforecast=1 yearsfproj=15.3 mobil_average=0
127: or
128: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
129: or
130: prevbackcast=1 yearsbproj=12.3 mobil_average=1
131: or
132: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
133:
1.296 brouard 134: Revision 1.295 2019/05/18 09:52:50 brouard
135: Summary: doxygen tex bug
136:
1.295 brouard 137: Revision 1.294 2019/05/16 14:54:33 brouard
138: Summary: There was some wrong lines added
139:
1.294 brouard 140: Revision 1.293 2019/05/09 15:17:34 brouard
141: *** empty log message ***
142:
1.293 brouard 143: Revision 1.292 2019/05/09 14:17:20 brouard
144: Summary: Some updates
145:
1.292 brouard 146: Revision 1.291 2019/05/09 13:44:18 brouard
147: Summary: Before ncovmax
148:
1.291 brouard 149: Revision 1.290 2019/05/09 13:39:37 brouard
150: Summary: 0.99r18 unlimited number of individuals
151:
152: 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.
153:
1.290 brouard 154: Revision 1.289 2018/12/13 09:16:26 brouard
155: Summary: Bug for young ages (<-30) will be in r17
156:
1.289 brouard 157: Revision 1.288 2018/05/02 20:58:27 brouard
158: Summary: Some bugs fixed
159:
1.288 brouard 160: Revision 1.287 2018/05/01 17:57:25 brouard
161: Summary: Bug fixed by providing frequencies only for non missing covariates
162:
1.287 brouard 163: Revision 1.286 2018/04/27 14:27:04 brouard
164: Summary: some minor bugs
165:
1.286 brouard 166: Revision 1.285 2018/04/21 21:02:16 brouard
167: Summary: Some bugs fixed, valgrind tested
168:
1.285 brouard 169: Revision 1.284 2018/04/20 05:22:13 brouard
170: Summary: Computing mean and stdeviation of fixed quantitative variables
171:
1.284 brouard 172: Revision 1.283 2018/04/19 14:49:16 brouard
173: Summary: Some minor bugs fixed
174:
1.283 brouard 175: Revision 1.282 2018/02/27 22:50:02 brouard
176: *** empty log message ***
177:
1.282 brouard 178: Revision 1.281 2018/02/27 19:25:23 brouard
179: Summary: Adding second argument for quitting
180:
1.281 brouard 181: Revision 1.280 2018/02/21 07:58:13 brouard
182: Summary: 0.99r15
183:
184: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
185:
1.280 brouard 186: Revision 1.279 2017/07/20 13:35:01 brouard
187: Summary: temporary working
188:
1.279 brouard 189: Revision 1.278 2017/07/19 14:09:02 brouard
190: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
191:
1.278 brouard 192: Revision 1.277 2017/07/17 08:53:49 brouard
193: Summary: BOM files can be read now
194:
1.277 brouard 195: Revision 1.276 2017/06/30 15:48:31 brouard
196: Summary: Graphs improvements
197:
1.276 brouard 198: Revision 1.275 2017/06/30 13:39:33 brouard
199: Summary: Saito's color
200:
1.275 brouard 201: Revision 1.274 2017/06/29 09:47:08 brouard
202: Summary: Version 0.99r14
203:
1.274 brouard 204: Revision 1.273 2017/06/27 11:06:02 brouard
205: Summary: More documentation on projections
206:
1.273 brouard 207: Revision 1.272 2017/06/27 10:22:40 brouard
208: Summary: Color of backprojection changed from 6 to 5(yellow)
209:
1.272 brouard 210: Revision 1.271 2017/06/27 10:17:50 brouard
211: Summary: Some bug with rint
212:
1.271 brouard 213: Revision 1.270 2017/05/24 05:45:29 brouard
214: *** empty log message ***
215:
1.270 brouard 216: Revision 1.269 2017/05/23 08:39:25 brouard
217: Summary: Code into subroutine, cleanings
218:
1.269 brouard 219: Revision 1.268 2017/05/18 20:09:32 brouard
220: Summary: backprojection and confidence intervals of backprevalence
221:
1.268 brouard 222: Revision 1.267 2017/05/13 10:25:05 brouard
223: Summary: temporary save for backprojection
224:
1.267 brouard 225: Revision 1.266 2017/05/13 07:26:12 brouard
226: Summary: Version 0.99r13 (improvements and bugs fixed)
227:
1.266 brouard 228: Revision 1.265 2017/04/26 16:22:11 brouard
229: Summary: imach 0.99r13 Some bugs fixed
230:
1.265 brouard 231: Revision 1.264 2017/04/26 06:01:29 brouard
232: Summary: Labels in graphs
233:
1.264 brouard 234: Revision 1.263 2017/04/24 15:23:15 brouard
235: Summary: to save
236:
1.263 brouard 237: Revision 1.262 2017/04/18 16:48:12 brouard
238: *** empty log message ***
239:
1.262 brouard 240: Revision 1.261 2017/04/05 10:14:09 brouard
241: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
242:
1.261 brouard 243: Revision 1.260 2017/04/04 17:46:59 brouard
244: Summary: Gnuplot indexations fixed (humm)
245:
1.260 brouard 246: Revision 1.259 2017/04/04 13:01:16 brouard
247: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
248:
1.259 brouard 249: Revision 1.258 2017/04/03 10:17:47 brouard
250: Summary: Version 0.99r12
251:
252: Some cleanings, conformed with updated documentation.
253:
1.258 brouard 254: Revision 1.257 2017/03/29 16:53:30 brouard
255: Summary: Temp
256:
1.257 brouard 257: Revision 1.256 2017/03/27 05:50:23 brouard
258: Summary: Temporary
259:
1.256 brouard 260: Revision 1.255 2017/03/08 16:02:28 brouard
261: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
262:
1.255 brouard 263: Revision 1.254 2017/03/08 07:13:00 brouard
264: Summary: Fixing data parameter line
265:
1.254 brouard 266: Revision 1.253 2016/12/15 11:59:41 brouard
267: Summary: 0.99 in progress
268:
1.253 brouard 269: Revision 1.252 2016/09/15 21:15:37 brouard
270: *** empty log message ***
271:
1.252 brouard 272: Revision 1.251 2016/09/15 15:01:13 brouard
273: Summary: not working
274:
1.251 brouard 275: Revision 1.250 2016/09/08 16:07:27 brouard
276: Summary: continue
277:
1.250 brouard 278: Revision 1.249 2016/09/07 17:14:18 brouard
279: Summary: Starting values from frequencies
280:
1.249 brouard 281: Revision 1.248 2016/09/07 14:10:18 brouard
282: *** empty log message ***
283:
1.248 brouard 284: Revision 1.247 2016/09/02 11:11:21 brouard
285: *** empty log message ***
286:
1.247 brouard 287: Revision 1.246 2016/09/02 08:49:22 brouard
288: *** empty log message ***
289:
1.246 brouard 290: Revision 1.245 2016/09/02 07:25:01 brouard
291: *** empty log message ***
292:
1.245 brouard 293: Revision 1.244 2016/09/02 07:17:34 brouard
294: *** empty log message ***
295:
1.244 brouard 296: Revision 1.243 2016/09/02 06:45:35 brouard
297: *** empty log message ***
298:
1.243 brouard 299: Revision 1.242 2016/08/30 15:01:20 brouard
300: Summary: Fixing a lots
301:
1.242 brouard 302: Revision 1.241 2016/08/29 17:17:25 brouard
303: Summary: gnuplot problem in Back projection to fix
304:
1.241 brouard 305: Revision 1.240 2016/08/29 07:53:18 brouard
306: Summary: Better
307:
1.240 brouard 308: Revision 1.239 2016/08/26 15:51:03 brouard
309: Summary: Improvement in Powell output in order to copy and paste
310:
311: Author:
312:
1.239 brouard 313: Revision 1.238 2016/08/26 14:23:35 brouard
314: Summary: Starting tests of 0.99
315:
1.238 brouard 316: Revision 1.237 2016/08/26 09:20:19 brouard
317: Summary: to valgrind
318:
1.237 brouard 319: Revision 1.236 2016/08/25 10:50:18 brouard
320: *** empty log message ***
321:
1.236 brouard 322: Revision 1.235 2016/08/25 06:59:23 brouard
323: *** empty log message ***
324:
1.235 brouard 325: Revision 1.234 2016/08/23 16:51:20 brouard
326: *** empty log message ***
327:
1.234 brouard 328: Revision 1.233 2016/08/23 07:40:50 brouard
329: Summary: not working
330:
1.233 brouard 331: Revision 1.232 2016/08/22 14:20:21 brouard
332: Summary: not working
333:
1.232 brouard 334: Revision 1.231 2016/08/22 07:17:15 brouard
335: Summary: not working
336:
1.231 brouard 337: Revision 1.230 2016/08/22 06:55:53 brouard
338: Summary: Not working
339:
1.230 brouard 340: Revision 1.229 2016/07/23 09:45:53 brouard
341: Summary: Completing for func too
342:
1.229 brouard 343: Revision 1.228 2016/07/22 17:45:30 brouard
344: Summary: Fixing some arrays, still debugging
345:
1.227 brouard 346: Revision 1.226 2016/07/12 18:42:34 brouard
347: Summary: temp
348:
1.226 brouard 349: Revision 1.225 2016/07/12 08:40:03 brouard
350: Summary: saving but not running
351:
1.225 brouard 352: Revision 1.224 2016/07/01 13:16:01 brouard
353: Summary: Fixes
354:
1.224 brouard 355: Revision 1.223 2016/02/19 09:23:35 brouard
356: Summary: temporary
357:
1.223 brouard 358: Revision 1.222 2016/02/17 08:14:50 brouard
359: Summary: Probably last 0.98 stable version 0.98r6
360:
1.222 brouard 361: Revision 1.221 2016/02/15 23:35:36 brouard
362: Summary: minor bug
363:
1.220 brouard 364: Revision 1.219 2016/02/15 00:48:12 brouard
365: *** empty log message ***
366:
1.219 brouard 367: Revision 1.218 2016/02/12 11:29:23 brouard
368: Summary: 0.99 Back projections
369:
1.218 brouard 370: Revision 1.217 2015/12/23 17:18:31 brouard
371: Summary: Experimental backcast
372:
1.217 brouard 373: Revision 1.216 2015/12/18 17:32:11 brouard
374: Summary: 0.98r4 Warning and status=-2
375:
376: Version 0.98r4 is now:
377: - displaying an error when status is -1, date of interview unknown and date of death known;
378: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
379: Older changes concerning s=-2, dating from 2005 have been supersed.
380:
1.216 brouard 381: Revision 1.215 2015/12/16 08:52:24 brouard
382: Summary: 0.98r4 working
383:
1.215 brouard 384: Revision 1.214 2015/12/16 06:57:54 brouard
385: Summary: temporary not working
386:
1.214 brouard 387: Revision 1.213 2015/12/11 18:22:17 brouard
388: Summary: 0.98r4
389:
1.213 brouard 390: Revision 1.212 2015/11/21 12:47:24 brouard
391: Summary: minor typo
392:
1.212 brouard 393: Revision 1.211 2015/11/21 12:41:11 brouard
394: Summary: 0.98r3 with some graph of projected cross-sectional
395:
396: Author: Nicolas Brouard
397:
1.211 brouard 398: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 399: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 400: Summary: Adding ftolpl parameter
401: Author: N Brouard
402:
403: We had difficulties to get smoothed confidence intervals. It was due
404: to the period prevalence which wasn't computed accurately. The inner
405: parameter ftolpl is now an outer parameter of the .imach parameter
406: file after estepm. If ftolpl is small 1.e-4 and estepm too,
407: computation are long.
408:
1.209 brouard 409: Revision 1.208 2015/11/17 14:31:57 brouard
410: Summary: temporary
411:
1.208 brouard 412: Revision 1.207 2015/10/27 17:36:57 brouard
413: *** empty log message ***
414:
1.207 brouard 415: Revision 1.206 2015/10/24 07:14:11 brouard
416: *** empty log message ***
417:
1.206 brouard 418: Revision 1.205 2015/10/23 15:50:53 brouard
419: Summary: 0.98r3 some clarification for graphs on likelihood contributions
420:
1.205 brouard 421: Revision 1.204 2015/10/01 16:20:26 brouard
422: Summary: Some new graphs of contribution to likelihood
423:
1.204 brouard 424: Revision 1.203 2015/09/30 17:45:14 brouard
425: Summary: looking at better estimation of the hessian
426:
427: Also a better criteria for convergence to the period prevalence And
428: therefore adding the number of years needed to converge. (The
429: prevalence in any alive state shold sum to one
430:
1.203 brouard 431: Revision 1.202 2015/09/22 19:45:16 brouard
432: Summary: Adding some overall graph on contribution to likelihood. Might change
433:
1.202 brouard 434: Revision 1.201 2015/09/15 17:34:58 brouard
435: Summary: 0.98r0
436:
437: - Some new graphs like suvival functions
438: - Some bugs fixed like model=1+age+V2.
439:
1.201 brouard 440: Revision 1.200 2015/09/09 16:53:55 brouard
441: Summary: Big bug thanks to Flavia
442:
443: Even model=1+age+V2. did not work anymore
444:
1.200 brouard 445: Revision 1.199 2015/09/07 14:09:23 brouard
446: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
447:
1.199 brouard 448: Revision 1.198 2015/09/03 07:14:39 brouard
449: Summary: 0.98q5 Flavia
450:
1.198 brouard 451: Revision 1.197 2015/09/01 18:24:39 brouard
452: *** empty log message ***
453:
1.197 brouard 454: Revision 1.196 2015/08/18 23:17:52 brouard
455: Summary: 0.98q5
456:
1.196 brouard 457: Revision 1.195 2015/08/18 16:28:39 brouard
458: Summary: Adding a hack for testing purpose
459:
460: After reading the title, ftol and model lines, if the comment line has
461: a q, starting with #q, the answer at the end of the run is quit. It
462: permits to run test files in batch with ctest. The former workaround was
463: $ echo q | imach foo.imach
464:
1.195 brouard 465: Revision 1.194 2015/08/18 13:32:00 brouard
466: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
467:
1.194 brouard 468: Revision 1.193 2015/08/04 07:17:42 brouard
469: Summary: 0.98q4
470:
1.193 brouard 471: Revision 1.192 2015/07/16 16:49:02 brouard
472: Summary: Fixing some outputs
473:
1.192 brouard 474: Revision 1.191 2015/07/14 10:00:33 brouard
475: Summary: Some fixes
476:
1.191 brouard 477: Revision 1.190 2015/05/05 08:51:13 brouard
478: Summary: Adding digits in output parameters (7 digits instead of 6)
479:
480: Fix 1+age+.
481:
1.190 brouard 482: Revision 1.189 2015/04/30 14:45:16 brouard
483: Summary: 0.98q2
484:
1.189 brouard 485: Revision 1.188 2015/04/30 08:27:53 brouard
486: *** empty log message ***
487:
1.188 brouard 488: Revision 1.187 2015/04/29 09:11:15 brouard
489: *** empty log message ***
490:
1.187 brouard 491: Revision 1.186 2015/04/23 12:01:52 brouard
492: Summary: V1*age is working now, version 0.98q1
493:
494: Some codes had been disabled in order to simplify and Vn*age was
495: working in the optimization phase, ie, giving correct MLE parameters,
496: but, as usual, outputs were not correct and program core dumped.
497:
1.186 brouard 498: Revision 1.185 2015/03/11 13:26:42 brouard
499: Summary: Inclusion of compile and links command line for Intel Compiler
500:
1.185 brouard 501: Revision 1.184 2015/03/11 11:52:39 brouard
502: Summary: Back from Windows 8. Intel Compiler
503:
1.184 brouard 504: Revision 1.183 2015/03/10 20:34:32 brouard
505: Summary: 0.98q0, trying with directest, mnbrak fixed
506:
507: We use directest instead of original Powell test; probably no
508: incidence on the results, but better justifications;
509: We fixed Numerical Recipes mnbrak routine which was wrong and gave
510: wrong results.
511:
1.183 brouard 512: Revision 1.182 2015/02/12 08:19:57 brouard
513: Summary: Trying to keep directest which seems simpler and more general
514: Author: Nicolas Brouard
515:
1.182 brouard 516: Revision 1.181 2015/02/11 23:22:24 brouard
517: Summary: Comments on Powell added
518:
519: Author:
520:
1.181 brouard 521: Revision 1.180 2015/02/11 17:33:45 brouard
522: Summary: Finishing move from main to function (hpijx and prevalence_limit)
523:
1.180 brouard 524: Revision 1.179 2015/01/04 09:57:06 brouard
525: Summary: back to OS/X
526:
1.179 brouard 527: Revision 1.178 2015/01/04 09:35:48 brouard
528: *** empty log message ***
529:
1.178 brouard 530: Revision 1.177 2015/01/03 18:40:56 brouard
531: Summary: Still testing ilc32 on OSX
532:
1.177 brouard 533: Revision 1.176 2015/01/03 16:45:04 brouard
534: *** empty log message ***
535:
1.176 brouard 536: Revision 1.175 2015/01/03 16:33:42 brouard
537: *** empty log message ***
538:
1.175 brouard 539: Revision 1.174 2015/01/03 16:15:49 brouard
540: Summary: Still in cross-compilation
541:
1.174 brouard 542: Revision 1.173 2015/01/03 12:06:26 brouard
543: Summary: trying to detect cross-compilation
544:
1.173 brouard 545: Revision 1.172 2014/12/27 12:07:47 brouard
546: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
547:
1.172 brouard 548: Revision 1.171 2014/12/23 13:26:59 brouard
549: Summary: Back from Visual C
550:
551: Still problem with utsname.h on Windows
552:
1.171 brouard 553: Revision 1.170 2014/12/23 11:17:12 brouard
554: Summary: Cleaning some \%% back to %%
555:
556: The escape was mandatory for a specific compiler (which one?), but too many warnings.
557:
1.170 brouard 558: Revision 1.169 2014/12/22 23:08:31 brouard
559: Summary: 0.98p
560:
561: Outputs some informations on compiler used, OS etc. Testing on different platforms.
562:
1.169 brouard 563: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 564: Summary: update
1.169 brouard 565:
1.168 brouard 566: Revision 1.167 2014/12/22 13:50:56 brouard
567: Summary: Testing uname and compiler version and if compiled 32 or 64
568:
569: Testing on Linux 64
570:
1.167 brouard 571: Revision 1.166 2014/12/22 11:40:47 brouard
572: *** empty log message ***
573:
1.166 brouard 574: Revision 1.165 2014/12/16 11:20:36 brouard
575: Summary: After compiling on Visual C
576:
577: * imach.c (Module): Merging 1.61 to 1.162
578:
1.165 brouard 579: Revision 1.164 2014/12/16 10:52:11 brouard
580: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
581:
582: * imach.c (Module): Merging 1.61 to 1.162
583:
1.164 brouard 584: Revision 1.163 2014/12/16 10:30:11 brouard
585: * imach.c (Module): Merging 1.61 to 1.162
586:
1.163 brouard 587: Revision 1.162 2014/09/25 11:43:39 brouard
588: Summary: temporary backup 0.99!
589:
1.162 brouard 590: Revision 1.1 2014/09/16 11:06:58 brouard
591: Summary: With some code (wrong) for nlopt
592:
593: Author:
594:
595: Revision 1.161 2014/09/15 20:41:41 brouard
596: Summary: Problem with macro SQR on Intel compiler
597:
1.161 brouard 598: Revision 1.160 2014/09/02 09:24:05 brouard
599: *** empty log message ***
600:
1.160 brouard 601: Revision 1.159 2014/09/01 10:34:10 brouard
602: Summary: WIN32
603: Author: Brouard
604:
1.159 brouard 605: Revision 1.158 2014/08/27 17:11:51 brouard
606: *** empty log message ***
607:
1.158 brouard 608: Revision 1.157 2014/08/27 16:26:55 brouard
609: Summary: Preparing windows Visual studio version
610: Author: Brouard
611:
612: In order to compile on Visual studio, time.h is now correct and time_t
613: and tm struct should be used. difftime should be used but sometimes I
614: just make the differences in raw time format (time(&now).
615: Trying to suppress #ifdef LINUX
616: Add xdg-open for __linux in order to open default browser.
617:
1.157 brouard 618: Revision 1.156 2014/08/25 20:10:10 brouard
619: *** empty log message ***
620:
1.156 brouard 621: Revision 1.155 2014/08/25 18:32:34 brouard
622: Summary: New compile, minor changes
623: Author: Brouard
624:
1.155 brouard 625: Revision 1.154 2014/06/20 17:32:08 brouard
626: Summary: Outputs now all graphs of convergence to period prevalence
627:
1.154 brouard 628: Revision 1.153 2014/06/20 16:45:46 brouard
629: Summary: If 3 live state, convergence to period prevalence on same graph
630: Author: Brouard
631:
1.153 brouard 632: Revision 1.152 2014/06/18 17:54:09 brouard
633: Summary: open browser, use gnuplot on same dir than imach if not found in the path
634:
1.152 brouard 635: Revision 1.151 2014/06/18 16:43:30 brouard
636: *** empty log message ***
637:
1.151 brouard 638: Revision 1.150 2014/06/18 16:42:35 brouard
639: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
640: Author: brouard
641:
1.150 brouard 642: Revision 1.149 2014/06/18 15:51:14 brouard
643: Summary: Some fixes in parameter files errors
644: Author: Nicolas Brouard
645:
1.149 brouard 646: Revision 1.148 2014/06/17 17:38:48 brouard
647: Summary: Nothing new
648: Author: Brouard
649:
650: Just a new packaging for OS/X version 0.98nS
651:
1.148 brouard 652: Revision 1.147 2014/06/16 10:33:11 brouard
653: *** empty log message ***
654:
1.147 brouard 655: Revision 1.146 2014/06/16 10:20:28 brouard
656: Summary: Merge
657: Author: Brouard
658:
659: Merge, before building revised version.
660:
1.146 brouard 661: Revision 1.145 2014/06/10 21:23:15 brouard
662: Summary: Debugging with valgrind
663: Author: Nicolas Brouard
664:
665: Lot of changes in order to output the results with some covariates
666: After the Edimburgh REVES conference 2014, it seems mandatory to
667: improve the code.
668: No more memory valgrind error but a lot has to be done in order to
669: continue the work of splitting the code into subroutines.
670: Also, decodemodel has been improved. Tricode is still not
671: optimal. nbcode should be improved. Documentation has been added in
672: the source code.
673:
1.144 brouard 674: Revision 1.143 2014/01/26 09:45:38 brouard
675: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
676:
677: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
678: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
679:
1.143 brouard 680: Revision 1.142 2014/01/26 03:57:36 brouard
681: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
682:
683: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
684:
1.142 brouard 685: Revision 1.141 2014/01/26 02:42:01 brouard
686: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
687:
1.141 brouard 688: Revision 1.140 2011/09/02 10:37:54 brouard
689: Summary: times.h is ok with mingw32 now.
690:
1.140 brouard 691: Revision 1.139 2010/06/14 07:50:17 brouard
692: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
693: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
694:
1.139 brouard 695: Revision 1.138 2010/04/30 18:19:40 brouard
696: *** empty log message ***
697:
1.138 brouard 698: Revision 1.137 2010/04/29 18:11:38 brouard
699: (Module): Checking covariates for more complex models
700: than V1+V2. A lot of change to be done. Unstable.
701:
1.137 brouard 702: Revision 1.136 2010/04/26 20:30:53 brouard
703: (Module): merging some libgsl code. Fixing computation
704: of likelione (using inter/intrapolation if mle = 0) in order to
705: get same likelihood as if mle=1.
706: Some cleaning of code and comments added.
707:
1.136 brouard 708: Revision 1.135 2009/10/29 15:33:14 brouard
709: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
710:
1.135 brouard 711: Revision 1.134 2009/10/29 13:18:53 brouard
712: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
713:
1.134 brouard 714: Revision 1.133 2009/07/06 10:21:25 brouard
715: just nforces
716:
1.133 brouard 717: Revision 1.132 2009/07/06 08:22:05 brouard
718: Many tings
719:
1.132 brouard 720: Revision 1.131 2009/06/20 16:22:47 brouard
721: Some dimensions resccaled
722:
1.131 brouard 723: Revision 1.130 2009/05/26 06:44:34 brouard
724: (Module): Max Covariate is now set to 20 instead of 8. A
725: lot of cleaning with variables initialized to 0. Trying to make
726: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
727:
1.130 brouard 728: Revision 1.129 2007/08/31 13:49:27 lievre
729: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
730:
1.129 lievre 731: Revision 1.128 2006/06/30 13:02:05 brouard
732: (Module): Clarifications on computing e.j
733:
1.128 brouard 734: Revision 1.127 2006/04/28 18:11:50 brouard
735: (Module): Yes the sum of survivors was wrong since
736: imach-114 because nhstepm was no more computed in the age
737: loop. Now we define nhstepma in the age loop.
738: (Module): In order to speed up (in case of numerous covariates) we
739: compute health expectancies (without variances) in a first step
740: and then all the health expectancies with variances or standard
741: deviation (needs data from the Hessian matrices) which slows the
742: computation.
743: In the future we should be able to stop the program is only health
744: expectancies and graph are needed without standard deviations.
745:
1.127 brouard 746: Revision 1.126 2006/04/28 17:23:28 brouard
747: (Module): Yes the sum of survivors was wrong since
748: imach-114 because nhstepm was no more computed in the age
749: loop. Now we define nhstepma in the age loop.
750: Version 0.98h
751:
1.126 brouard 752: Revision 1.125 2006/04/04 15:20:31 lievre
753: Errors in calculation of health expectancies. Age was not initialized.
754: Forecasting file added.
755:
756: Revision 1.124 2006/03/22 17:13:53 lievre
757: Parameters are printed with %lf instead of %f (more numbers after the comma).
758: The log-likelihood is printed in the log file
759:
760: Revision 1.123 2006/03/20 10:52:43 brouard
761: * imach.c (Module): <title> changed, corresponds to .htm file
762: name. <head> headers where missing.
763:
764: * imach.c (Module): Weights can have a decimal point as for
765: English (a comma might work with a correct LC_NUMERIC environment,
766: otherwise the weight is truncated).
767: Modification of warning when the covariates values are not 0 or
768: 1.
769: Version 0.98g
770:
771: Revision 1.122 2006/03/20 09:45:41 brouard
772: (Module): Weights can have a decimal point as for
773: English (a comma might work with a correct LC_NUMERIC environment,
774: otherwise the weight is truncated).
775: Modification of warning when the covariates values are not 0 or
776: 1.
777: Version 0.98g
778:
779: Revision 1.121 2006/03/16 17:45:01 lievre
780: * imach.c (Module): Comments concerning covariates added
781:
782: * imach.c (Module): refinements in the computation of lli if
783: status=-2 in order to have more reliable computation if stepm is
784: not 1 month. Version 0.98f
785:
786: Revision 1.120 2006/03/16 15:10:38 lievre
787: (Module): refinements in the computation of lli if
788: status=-2 in order to have more reliable computation if stepm is
789: not 1 month. Version 0.98f
790:
791: Revision 1.119 2006/03/15 17:42:26 brouard
792: (Module): Bug if status = -2, the loglikelihood was
793: computed as likelihood omitting the logarithm. Version O.98e
794:
795: Revision 1.118 2006/03/14 18:20:07 brouard
796: (Module): varevsij Comments added explaining the second
797: table of variances if popbased=1 .
798: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
799: (Module): Function pstamp added
800: (Module): Version 0.98d
801:
802: Revision 1.117 2006/03/14 17:16:22 brouard
803: (Module): varevsij Comments added explaining the second
804: table of variances if popbased=1 .
805: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
806: (Module): Function pstamp added
807: (Module): Version 0.98d
808:
809: Revision 1.116 2006/03/06 10:29:27 brouard
810: (Module): Variance-covariance wrong links and
811: varian-covariance of ej. is needed (Saito).
812:
813: Revision 1.115 2006/02/27 12:17:45 brouard
814: (Module): One freematrix added in mlikeli! 0.98c
815:
816: Revision 1.114 2006/02/26 12:57:58 brouard
817: (Module): Some improvements in processing parameter
818: filename with strsep.
819:
820: Revision 1.113 2006/02/24 14:20:24 brouard
821: (Module): Memory leaks checks with valgrind and:
822: datafile was not closed, some imatrix were not freed and on matrix
823: allocation too.
824:
825: Revision 1.112 2006/01/30 09:55:26 brouard
826: (Module): Back to gnuplot.exe instead of wgnuplot.exe
827:
828: Revision 1.111 2006/01/25 20:38:18 brouard
829: (Module): Lots of cleaning and bugs added (Gompertz)
830: (Module): Comments can be added in data file. Missing date values
831: can be a simple dot '.'.
832:
833: Revision 1.110 2006/01/25 00:51:50 brouard
834: (Module): Lots of cleaning and bugs added (Gompertz)
835:
836: Revision 1.109 2006/01/24 19:37:15 brouard
837: (Module): Comments (lines starting with a #) are allowed in data.
838:
839: Revision 1.108 2006/01/19 18:05:42 lievre
840: Gnuplot problem appeared...
841: To be fixed
842:
843: Revision 1.107 2006/01/19 16:20:37 brouard
844: Test existence of gnuplot in imach path
845:
846: Revision 1.106 2006/01/19 13:24:36 brouard
847: Some cleaning and links added in html output
848:
849: Revision 1.105 2006/01/05 20:23:19 lievre
850: *** empty log message ***
851:
852: Revision 1.104 2005/09/30 16:11:43 lievre
853: (Module): sump fixed, loop imx fixed, and simplifications.
854: (Module): If the status is missing at the last wave but we know
855: that the person is alive, then we can code his/her status as -2
856: (instead of missing=-1 in earlier versions) and his/her
857: contributions to the likelihood is 1 - Prob of dying from last
858: health status (= 1-p13= p11+p12 in the easiest case of somebody in
859: the healthy state at last known wave). Version is 0.98
860:
861: Revision 1.103 2005/09/30 15:54:49 lievre
862: (Module): sump fixed, loop imx fixed, and simplifications.
863:
864: Revision 1.102 2004/09/15 17:31:30 brouard
865: Add the possibility to read data file including tab characters.
866:
867: Revision 1.101 2004/09/15 10:38:38 brouard
868: Fix on curr_time
869:
870: Revision 1.100 2004/07/12 18:29:06 brouard
871: Add version for Mac OS X. Just define UNIX in Makefile
872:
873: Revision 1.99 2004/06/05 08:57:40 brouard
874: *** empty log message ***
875:
876: Revision 1.98 2004/05/16 15:05:56 brouard
877: New version 0.97 . First attempt to estimate force of mortality
878: directly from the data i.e. without the need of knowing the health
879: state at each age, but using a Gompertz model: log u =a + b*age .
880: This is the basic analysis of mortality and should be done before any
881: other analysis, in order to test if the mortality estimated from the
882: cross-longitudinal survey is different from the mortality estimated
883: from other sources like vital statistic data.
884:
885: The same imach parameter file can be used but the option for mle should be -3.
886:
1.324 brouard 887: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 888: former routines in order to include the new code within the former code.
889:
890: The output is very simple: only an estimate of the intercept and of
891: the slope with 95% confident intervals.
892:
893: Current limitations:
894: A) Even if you enter covariates, i.e. with the
895: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
896: B) There is no computation of Life Expectancy nor Life Table.
897:
898: Revision 1.97 2004/02/20 13:25:42 lievre
899: Version 0.96d. Population forecasting command line is (temporarily)
900: suppressed.
901:
902: Revision 1.96 2003/07/15 15:38:55 brouard
903: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
904: rewritten within the same printf. Workaround: many printfs.
905:
906: Revision 1.95 2003/07/08 07:54:34 brouard
907: * imach.c (Repository):
908: (Repository): Using imachwizard code to output a more meaningful covariance
909: matrix (cov(a12,c31) instead of numbers.
910:
911: Revision 1.94 2003/06/27 13:00:02 brouard
912: Just cleaning
913:
914: Revision 1.93 2003/06/25 16:33:55 brouard
915: (Module): On windows (cygwin) function asctime_r doesn't
916: exist so I changed back to asctime which exists.
917: (Module): Version 0.96b
918:
919: Revision 1.92 2003/06/25 16:30:45 brouard
920: (Module): On windows (cygwin) function asctime_r doesn't
921: exist so I changed back to asctime which exists.
922:
923: Revision 1.91 2003/06/25 15:30:29 brouard
924: * imach.c (Repository): Duplicated warning errors corrected.
925: (Repository): Elapsed time after each iteration is now output. It
926: helps to forecast when convergence will be reached. Elapsed time
927: is stamped in powell. We created a new html file for the graphs
928: concerning matrix of covariance. It has extension -cov.htm.
929:
930: Revision 1.90 2003/06/24 12:34:15 brouard
931: (Module): Some bugs corrected for windows. Also, when
932: mle=-1 a template is output in file "or"mypar.txt with the design
933: of the covariance matrix to be input.
934:
935: Revision 1.89 2003/06/24 12:30:52 brouard
936: (Module): Some bugs corrected for windows. Also, when
937: mle=-1 a template is output in file "or"mypar.txt with the design
938: of the covariance matrix to be input.
939:
940: Revision 1.88 2003/06/23 17:54:56 brouard
941: * 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.
942:
943: Revision 1.87 2003/06/18 12:26:01 brouard
944: Version 0.96
945:
946: Revision 1.86 2003/06/17 20:04:08 brouard
947: (Module): Change position of html and gnuplot routines and added
948: routine fileappend.
949:
950: Revision 1.85 2003/06/17 13:12:43 brouard
951: * imach.c (Repository): Check when date of death was earlier that
952: current date of interview. It may happen when the death was just
953: prior to the death. In this case, dh was negative and likelihood
954: was wrong (infinity). We still send an "Error" but patch by
955: assuming that the date of death was just one stepm after the
956: interview.
957: (Repository): Because some people have very long ID (first column)
958: we changed int to long in num[] and we added a new lvector for
959: memory allocation. But we also truncated to 8 characters (left
960: truncation)
961: (Repository): No more line truncation errors.
962:
963: Revision 1.84 2003/06/13 21:44:43 brouard
964: * imach.c (Repository): Replace "freqsummary" at a correct
965: place. It differs from routine "prevalence" which may be called
966: many times. Probs is memory consuming and must be used with
967: parcimony.
968: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
969:
970: Revision 1.83 2003/06/10 13:39:11 lievre
971: *** empty log message ***
972:
973: Revision 1.82 2003/06/05 15:57:20 brouard
974: Add log in imach.c and fullversion number is now printed.
975:
976: */
977: /*
978: Interpolated Markov Chain
979:
980: Short summary of the programme:
981:
1.227 brouard 982: This program computes Healthy Life Expectancies or State-specific
983: (if states aren't health statuses) Expectancies from
984: cross-longitudinal data. Cross-longitudinal data consist in:
985:
986: -1- a first survey ("cross") where individuals from different ages
987: are interviewed on their health status or degree of disability (in
988: the case of a health survey which is our main interest)
989:
990: -2- at least a second wave of interviews ("longitudinal") which
991: measure each change (if any) in individual health status. Health
992: expectancies are computed from the time spent in each health state
993: according to a model. More health states you consider, more time is
994: necessary to reach the Maximum Likelihood of the parameters involved
995: in the model. The simplest model is the multinomial logistic model
996: where pij is the probability to be observed in state j at the second
997: wave conditional to be observed in state i at the first
998: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
999: etc , where 'age' is age and 'sex' is a covariate. If you want to
1000: have a more complex model than "constant and age", you should modify
1001: the program where the markup *Covariates have to be included here
1002: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1003: convergence.
1004:
1005: The advantage of this computer programme, compared to a simple
1006: multinomial logistic model, is clear when the delay between waves is not
1007: identical for each individual. Also, if a individual missed an
1008: intermediate interview, the information is lost, but taken into
1009: account using an interpolation or extrapolation.
1010:
1011: hPijx is the probability to be observed in state i at age x+h
1012: conditional to the observed state i at age x. The delay 'h' can be
1013: split into an exact number (nh*stepm) of unobserved intermediate
1014: states. This elementary transition (by month, quarter,
1015: semester or year) is modelled as a multinomial logistic. The hPx
1016: matrix is simply the matrix product of nh*stepm elementary matrices
1017: and the contribution of each individual to the likelihood is simply
1018: hPijx.
1019:
1020: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1021: of the life expectancies. It also computes the period (stable) prevalence.
1022:
1023: Back prevalence and projections:
1.227 brouard 1024:
1025: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1026: double agemaxpar, double ftolpl, int *ncvyearp, double
1027: dateprev1,double dateprev2, int firstpass, int lastpass, int
1028: mobilavproj)
1029:
1030: Computes the back prevalence limit for any combination of
1031: covariate values k at any age between ageminpar and agemaxpar and
1032: returns it in **bprlim. In the loops,
1033:
1034: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1035: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1036:
1037: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1038: Computes for any combination of covariates k and any age between bage and fage
1039: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1040: oldm=oldms;savm=savms;
1.227 brouard 1041:
1.267 brouard 1042: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1043: Computes the transition matrix starting at age 'age' over
1044: 'nhstepm*hstepm*stepm' months (i.e. until
1045: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1046: nhstepm*hstepm matrices.
1047:
1048: Returns p3mat[i][j][h] after calling
1049: p3mat[i][j][h]=matprod2(newm,
1050: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1051: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1052: oldm);
1.226 brouard 1053:
1054: Important routines
1055:
1056: - func (or funcone), computes logit (pij) distinguishing
1057: o fixed variables (single or product dummies or quantitative);
1058: o varying variables by:
1059: (1) wave (single, product dummies, quantitative),
1060: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1061: % fixed dummy (treated) or quantitative (not done because time-consuming);
1062: % varying dummy (not done) or quantitative (not done);
1063: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1064: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1065: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1066: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1067: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1068:
1.226 brouard 1069:
1070:
1.324 brouard 1071: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1072: Institut national d'études démographiques, Paris.
1.126 brouard 1073: This software have been partly granted by Euro-REVES, a concerted action
1074: from the European Union.
1075: It is copyrighted identically to a GNU software product, ie programme and
1076: software can be distributed freely for non commercial use. Latest version
1077: can be accessed at http://euroreves.ined.fr/imach .
1078:
1079: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1080: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1081:
1082: **********************************************************************/
1083: /*
1084: main
1085: read parameterfile
1086: read datafile
1087: concatwav
1088: freqsummary
1089: if (mle >= 1)
1090: mlikeli
1091: print results files
1092: if mle==1
1093: computes hessian
1094: read end of parameter file: agemin, agemax, bage, fage, estepm
1095: begin-prev-date,...
1096: open gnuplot file
1097: open html file
1.145 brouard 1098: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1099: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1100: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1101: freexexit2 possible for memory heap.
1102:
1103: h Pij x | pij_nom ficrestpij
1104: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1105: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1106: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1107:
1108: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1109: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1110: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1111: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1112: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1113:
1.126 brouard 1114: forecasting if prevfcast==1 prevforecast call prevalence()
1115: health expectancies
1116: Variance-covariance of DFLE
1117: prevalence()
1118: movingaverage()
1119: varevsij()
1120: if popbased==1 varevsij(,popbased)
1121: total life expectancies
1122: Variance of period (stable) prevalence
1123: end
1124: */
1125:
1.187 brouard 1126: /* #define DEBUG */
1127: /* #define DEBUGBRENT */
1.203 brouard 1128: /* #define DEBUGLINMIN */
1129: /* #define DEBUGHESS */
1130: #define DEBUGHESSIJ
1.224 brouard 1131: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1132: #define POWELL /* Instead of NLOPT */
1.224 brouard 1133: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1134: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1135: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1136: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1137:
1138: #include <math.h>
1139: #include <stdio.h>
1140: #include <stdlib.h>
1141: #include <string.h>
1.226 brouard 1142: #include <ctype.h>
1.159 brouard 1143:
1144: #ifdef _WIN32
1145: #include <io.h>
1.172 brouard 1146: #include <windows.h>
1147: #include <tchar.h>
1.159 brouard 1148: #else
1.126 brouard 1149: #include <unistd.h>
1.159 brouard 1150: #endif
1.126 brouard 1151:
1152: #include <limits.h>
1153: #include <sys/types.h>
1.171 brouard 1154:
1155: #if defined(__GNUC__)
1156: #include <sys/utsname.h> /* Doesn't work on Windows */
1157: #endif
1158:
1.126 brouard 1159: #include <sys/stat.h>
1160: #include <errno.h>
1.159 brouard 1161: /* extern int errno; */
1.126 brouard 1162:
1.157 brouard 1163: /* #ifdef LINUX */
1164: /* #include <time.h> */
1165: /* #include "timeval.h" */
1166: /* #else */
1167: /* #include <sys/time.h> */
1168: /* #endif */
1169:
1.126 brouard 1170: #include <time.h>
1171:
1.136 brouard 1172: #ifdef GSL
1173: #include <gsl/gsl_errno.h>
1174: #include <gsl/gsl_multimin.h>
1175: #endif
1176:
1.167 brouard 1177:
1.162 brouard 1178: #ifdef NLOPT
1179: #include <nlopt.h>
1180: typedef struct {
1181: double (* function)(double [] );
1182: } myfunc_data ;
1183: #endif
1184:
1.126 brouard 1185: /* #include <libintl.h> */
1186: /* #define _(String) gettext (String) */
1187:
1.251 brouard 1188: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1189:
1190: #define GNUPLOTPROGRAM "gnuplot"
1191: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1192: #define FILENAMELENGTH 132
1193:
1194: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1195: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1196:
1.144 brouard 1197: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1198: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1199:
1200: #define NINTERVMAX 8
1.144 brouard 1201: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1202: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1203: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1204: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1205: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1206: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1207: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1208: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1209: /* #define AGESUP 130 */
1.288 brouard 1210: /* #define AGESUP 150 */
1211: #define AGESUP 200
1.268 brouard 1212: #define AGEINF 0
1.218 brouard 1213: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1214: #define AGEBASE 40
1.194 brouard 1215: #define AGEOVERFLOW 1.e20
1.164 brouard 1216: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1217: #ifdef _WIN32
1218: #define DIRSEPARATOR '\\'
1219: #define CHARSEPARATOR "\\"
1220: #define ODIRSEPARATOR '/'
1221: #else
1.126 brouard 1222: #define DIRSEPARATOR '/'
1223: #define CHARSEPARATOR "/"
1224: #define ODIRSEPARATOR '\\'
1225: #endif
1226:
1.328 ! brouard 1227: /* $Id: imach.c,v 1.327 2022/07/27 14:47:35 brouard Exp $ */
1.126 brouard 1228: /* $State: Exp $ */
1.196 brouard 1229: #include "version.h"
1230: char version[]=__IMACH_VERSION__;
1.323 brouard 1231: 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";
1.328 ! brouard 1232: char fullversion[]="$Revision: 1.327 $ $Date: 2022/07/27 14:47:35 $";
1.126 brouard 1233: char strstart[80];
1234: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1235: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1236: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1237: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1238: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1239: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1240: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1241: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1242: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1243: int cptcovprodnoage=0; /**< Number of covariate products without age */
1244: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1245: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1246: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1247: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1248: int nsd=0; /**< Total number of single dummy variables (output) */
1249: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1250: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1251: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1252: int ntveff=0; /**< ntveff number of effective time varying variables */
1253: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1254: int cptcov=0; /* Working variable */
1.290 brouard 1255: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1256: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1257: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1258: int nlstate=2; /* Number of live states */
1259: int ndeath=1; /* Number of dead states */
1.130 brouard 1260: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1261: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1262: int popbased=0;
1263:
1264: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1265: int maxwav=0; /* Maxim number of waves */
1266: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1267: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1268: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1269: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1270: int mle=1, weightopt=0;
1.126 brouard 1271: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1272: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1273: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1274: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1275: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1276: int selected(int kvar); /* Is covariate kvar selected for printing results */
1277:
1.130 brouard 1278: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1279: double **matprod2(); /* test */
1.126 brouard 1280: double **oldm, **newm, **savm; /* Working pointers to matrices */
1281: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1282: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1283:
1.136 brouard 1284: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1285: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1286: FILE *ficlog, *ficrespow;
1.130 brouard 1287: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1288: double fretone; /* Only one call to likelihood */
1.130 brouard 1289: long ipmx=0; /* Number of contributions */
1.126 brouard 1290: double sw; /* Sum of weights */
1291: char filerespow[FILENAMELENGTH];
1292: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1293: FILE *ficresilk;
1294: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1295: FILE *ficresprobmorprev;
1296: FILE *fichtm, *fichtmcov; /* Html File */
1297: FILE *ficreseij;
1298: char filerese[FILENAMELENGTH];
1299: FILE *ficresstdeij;
1300: char fileresstde[FILENAMELENGTH];
1301: FILE *ficrescveij;
1302: char filerescve[FILENAMELENGTH];
1303: FILE *ficresvij;
1304: char fileresv[FILENAMELENGTH];
1.269 brouard 1305:
1.126 brouard 1306: char title[MAXLINE];
1.234 brouard 1307: char model[MAXLINE]; /**< The model line */
1.217 brouard 1308: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1309: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1310: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1311: char command[FILENAMELENGTH];
1312: int outcmd=0;
1313:
1.217 brouard 1314: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1315: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1316: char filelog[FILENAMELENGTH]; /* Log file */
1317: char filerest[FILENAMELENGTH];
1318: char fileregp[FILENAMELENGTH];
1319: char popfile[FILENAMELENGTH];
1320:
1321: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1322:
1.157 brouard 1323: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1324: /* struct timezone tzp; */
1325: /* extern int gettimeofday(); */
1326: struct tm tml, *gmtime(), *localtime();
1327:
1328: extern time_t time();
1329:
1330: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1331: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1332: struct tm tm;
1333:
1.126 brouard 1334: char strcurr[80], strfor[80];
1335:
1336: char *endptr;
1337: long lval;
1338: double dval;
1339:
1340: #define NR_END 1
1341: #define FREE_ARG char*
1342: #define FTOL 1.0e-10
1343:
1344: #define NRANSI
1.240 brouard 1345: #define ITMAX 200
1346: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1347:
1348: #define TOL 2.0e-4
1349:
1350: #define CGOLD 0.3819660
1351: #define ZEPS 1.0e-10
1352: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1353:
1354: #define GOLD 1.618034
1355: #define GLIMIT 100.0
1356: #define TINY 1.0e-20
1357:
1358: static double maxarg1,maxarg2;
1359: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1360: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1361:
1362: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1363: #define rint(a) floor(a+0.5)
1.166 brouard 1364: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1365: #define mytinydouble 1.0e-16
1.166 brouard 1366: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1367: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1368: /* static double dsqrarg; */
1369: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1370: static double sqrarg;
1371: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1372: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1373: int agegomp= AGEGOMP;
1374:
1375: int imx;
1376: int stepm=1;
1377: /* Stepm, step in month: minimum step interpolation*/
1378:
1379: int estepm;
1380: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1381:
1382: int m,nb;
1383: long *num;
1.197 brouard 1384: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1385: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1386: covariate for which somebody answered excluding
1387: undefined. Usually 2: 0 and 1. */
1388: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1389: covariate for which somebody answered including
1390: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1391: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1392: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1393: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1394: double *ageexmed,*agecens;
1395: double dateintmean=0;
1.296 brouard 1396: double anprojd, mprojd, jprojd; /* For eventual projections */
1397: double anprojf, mprojf, jprojf;
1.126 brouard 1398:
1.296 brouard 1399: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1400: double anbackf, mbackf, jbackf;
1401: double jintmean,mintmean,aintmean;
1.126 brouard 1402: double *weight;
1403: int **s; /* Status */
1.141 brouard 1404: double *agedc;
1.145 brouard 1405: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1406: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1407: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1408: double **coqvar; /* Fixed quantitative covariate nqv */
1409: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1410: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1411: double idx;
1412: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1413: /* Some documentation */
1414: /* Design original data
1415: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1416: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1417: * ntv=3 nqtv=1
1418: * cptcovn number of covariates (not including constant and age) = # of + plus 1 = 10+1=11
1419: * For time varying covariate, quanti or dummies
1420: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1421: * cotvar[wav][ntv+iv][i]= [3+(1 to nqtv)][i]=(V12) quanti
1422: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1423: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1424: * covar[k,i], value of kth fixed covariate dummy or quanti :
1425: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1426: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1427: * k= 1 2 3 4 5 6 7 8 9 10 11
1428: */
1429: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1430: /* 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
1431: # States 1=Coresidence, 2 Living alone, 3 Institution
1432: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1433: */
1.319 brouard 1434: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1435: /* k 1 2 3 4 5 6 7 8 9 */
1436: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1437: /* fixed or varying), 1 for age product, 2 for*/
1438: /* product */
1439: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1440: /*(single or product without age), 2 dummy*/
1441: /* with age product, 3 quant with age product*/
1442: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1443: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1444: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1445: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1446: /* nsq 1 2 */ /* Counting single quantit tv */
1447: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1448: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1449: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1450: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1451: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1452: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1453: /* 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 1454: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1455: /* Type */
1456: /* V 1 2 3 4 5 */
1457: /* F F V V V */
1458: /* D Q D D Q */
1459: /* */
1460: int *TvarsD;
1461: int *TvarsDind;
1462: int *TvarsQ;
1463: int *TvarsQind;
1464:
1.318 brouard 1465: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1466: int nresult=0;
1.258 brouard 1467: int parameterline=0; /* # of the parameter (type) line */
1.318 brouard 1468: int TKresult[MAXRESULTLINESPONE];
1469: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1470: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1471: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For dummy variable , variable # (output) */
1472: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1473: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1474: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , variable # (output) */
1475:
1476: /* 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
1477: # States 1=Coresidence, 2 Living alone, 3 Institution
1478: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1479: */
1.234 brouard 1480: /* 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 1481: 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 */
1482: 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 */
1483: 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 */
1484: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1485: 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 */
1486: 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 1487: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1488: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1489: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1490: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1491: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1492: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1493: 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 */
1494: 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 */
1495:
1.230 brouard 1496: int *Tvarsel; /**< Selected covariates for output */
1497: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1498: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1499: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1500: 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 1501: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1502: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1503: int *Tage;
1.227 brouard 1504: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1505: 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 1506: 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*/
1507: 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 1508: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1509: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1510: int **Tvard;
1511: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1512: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1513: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1514: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1515: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1516: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1517: double *lsurv, *lpop, *tpop;
1518:
1.231 brouard 1519: #define FD 1; /* Fixed dummy covariate */
1520: #define FQ 2; /* Fixed quantitative covariate */
1521: #define FP 3; /* Fixed product covariate */
1522: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1523: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1524: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1525: #define VD 10; /* Varying dummy covariate */
1526: #define VQ 11; /* Varying quantitative covariate */
1527: #define VP 12; /* Varying product covariate */
1528: #define VPDD 13; /* Varying product dummy*dummy covariate */
1529: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1530: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1531: #define APFD 16; /* Age product * fixed dummy covariate */
1532: #define APFQ 17; /* Age product * fixed quantitative covariate */
1533: #define APVD 18; /* Age product * varying dummy covariate */
1534: #define APVQ 19; /* Age product * varying quantitative covariate */
1535:
1536: #define FTYPE 1; /* Fixed covariate */
1537: #define VTYPE 2; /* Varying covariate (loop in wave) */
1538: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1539:
1540: struct kmodel{
1541: int maintype; /* main type */
1542: int subtype; /* subtype */
1543: };
1544: struct kmodel modell[NCOVMAX];
1545:
1.143 brouard 1546: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1547: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1548:
1549: /**************** split *************************/
1550: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1551: {
1552: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1553: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1554: */
1555: char *ss; /* pointer */
1.186 brouard 1556: int l1=0, l2=0; /* length counters */
1.126 brouard 1557:
1558: l1 = strlen(path ); /* length of path */
1559: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1560: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1561: if ( ss == NULL ) { /* no directory, so determine current directory */
1562: strcpy( name, path ); /* we got the fullname name because no directory */
1563: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1564: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1565: /* get current working directory */
1566: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1567: #ifdef WIN32
1568: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1569: #else
1570: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1571: #endif
1.126 brouard 1572: return( GLOCK_ERROR_GETCWD );
1573: }
1574: /* got dirc from getcwd*/
1575: printf(" DIRC = %s \n",dirc);
1.205 brouard 1576: } else { /* strip directory from path */
1.126 brouard 1577: ss++; /* after this, the filename */
1578: l2 = strlen( ss ); /* length of filename */
1579: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1580: strcpy( name, ss ); /* save file name */
1581: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1582: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1583: printf(" DIRC2 = %s \n",dirc);
1584: }
1585: /* We add a separator at the end of dirc if not exists */
1586: l1 = strlen( dirc ); /* length of directory */
1587: if( dirc[l1-1] != DIRSEPARATOR ){
1588: dirc[l1] = DIRSEPARATOR;
1589: dirc[l1+1] = 0;
1590: printf(" DIRC3 = %s \n",dirc);
1591: }
1592: ss = strrchr( name, '.' ); /* find last / */
1593: if (ss >0){
1594: ss++;
1595: strcpy(ext,ss); /* save extension */
1596: l1= strlen( name);
1597: l2= strlen(ss)+1;
1598: strncpy( finame, name, l1-l2);
1599: finame[l1-l2]= 0;
1600: }
1601:
1602: return( 0 ); /* we're done */
1603: }
1604:
1605:
1606: /******************************************/
1607:
1608: void replace_back_to_slash(char *s, char*t)
1609: {
1610: int i;
1611: int lg=0;
1612: i=0;
1613: lg=strlen(t);
1614: for(i=0; i<= lg; i++) {
1615: (s[i] = t[i]);
1616: if (t[i]== '\\') s[i]='/';
1617: }
1618: }
1619:
1.132 brouard 1620: char *trimbb(char *out, char *in)
1.137 brouard 1621: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1622: char *s;
1623: s=out;
1624: while (*in != '\0'){
1.137 brouard 1625: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1626: in++;
1627: }
1628: *out++ = *in++;
1629: }
1630: *out='\0';
1631: return s;
1632: }
1633:
1.187 brouard 1634: /* char *substrchaine(char *out, char *in, char *chain) */
1635: /* { */
1636: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1637: /* char *s, *t; */
1638: /* t=in;s=out; */
1639: /* while ((*in != *chain) && (*in != '\0')){ */
1640: /* *out++ = *in++; */
1641: /* } */
1642:
1643: /* /\* *in matches *chain *\/ */
1644: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1645: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1646: /* } */
1647: /* in--; chain--; */
1648: /* while ( (*in != '\0')){ */
1649: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1650: /* *out++ = *in++; */
1651: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1652: /* } */
1653: /* *out='\0'; */
1654: /* out=s; */
1655: /* return out; */
1656: /* } */
1657: char *substrchaine(char *out, char *in, char *chain)
1658: {
1659: /* Substract chain 'chain' from 'in', return and output 'out' */
1660: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1661:
1662: char *strloc;
1663:
1664: strcpy (out, in);
1665: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1666: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1667: if(strloc != NULL){
1668: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1669: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1670: /* strcpy (strloc, strloc +strlen(chain));*/
1671: }
1672: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1673: return out;
1674: }
1675:
1676:
1.145 brouard 1677: char *cutl(char *blocc, char *alocc, char *in, char occ)
1678: {
1.187 brouard 1679: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1680: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1681: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1682: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1683: */
1.160 brouard 1684: char *s, *t;
1.145 brouard 1685: t=in;s=in;
1686: while ((*in != occ) && (*in != '\0')){
1687: *alocc++ = *in++;
1688: }
1689: if( *in == occ){
1690: *(alocc)='\0';
1691: s=++in;
1692: }
1693:
1694: if (s == t) {/* occ not found */
1695: *(alocc-(in-s))='\0';
1696: in=s;
1697: }
1698: while ( *in != '\0'){
1699: *blocc++ = *in++;
1700: }
1701:
1702: *blocc='\0';
1703: return t;
1704: }
1.137 brouard 1705: char *cutv(char *blocc, char *alocc, char *in, char occ)
1706: {
1.187 brouard 1707: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1708: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1709: gives blocc="abcdef2ghi" and alocc="j".
1710: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1711: */
1712: char *s, *t;
1713: t=in;s=in;
1714: while (*in != '\0'){
1715: while( *in == occ){
1716: *blocc++ = *in++;
1717: s=in;
1718: }
1719: *blocc++ = *in++;
1720: }
1721: if (s == t) /* occ not found */
1722: *(blocc-(in-s))='\0';
1723: else
1724: *(blocc-(in-s)-1)='\0';
1725: in=s;
1726: while ( *in != '\0'){
1727: *alocc++ = *in++;
1728: }
1729:
1730: *alocc='\0';
1731: return s;
1732: }
1733:
1.126 brouard 1734: int nbocc(char *s, char occ)
1735: {
1736: int i,j=0;
1737: int lg=20;
1738: i=0;
1739: lg=strlen(s);
1740: for(i=0; i<= lg; i++) {
1.234 brouard 1741: if (s[i] == occ ) j++;
1.126 brouard 1742: }
1743: return j;
1744: }
1745:
1.137 brouard 1746: /* void cutv(char *u,char *v, char*t, char occ) */
1747: /* { */
1748: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1749: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1750: /* gives u="abcdef2ghi" and v="j" *\/ */
1751: /* int i,lg,j,p=0; */
1752: /* i=0; */
1753: /* lg=strlen(t); */
1754: /* for(j=0; j<=lg-1; j++) { */
1755: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1756: /* } */
1.126 brouard 1757:
1.137 brouard 1758: /* for(j=0; j<p; j++) { */
1759: /* (u[j] = t[j]); */
1760: /* } */
1761: /* u[p]='\0'; */
1.126 brouard 1762:
1.137 brouard 1763: /* for(j=0; j<= lg; j++) { */
1764: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1765: /* } */
1766: /* } */
1.126 brouard 1767:
1.160 brouard 1768: #ifdef _WIN32
1769: char * strsep(char **pp, const char *delim)
1770: {
1771: char *p, *q;
1772:
1773: if ((p = *pp) == NULL)
1774: return 0;
1775: if ((q = strpbrk (p, delim)) != NULL)
1776: {
1777: *pp = q + 1;
1778: *q = '\0';
1779: }
1780: else
1781: *pp = 0;
1782: return p;
1783: }
1784: #endif
1785:
1.126 brouard 1786: /********************** nrerror ********************/
1787:
1788: void nrerror(char error_text[])
1789: {
1790: fprintf(stderr,"ERREUR ...\n");
1791: fprintf(stderr,"%s\n",error_text);
1792: exit(EXIT_FAILURE);
1793: }
1794: /*********************** vector *******************/
1795: double *vector(int nl, int nh)
1796: {
1797: double *v;
1798: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1799: if (!v) nrerror("allocation failure in vector");
1800: return v-nl+NR_END;
1801: }
1802:
1803: /************************ free vector ******************/
1804: void free_vector(double*v, int nl, int nh)
1805: {
1806: free((FREE_ARG)(v+nl-NR_END));
1807: }
1808:
1809: /************************ivector *******************************/
1810: int *ivector(long nl,long nh)
1811: {
1812: int *v;
1813: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1814: if (!v) nrerror("allocation failure in ivector");
1815: return v-nl+NR_END;
1816: }
1817:
1818: /******************free ivector **************************/
1819: void free_ivector(int *v, long nl, long nh)
1820: {
1821: free((FREE_ARG)(v+nl-NR_END));
1822: }
1823:
1824: /************************lvector *******************************/
1825: long *lvector(long nl,long nh)
1826: {
1827: long *v;
1828: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1829: if (!v) nrerror("allocation failure in ivector");
1830: return v-nl+NR_END;
1831: }
1832:
1833: /******************free lvector **************************/
1834: void free_lvector(long *v, long nl, long nh)
1835: {
1836: free((FREE_ARG)(v+nl-NR_END));
1837: }
1838:
1839: /******************* imatrix *******************************/
1840: int **imatrix(long nrl, long nrh, long ncl, long nch)
1841: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1842: {
1843: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1844: int **m;
1845:
1846: /* allocate pointers to rows */
1847: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1848: if (!m) nrerror("allocation failure 1 in matrix()");
1849: m += NR_END;
1850: m -= nrl;
1851:
1852:
1853: /* allocate rows and set pointers to them */
1854: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1855: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1856: m[nrl] += NR_END;
1857: m[nrl] -= ncl;
1858:
1859: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1860:
1861: /* return pointer to array of pointers to rows */
1862: return m;
1863: }
1864:
1865: /****************** free_imatrix *************************/
1866: void free_imatrix(m,nrl,nrh,ncl,nch)
1867: int **m;
1868: long nch,ncl,nrh,nrl;
1869: /* free an int matrix allocated by imatrix() */
1870: {
1871: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1872: free((FREE_ARG) (m+nrl-NR_END));
1873: }
1874:
1875: /******************* matrix *******************************/
1876: double **matrix(long nrl, long nrh, long ncl, long nch)
1877: {
1878: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1879: double **m;
1880:
1881: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1882: if (!m) nrerror("allocation failure 1 in matrix()");
1883: m += NR_END;
1884: m -= nrl;
1885:
1886: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1887: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1888: m[nrl] += NR_END;
1889: m[nrl] -= ncl;
1890:
1891: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1892: return m;
1.145 brouard 1893: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1894: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1895: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1896: */
1897: }
1898:
1899: /*************************free matrix ************************/
1900: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1901: {
1902: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1903: free((FREE_ARG)(m+nrl-NR_END));
1904: }
1905:
1906: /******************* ma3x *******************************/
1907: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1908: {
1909: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1910: double ***m;
1911:
1912: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1913: if (!m) nrerror("allocation failure 1 in matrix()");
1914: m += NR_END;
1915: m -= nrl;
1916:
1917: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1918: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1919: m[nrl] += NR_END;
1920: m[nrl] -= ncl;
1921:
1922: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1923:
1924: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1925: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1926: m[nrl][ncl] += NR_END;
1927: m[nrl][ncl] -= nll;
1928: for (j=ncl+1; j<=nch; j++)
1929: m[nrl][j]=m[nrl][j-1]+nlay;
1930:
1931: for (i=nrl+1; i<=nrh; i++) {
1932: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1933: for (j=ncl+1; j<=nch; j++)
1934: m[i][j]=m[i][j-1]+nlay;
1935: }
1936: return m;
1937: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1938: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1939: */
1940: }
1941:
1942: /*************************free ma3x ************************/
1943: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1944: {
1945: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1946: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1947: free((FREE_ARG)(m+nrl-NR_END));
1948: }
1949:
1950: /*************** function subdirf ***********/
1951: char *subdirf(char fileres[])
1952: {
1953: /* Caution optionfilefiname is hidden */
1954: strcpy(tmpout,optionfilefiname);
1955: strcat(tmpout,"/"); /* Add to the right */
1956: strcat(tmpout,fileres);
1957: return tmpout;
1958: }
1959:
1960: /*************** function subdirf2 ***********/
1961: char *subdirf2(char fileres[], char *preop)
1962: {
1.314 brouard 1963: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
1964: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 1965: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 1966: /* Caution optionfilefiname is hidden */
1967: strcpy(tmpout,optionfilefiname);
1968: strcat(tmpout,"/");
1969: strcat(tmpout,preop);
1970: strcat(tmpout,fileres);
1971: return tmpout;
1972: }
1973:
1974: /*************** function subdirf3 ***********/
1975: char *subdirf3(char fileres[], char *preop, char *preop2)
1976: {
1977:
1978: /* Caution optionfilefiname is hidden */
1979: strcpy(tmpout,optionfilefiname);
1980: strcat(tmpout,"/");
1981: strcat(tmpout,preop);
1982: strcat(tmpout,preop2);
1983: strcat(tmpout,fileres);
1984: return tmpout;
1985: }
1.213 brouard 1986:
1987: /*************** function subdirfext ***********/
1988: char *subdirfext(char fileres[], char *preop, char *postop)
1989: {
1990:
1991: strcpy(tmpout,preop);
1992: strcat(tmpout,fileres);
1993: strcat(tmpout,postop);
1994: return tmpout;
1995: }
1.126 brouard 1996:
1.213 brouard 1997: /*************** function subdirfext3 ***********/
1998: char *subdirfext3(char fileres[], char *preop, char *postop)
1999: {
2000:
2001: /* Caution optionfilefiname is hidden */
2002: strcpy(tmpout,optionfilefiname);
2003: strcat(tmpout,"/");
2004: strcat(tmpout,preop);
2005: strcat(tmpout,fileres);
2006: strcat(tmpout,postop);
2007: return tmpout;
2008: }
2009:
1.162 brouard 2010: char *asc_diff_time(long time_sec, char ascdiff[])
2011: {
2012: long sec_left, days, hours, minutes;
2013: days = (time_sec) / (60*60*24);
2014: sec_left = (time_sec) % (60*60*24);
2015: hours = (sec_left) / (60*60) ;
2016: sec_left = (sec_left) %(60*60);
2017: minutes = (sec_left) /60;
2018: sec_left = (sec_left) % (60);
2019: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2020: return ascdiff;
2021: }
2022:
1.126 brouard 2023: /***************** f1dim *************************/
2024: extern int ncom;
2025: extern double *pcom,*xicom;
2026: extern double (*nrfunc)(double []);
2027:
2028: double f1dim(double x)
2029: {
2030: int j;
2031: double f;
2032: double *xt;
2033:
2034: xt=vector(1,ncom);
2035: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2036: f=(*nrfunc)(xt);
2037: free_vector(xt,1,ncom);
2038: return f;
2039: }
2040:
2041: /*****************brent *************************/
2042: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2043: {
2044: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2045: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2046: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2047: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2048: * returned function value.
2049: */
1.126 brouard 2050: int iter;
2051: double a,b,d,etemp;
1.159 brouard 2052: double fu=0,fv,fw,fx;
1.164 brouard 2053: double ftemp=0.;
1.126 brouard 2054: double p,q,r,tol1,tol2,u,v,w,x,xm;
2055: double e=0.0;
2056:
2057: a=(ax < cx ? ax : cx);
2058: b=(ax > cx ? ax : cx);
2059: x=w=v=bx;
2060: fw=fv=fx=(*f)(x);
2061: for (iter=1;iter<=ITMAX;iter++) {
2062: xm=0.5*(a+b);
2063: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2064: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2065: printf(".");fflush(stdout);
2066: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2067: #ifdef DEBUGBRENT
1.126 brouard 2068: 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);
2069: 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);
2070: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2071: #endif
2072: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2073: *xmin=x;
2074: return fx;
2075: }
2076: ftemp=fu;
2077: if (fabs(e) > tol1) {
2078: r=(x-w)*(fx-fv);
2079: q=(x-v)*(fx-fw);
2080: p=(x-v)*q-(x-w)*r;
2081: q=2.0*(q-r);
2082: if (q > 0.0) p = -p;
2083: q=fabs(q);
2084: etemp=e;
2085: e=d;
2086: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2087: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2088: else {
1.224 brouard 2089: d=p/q;
2090: u=x+d;
2091: if (u-a < tol2 || b-u < tol2)
2092: d=SIGN(tol1,xm-x);
1.126 brouard 2093: }
2094: } else {
2095: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2096: }
2097: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2098: fu=(*f)(u);
2099: if (fu <= fx) {
2100: if (u >= x) a=x; else b=x;
2101: SHFT(v,w,x,u)
1.183 brouard 2102: SHFT(fv,fw,fx,fu)
2103: } else {
2104: if (u < x) a=u; else b=u;
2105: if (fu <= fw || w == x) {
1.224 brouard 2106: v=w;
2107: w=u;
2108: fv=fw;
2109: fw=fu;
1.183 brouard 2110: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2111: v=u;
2112: fv=fu;
1.183 brouard 2113: }
2114: }
1.126 brouard 2115: }
2116: nrerror("Too many iterations in brent");
2117: *xmin=x;
2118: return fx;
2119: }
2120:
2121: /****************** mnbrak ***********************/
2122:
2123: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2124: double (*func)(double))
1.183 brouard 2125: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2126: the downhill direction (defined by the function as evaluated at the initial points) and returns
2127: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2128: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2129: */
1.126 brouard 2130: double ulim,u,r,q, dum;
2131: double fu;
1.187 brouard 2132:
2133: double scale=10.;
2134: int iterscale=0;
2135:
2136: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2137: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2138:
2139:
2140: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2141: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2142: /* *bx = *ax - (*ax - *bx)/scale; */
2143: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2144: /* } */
2145:
1.126 brouard 2146: if (*fb > *fa) {
2147: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2148: SHFT(dum,*fb,*fa,dum)
2149: }
1.126 brouard 2150: *cx=(*bx)+GOLD*(*bx-*ax);
2151: *fc=(*func)(*cx);
1.183 brouard 2152: #ifdef DEBUG
1.224 brouard 2153: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2154: 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 2155: #endif
1.224 brouard 2156: 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 2157: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2158: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2159: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2160: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2161: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2162: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2163: fu=(*func)(u);
1.163 brouard 2164: #ifdef DEBUG
2165: /* f(x)=A(x-u)**2+f(u) */
2166: double A, fparabu;
2167: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2168: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2169: 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);
2170: 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 2171: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2172: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2173: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2174: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2175: #endif
1.184 brouard 2176: #ifdef MNBRAKORIGINAL
1.183 brouard 2177: #else
1.191 brouard 2178: /* if (fu > *fc) { */
2179: /* #ifdef DEBUG */
2180: /* printf("mnbrak4 fu > fc \n"); */
2181: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2182: /* #endif */
2183: /* /\* 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 *\\/ *\/ */
2184: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2185: /* dum=u; /\* Shifting c and u *\/ */
2186: /* u = *cx; */
2187: /* *cx = dum; */
2188: /* dum = fu; */
2189: /* fu = *fc; */
2190: /* *fc =dum; */
2191: /* } else { /\* end *\/ */
2192: /* #ifdef DEBUG */
2193: /* printf("mnbrak3 fu < fc \n"); */
2194: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2195: /* #endif */
2196: /* dum=u; /\* Shifting c and u *\/ */
2197: /* u = *cx; */
2198: /* *cx = dum; */
2199: /* dum = fu; */
2200: /* fu = *fc; */
2201: /* *fc =dum; */
2202: /* } */
1.224 brouard 2203: #ifdef DEBUGMNBRAK
2204: double A, fparabu;
2205: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2206: fparabu= *fa - A*(*ax-u)*(*ax-u);
2207: 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);
2208: 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 2209: #endif
1.191 brouard 2210: dum=u; /* Shifting c and u */
2211: u = *cx;
2212: *cx = dum;
2213: dum = fu;
2214: fu = *fc;
2215: *fc =dum;
1.183 brouard 2216: #endif
1.162 brouard 2217: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2218: #ifdef DEBUG
1.224 brouard 2219: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2220: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2221: #endif
1.126 brouard 2222: fu=(*func)(u);
2223: if (fu < *fc) {
1.183 brouard 2224: #ifdef DEBUG
1.224 brouard 2225: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2226: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2227: #endif
2228: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2229: SHFT(*fb,*fc,fu,(*func)(u))
2230: #ifdef DEBUG
2231: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2232: #endif
2233: }
1.162 brouard 2234: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2235: #ifdef DEBUG
1.224 brouard 2236: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2237: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2238: #endif
1.126 brouard 2239: u=ulim;
2240: fu=(*func)(u);
1.183 brouard 2241: } else { /* u could be left to b (if r > q parabola has a maximum) */
2242: #ifdef DEBUG
1.224 brouard 2243: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2244: 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 2245: #endif
1.126 brouard 2246: u=(*cx)+GOLD*(*cx-*bx);
2247: fu=(*func)(u);
1.224 brouard 2248: #ifdef DEBUG
2249: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2250: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2251: #endif
1.183 brouard 2252: } /* end tests */
1.126 brouard 2253: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2254: SHFT(*fa,*fb,*fc,fu)
2255: #ifdef DEBUG
1.224 brouard 2256: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2257: 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 2258: #endif
2259: } /* 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 2260: }
2261:
2262: /*************** linmin ************************/
1.162 brouard 2263: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2264: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2265: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2266: the value of func at the returned location p . This is actually all accomplished by calling the
2267: routines mnbrak and brent .*/
1.126 brouard 2268: int ncom;
2269: double *pcom,*xicom;
2270: double (*nrfunc)(double []);
2271:
1.224 brouard 2272: #ifdef LINMINORIGINAL
1.126 brouard 2273: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2274: #else
2275: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2276: #endif
1.126 brouard 2277: {
2278: double brent(double ax, double bx, double cx,
2279: double (*f)(double), double tol, double *xmin);
2280: double f1dim(double x);
2281: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2282: double *fc, double (*func)(double));
2283: int j;
2284: double xx,xmin,bx,ax;
2285: double fx,fb,fa;
1.187 brouard 2286:
1.203 brouard 2287: #ifdef LINMINORIGINAL
2288: #else
2289: double scale=10., axs, xxs; /* Scale added for infinity */
2290: #endif
2291:
1.126 brouard 2292: ncom=n;
2293: pcom=vector(1,n);
2294: xicom=vector(1,n);
2295: nrfunc=func;
2296: for (j=1;j<=n;j++) {
2297: pcom[j]=p[j];
1.202 brouard 2298: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2299: }
1.187 brouard 2300:
1.203 brouard 2301: #ifdef LINMINORIGINAL
2302: xx=1.;
2303: #else
2304: axs=0.0;
2305: xxs=1.;
2306: do{
2307: xx= xxs;
2308: #endif
1.187 brouard 2309: ax=0.;
2310: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2311: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2312: /* 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)) */
2313: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2314: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2315: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2316: /* 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 2317: #ifdef LINMINORIGINAL
2318: #else
2319: if (fx != fx){
1.224 brouard 2320: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2321: printf("|");
2322: fprintf(ficlog,"|");
1.203 brouard 2323: #ifdef DEBUGLINMIN
1.224 brouard 2324: 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 2325: #endif
2326: }
1.224 brouard 2327: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2328: #endif
2329:
1.191 brouard 2330: #ifdef DEBUGLINMIN
2331: 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 2332: 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 2333: #endif
1.224 brouard 2334: #ifdef LINMINORIGINAL
2335: #else
1.317 brouard 2336: if(fb == fx){ /* Flat function in the direction */
2337: xmin=xx;
1.224 brouard 2338: *flat=1;
1.317 brouard 2339: }else{
1.224 brouard 2340: *flat=0;
2341: #endif
2342: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2343: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2344: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2345: /* fmin = f(p[j] + xmin * xi[j]) */
2346: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2347: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2348: #ifdef DEBUG
1.224 brouard 2349: 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);
2350: 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);
2351: #endif
2352: #ifdef LINMINORIGINAL
2353: #else
2354: }
1.126 brouard 2355: #endif
1.191 brouard 2356: #ifdef DEBUGLINMIN
2357: printf("linmin end ");
1.202 brouard 2358: fprintf(ficlog,"linmin end ");
1.191 brouard 2359: #endif
1.126 brouard 2360: for (j=1;j<=n;j++) {
1.203 brouard 2361: #ifdef LINMINORIGINAL
2362: xi[j] *= xmin;
2363: #else
2364: #ifdef DEBUGLINMIN
2365: if(xxs <1.0)
2366: printf(" before xi[%d]=%12.8f", j,xi[j]);
2367: #endif
2368: 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) */
2369: #ifdef DEBUGLINMIN
2370: if(xxs <1.0)
2371: 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 );
2372: #endif
2373: #endif
1.187 brouard 2374: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2375: }
1.191 brouard 2376: #ifdef DEBUGLINMIN
1.203 brouard 2377: printf("\n");
1.191 brouard 2378: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2379: 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 2380: for (j=1;j<=n;j++) {
1.202 brouard 2381: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2382: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2383: if(j % ncovmodel == 0){
1.191 brouard 2384: printf("\n");
1.202 brouard 2385: fprintf(ficlog,"\n");
2386: }
1.191 brouard 2387: }
1.203 brouard 2388: #else
1.191 brouard 2389: #endif
1.126 brouard 2390: free_vector(xicom,1,n);
2391: free_vector(pcom,1,n);
2392: }
2393:
2394:
2395: /*************** powell ************************/
1.162 brouard 2396: /*
1.317 brouard 2397: Minimization of a function func of n variables. Input consists in an initial starting point
2398: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2399: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2400: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2401: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2402: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2403: */
1.224 brouard 2404: #ifdef LINMINORIGINAL
2405: #else
2406: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2407: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2408: #endif
1.126 brouard 2409: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2410: double (*func)(double []))
2411: {
1.224 brouard 2412: #ifdef LINMINORIGINAL
2413: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2414: double (*func)(double []));
1.224 brouard 2415: #else
1.241 brouard 2416: void linmin(double p[], double xi[], int n, double *fret,
2417: double (*func)(double []),int *flat);
1.224 brouard 2418: #endif
1.239 brouard 2419: int i,ibig,j,jk,k;
1.126 brouard 2420: double del,t,*pt,*ptt,*xit;
1.181 brouard 2421: double directest;
1.126 brouard 2422: double fp,fptt;
2423: double *xits;
2424: int niterf, itmp;
2425:
2426: pt=vector(1,n);
2427: ptt=vector(1,n);
2428: xit=vector(1,n);
2429: xits=vector(1,n);
2430: *fret=(*func)(p);
2431: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2432: rcurr_time = time(NULL);
1.126 brouard 2433: for (*iter=1;;++(*iter)) {
2434: ibig=0;
2435: del=0.0;
1.157 brouard 2436: rlast_time=rcurr_time;
2437: /* (void) gettimeofday(&curr_time,&tzp); */
2438: rcurr_time = time(NULL);
2439: curr_time = *localtime(&rcurr_time);
1.324 brouard 2440: 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);
2441: 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 2442: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2443: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2444: for (i=1;i<=n;i++) {
1.126 brouard 2445: fprintf(ficrespow," %.12lf", p[i]);
2446: }
1.239 brouard 2447: fprintf(ficrespow,"\n");fflush(ficrespow);
2448: printf("\n#model= 1 + age ");
2449: fprintf(ficlog,"\n#model= 1 + age ");
2450: if(nagesqr==1){
1.241 brouard 2451: printf(" + age*age ");
2452: fprintf(ficlog," + age*age ");
1.239 brouard 2453: }
2454: for(j=1;j <=ncovmodel-2;j++){
2455: if(Typevar[j]==0) {
2456: printf(" + V%d ",Tvar[j]);
2457: fprintf(ficlog," + V%d ",Tvar[j]);
2458: }else if(Typevar[j]==1) {
2459: printf(" + V%d*age ",Tvar[j]);
2460: fprintf(ficlog," + V%d*age ",Tvar[j]);
2461: }else if(Typevar[j]==2) {
2462: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2463: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2464: }
2465: }
1.126 brouard 2466: printf("\n");
1.239 brouard 2467: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2468: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2469: fprintf(ficlog,"\n");
1.239 brouard 2470: for(i=1,jk=1; i <=nlstate; i++){
2471: for(k=1; k <=(nlstate+ndeath); k++){
2472: if (k != i) {
2473: printf("%d%d ",i,k);
2474: fprintf(ficlog,"%d%d ",i,k);
2475: for(j=1; j <=ncovmodel; j++){
2476: printf("%12.7f ",p[jk]);
2477: fprintf(ficlog,"%12.7f ",p[jk]);
2478: jk++;
2479: }
2480: printf("\n");
2481: fprintf(ficlog,"\n");
2482: }
2483: }
2484: }
1.241 brouard 2485: if(*iter <=3 && *iter >1){
1.157 brouard 2486: tml = *localtime(&rcurr_time);
2487: strcpy(strcurr,asctime(&tml));
2488: rforecast_time=rcurr_time;
1.126 brouard 2489: itmp = strlen(strcurr);
2490: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2491: strcurr[itmp-1]='\0';
1.162 brouard 2492: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2493: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2494: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2495: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2496: forecast_time = *localtime(&rforecast_time);
2497: strcpy(strfor,asctime(&forecast_time));
2498: itmp = strlen(strfor);
2499: if(strfor[itmp-1]=='\n')
2500: strfor[itmp-1]='\0';
2501: 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);
2502: 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 2503: }
2504: }
1.187 brouard 2505: for (i=1;i<=n;i++) { /* For each direction i */
2506: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2507: fptt=(*fret);
2508: #ifdef DEBUG
1.203 brouard 2509: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2510: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2511: #endif
1.203 brouard 2512: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2513: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2514: #ifdef LINMINORIGINAL
1.188 brouard 2515: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2516: #else
2517: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2518: flatdir[i]=flat; /* Function is vanishing in that direction i */
2519: #endif
2520: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2521: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2522: /* because that direction will be replaced unless the gain del is small */
2523: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2524: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2525: /* with the new direction. */
2526: del=fabs(fptt-(*fret));
2527: ibig=i;
1.126 brouard 2528: }
2529: #ifdef DEBUG
2530: printf("%d %.12e",i,(*fret));
2531: fprintf(ficlog,"%d %.12e",i,(*fret));
2532: for (j=1;j<=n;j++) {
1.224 brouard 2533: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2534: printf(" x(%d)=%.12e",j,xit[j]);
2535: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2536: }
2537: for(j=1;j<=n;j++) {
1.225 brouard 2538: printf(" p(%d)=%.12e",j,p[j]);
2539: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2540: }
2541: printf("\n");
2542: fprintf(ficlog,"\n");
2543: #endif
1.187 brouard 2544: } /* end loop on each direction i */
2545: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2546: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2547: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2548: for(j=1;j<=n;j++) {
2549: if(flatdir[j] >0){
2550: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2551: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2552: }
1.319 brouard 2553: /* printf("\n"); */
2554: /* fprintf(ficlog,"\n"); */
2555: }
1.243 brouard 2556: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2557: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2558: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2559: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2560: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2561: /* decreased of more than 3.84 */
2562: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2563: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2564: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2565:
1.188 brouard 2566: /* Starting the program with initial values given by a former maximization will simply change */
2567: /* the scales of the directions and the directions, because the are reset to canonical directions */
2568: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2569: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2570: #ifdef DEBUG
2571: int k[2],l;
2572: k[0]=1;
2573: k[1]=-1;
2574: printf("Max: %.12e",(*func)(p));
2575: fprintf(ficlog,"Max: %.12e",(*func)(p));
2576: for (j=1;j<=n;j++) {
2577: printf(" %.12e",p[j]);
2578: fprintf(ficlog," %.12e",p[j]);
2579: }
2580: printf("\n");
2581: fprintf(ficlog,"\n");
2582: for(l=0;l<=1;l++) {
2583: for (j=1;j<=n;j++) {
2584: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2585: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2586: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2587: }
2588: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2589: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2590: }
2591: #endif
2592:
2593: free_vector(xit,1,n);
2594: free_vector(xits,1,n);
2595: free_vector(ptt,1,n);
2596: free_vector(pt,1,n);
2597: return;
1.192 brouard 2598: } /* enough precision */
1.240 brouard 2599: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2600: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2601: ptt[j]=2.0*p[j]-pt[j];
2602: xit[j]=p[j]-pt[j];
2603: pt[j]=p[j];
2604: }
1.181 brouard 2605: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2606: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2607: if (*iter <=4) {
1.225 brouard 2608: #else
2609: #endif
1.224 brouard 2610: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2611: #else
1.161 brouard 2612: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2613: #endif
1.162 brouard 2614: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2615: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2616: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2617: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2618: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2619: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2620: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2621: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2622: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2623: /* Even if f3 <f1, directest can be negative and t >0 */
2624: /* mu² and del² are equal when f3=f1 */
2625: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2626: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2627: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2628: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2629: #ifdef NRCORIGINAL
2630: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2631: #else
2632: 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 2633: t= t- del*SQR(fp-fptt);
1.183 brouard 2634: #endif
1.202 brouard 2635: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2636: #ifdef DEBUG
1.181 brouard 2637: 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);
2638: 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 2639: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2640: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2641: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2642: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2643: 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);
2644: 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);
2645: #endif
1.183 brouard 2646: #ifdef POWELLORIGINAL
2647: if (t < 0.0) { /* Then we use it for new direction */
2648: #else
1.182 brouard 2649: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2650: 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 2651: 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 2652: 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 2653: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2654: }
1.181 brouard 2655: if (directest < 0.0) { /* Then we use it for new direction */
2656: #endif
1.191 brouard 2657: #ifdef DEBUGLINMIN
1.234 brouard 2658: printf("Before linmin in direction P%d-P0\n",n);
2659: for (j=1;j<=n;j++) {
2660: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2661: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2662: if(j % ncovmodel == 0){
2663: printf("\n");
2664: fprintf(ficlog,"\n");
2665: }
2666: }
1.224 brouard 2667: #endif
2668: #ifdef LINMINORIGINAL
1.234 brouard 2669: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2670: #else
1.234 brouard 2671: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2672: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2673: #endif
1.234 brouard 2674:
1.191 brouard 2675: #ifdef DEBUGLINMIN
1.234 brouard 2676: for (j=1;j<=n;j++) {
2677: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2678: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2679: if(j % ncovmodel == 0){
2680: printf("\n");
2681: fprintf(ficlog,"\n");
2682: }
2683: }
1.224 brouard 2684: #endif
1.234 brouard 2685: for (j=1;j<=n;j++) {
2686: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2687: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2688: }
1.224 brouard 2689: #ifdef LINMINORIGINAL
2690: #else
1.234 brouard 2691: for (j=1, flatd=0;j<=n;j++) {
2692: if(flatdir[j]>0)
2693: flatd++;
2694: }
2695: if(flatd >0){
1.255 brouard 2696: printf("%d flat directions: ",flatd);
2697: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2698: for (j=1;j<=n;j++) {
2699: if(flatdir[j]>0){
2700: printf("%d ",j);
2701: fprintf(ficlog,"%d ",j);
2702: }
2703: }
2704: printf("\n");
2705: fprintf(ficlog,"\n");
1.319 brouard 2706: #ifdef FLATSUP
2707: free_vector(xit,1,n);
2708: free_vector(xits,1,n);
2709: free_vector(ptt,1,n);
2710: free_vector(pt,1,n);
2711: return;
2712: #endif
1.234 brouard 2713: }
1.191 brouard 2714: #endif
1.234 brouard 2715: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2716: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2717:
1.126 brouard 2718: #ifdef DEBUG
1.234 brouard 2719: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2720: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2721: for(j=1;j<=n;j++){
2722: printf(" %lf",xit[j]);
2723: fprintf(ficlog," %lf",xit[j]);
2724: }
2725: printf("\n");
2726: fprintf(ficlog,"\n");
1.126 brouard 2727: #endif
1.192 brouard 2728: } /* end of t or directest negative */
1.224 brouard 2729: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2730: #else
1.234 brouard 2731: } /* end if (fptt < fp) */
1.192 brouard 2732: #endif
1.225 brouard 2733: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2734: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2735: #else
1.224 brouard 2736: #endif
1.234 brouard 2737: } /* loop iteration */
1.126 brouard 2738: }
1.234 brouard 2739:
1.126 brouard 2740: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2741:
1.235 brouard 2742: 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 2743: {
1.279 brouard 2744: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2745: * (and selected quantitative values in nres)
2746: * by left multiplying the unit
2747: * matrix by transitions matrix until convergence is reached with precision ftolpl
2748: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2749: * Wx is row vector: population in state 1, population in state 2, population dead
2750: * or prevalence in state 1, prevalence in state 2, 0
2751: * newm is the matrix after multiplications, its rows are identical at a factor.
2752: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2753: * Output is prlim.
2754: * Initial matrix pimij
2755: */
1.206 brouard 2756: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2757: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2758: /* 0, 0 , 1} */
2759: /*
2760: * and after some iteration: */
2761: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2762: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2763: /* 0, 0 , 1} */
2764: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2765: /* {0.51571254859325999, 0.4842874514067399, */
2766: /* 0.51326036147820708, 0.48673963852179264} */
2767: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2768:
1.126 brouard 2769: int i, ii,j,k;
1.209 brouard 2770: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2771: /* double **matprod2(); */ /* test */
1.218 brouard 2772: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2773: double **newm;
1.209 brouard 2774: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2775: int ncvloop=0;
1.288 brouard 2776: int first=0;
1.169 brouard 2777:
1.209 brouard 2778: min=vector(1,nlstate);
2779: max=vector(1,nlstate);
2780: meandiff=vector(1,nlstate);
2781:
1.218 brouard 2782: /* Starting with matrix unity */
1.126 brouard 2783: for (ii=1;ii<=nlstate+ndeath;ii++)
2784: for (j=1;j<=nlstate+ndeath;j++){
2785: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2786: }
1.169 brouard 2787:
2788: cov[1]=1.;
2789:
2790: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2791: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2792: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2793: ncvloop++;
1.126 brouard 2794: newm=savm;
2795: /* Covariates have to be included here again */
1.138 brouard 2796: cov[2]=agefin;
1.319 brouard 2797: if(nagesqr==1){
2798: cov[3]= agefin*agefin;
2799: }
1.234 brouard 2800: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2801: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2802: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.319 brouard 2803: /* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; */
1.235 brouard 2804: /* 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 2805: }
2806: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2807: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.319 brouard 2808: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2809: /* cov[++k1]=Tqresult[nres][k]; */
1.235 brouard 2810: /* 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 2811: }
1.237 brouard 2812: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 2813: if(Dummy[Tage[k]]==2){ /* dummy with age */
1.234 brouard 2814: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 2815: /* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
2816: } else if(Dummy[Tage[k]]==3){ /* quantitative with age */
2817: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2818: /* cov[++k1]=Tqresult[nres][k]; */
1.234 brouard 2819: }
1.235 brouard 2820: /* 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 2821: }
1.237 brouard 2822: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2823: /* printf("prevalim Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2]); */
1.237 brouard 2824: if(Dummy[Tvard[k][1]==0]){
2825: if(Dummy[Tvard[k][2]==0]){
2826: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
1.319 brouard 2827: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.237 brouard 2828: }else{
2829: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
1.319 brouard 2830: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; */
1.237 brouard 2831: }
2832: }else{
2833: if(Dummy[Tvard[k][2]==0]){
2834: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
1.319 brouard 2835: /* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; */
1.237 brouard 2836: }else{
2837: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
1.319 brouard 2838: /* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
1.237 brouard 2839: }
2840: }
1.234 brouard 2841: }
1.138 brouard 2842: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2843: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2844: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2845: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2846: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 2847: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2848: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2849:
1.126 brouard 2850: savm=oldm;
2851: oldm=newm;
1.209 brouard 2852:
2853: for(j=1; j<=nlstate; j++){
2854: max[j]=0.;
2855: min[j]=1.;
2856: }
2857: for(i=1;i<=nlstate;i++){
2858: sumnew=0;
2859: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2860: for(j=1; j<=nlstate; j++){
2861: prlim[i][j]= newm[i][j]/(1-sumnew);
2862: max[j]=FMAX(max[j],prlim[i][j]);
2863: min[j]=FMIN(min[j],prlim[i][j]);
2864: }
2865: }
2866:
1.126 brouard 2867: maxmax=0.;
1.209 brouard 2868: for(j=1; j<=nlstate; j++){
2869: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2870: maxmax=FMAX(maxmax,meandiff[j]);
2871: /* 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 2872: } /* j loop */
1.203 brouard 2873: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2874: /* 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 2875: if(maxmax < ftolpl){
1.209 brouard 2876: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2877: free_vector(min,1,nlstate);
2878: free_vector(max,1,nlstate);
2879: free_vector(meandiff,1,nlstate);
1.126 brouard 2880: return prlim;
2881: }
1.288 brouard 2882: } /* agefin loop */
1.208 brouard 2883: /* After some age loop it doesn't converge */
1.288 brouard 2884: if(!first){
2885: first=1;
2886: 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 2887: 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);
2888: }else if (first >=1 && first <10){
2889: 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);
2890: first++;
2891: }else if (first ==10){
2892: 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);
2893: 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");
2894: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
2895: first++;
1.288 brouard 2896: }
2897:
1.209 brouard 2898: /* 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); */
2899: free_vector(min,1,nlstate);
2900: free_vector(max,1,nlstate);
2901: free_vector(meandiff,1,nlstate);
1.208 brouard 2902:
1.169 brouard 2903: return prlim; /* should not reach here */
1.126 brouard 2904: }
2905:
1.217 brouard 2906:
2907: /**** Back Prevalence limit (stable or period prevalence) ****************/
2908:
1.218 brouard 2909: /* 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) */
2910: /* 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 2911: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2912: {
1.264 brouard 2913: /* 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 2914: matrix by transitions matrix until convergence is reached with precision ftolpl */
2915: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2916: /* Wx is row vector: population in state 1, population in state 2, population dead */
2917: /* or prevalence in state 1, prevalence in state 2, 0 */
2918: /* newm is the matrix after multiplications, its rows are identical at a factor */
2919: /* Initial matrix pimij */
2920: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2921: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2922: /* 0, 0 , 1} */
2923: /*
2924: * and after some iteration: */
2925: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2926: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2927: /* 0, 0 , 1} */
2928: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2929: /* {0.51571254859325999, 0.4842874514067399, */
2930: /* 0.51326036147820708, 0.48673963852179264} */
2931: /* If we start from prlim again, prlim tends to a constant matrix */
2932:
2933: int i, ii,j,k;
1.247 brouard 2934: int first=0;
1.217 brouard 2935: double *min, *max, *meandiff, maxmax,sumnew=0.;
2936: /* double **matprod2(); */ /* test */
2937: double **out, cov[NCOVMAX+1], **bmij();
2938: double **newm;
1.218 brouard 2939: double **dnewm, **doldm, **dsavm; /* for use */
2940: double **oldm, **savm; /* for use */
2941:
1.217 brouard 2942: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2943: int ncvloop=0;
2944:
2945: min=vector(1,nlstate);
2946: max=vector(1,nlstate);
2947: meandiff=vector(1,nlstate);
2948:
1.266 brouard 2949: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2950: oldm=oldms; savm=savms;
2951:
2952: /* Starting with matrix unity */
2953: for (ii=1;ii<=nlstate+ndeath;ii++)
2954: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2955: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2956: }
2957:
2958: cov[1]=1.;
2959:
2960: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2961: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2962: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2963: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2964: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2965: ncvloop++;
1.218 brouard 2966: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2967: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2968: /* Covariates have to be included here again */
2969: cov[2]=agefin;
1.319 brouard 2970: if(nagesqr==1){
1.217 brouard 2971: cov[3]= agefin*agefin;;
1.319 brouard 2972: }
1.242 brouard 2973: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2974: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2975: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2976: /* 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 2977: }
2978: /* for (k=1; k<=cptcovn;k++) { */
2979: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2980: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2981: /* /\* 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])]); *\/ */
2982: /* } */
2983: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2984: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2985: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2986: /* printf("prevalim Quantitative k=%d TvarsQind[%d]=%d, TvarsQ[%d]=V%d,Tqresult[%d][%d]=%f\n",k,k,TvarsQind[k],k,TvarsQ[k],nres,k,Tqresult[nres][k]); */
2987: }
2988: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2989: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2990: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2991: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2992: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.319 brouard 2993: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age *\/ ERROR ???*/
2994: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.242 brouard 2995: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 2996: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
2997: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.242 brouard 2998: }
2999: /* 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]); */
3000: }
3001: for (k=1; k<=cptcovprod;k++){ /* For product without age */
3002: /* 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]); */
3003: if(Dummy[Tvard[k][1]==0]){
3004: if(Dummy[Tvard[k][2]==0]){
3005: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3006: }else{
3007: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
3008: }
3009: }else{
3010: if(Dummy[Tvard[k][2]==0]){
3011: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
3012: }else{
3013: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3014: }
3015: }
1.217 brouard 3016: }
3017:
3018: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3019: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3020: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3021: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3022: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3023: /* ij should be linked to the correct index of cov */
3024: /* age and covariate values ij are in 'cov', but we need to pass
3025: * ij for the observed prevalence at age and status and covariate
3026: * number: prevacurrent[(int)agefin][ii][ij]
3027: */
3028: /* 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 *\/ */
3029: /* 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 *\/ */
3030: 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 3031: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3032: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3033: /* for(i=1; i<=nlstate+ndeath; i++) { */
3034: /* printf("%d newm= ",i); */
3035: /* for(j=1;j<=nlstate+ndeath;j++) { */
3036: /* printf("%f ",newm[i][j]); */
3037: /* } */
3038: /* printf("oldm * "); */
3039: /* for(j=1;j<=nlstate+ndeath;j++) { */
3040: /* printf("%f ",oldm[i][j]); */
3041: /* } */
1.268 brouard 3042: /* printf(" bmmij "); */
1.266 brouard 3043: /* for(j=1;j<=nlstate+ndeath;j++) { */
3044: /* printf("%f ",pmmij[i][j]); */
3045: /* } */
3046: /* printf("\n"); */
3047: /* } */
3048: /* } */
1.217 brouard 3049: savm=oldm;
3050: oldm=newm;
1.266 brouard 3051:
1.217 brouard 3052: for(j=1; j<=nlstate; j++){
3053: max[j]=0.;
3054: min[j]=1.;
3055: }
3056: for(j=1; j<=nlstate; j++){
3057: for(i=1;i<=nlstate;i++){
1.234 brouard 3058: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3059: bprlim[i][j]= newm[i][j];
3060: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3061: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3062: }
3063: }
1.218 brouard 3064:
1.217 brouard 3065: maxmax=0.;
3066: for(i=1; i<=nlstate; i++){
1.318 brouard 3067: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3068: maxmax=FMAX(maxmax,meandiff[i]);
3069: /* 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 3070: } /* i loop */
1.217 brouard 3071: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3072: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3073: if(maxmax < ftolpl){
1.220 brouard 3074: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3075: free_vector(min,1,nlstate);
3076: free_vector(max,1,nlstate);
3077: free_vector(meandiff,1,nlstate);
3078: return bprlim;
3079: }
1.288 brouard 3080: } /* agefin loop */
1.217 brouard 3081: /* After some age loop it doesn't converge */
1.288 brouard 3082: if(!first){
1.247 brouard 3083: first=1;
3084: 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\
3085: 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);
3086: }
3087: 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 3088: 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);
3089: /* 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); */
3090: free_vector(min,1,nlstate);
3091: free_vector(max,1,nlstate);
3092: free_vector(meandiff,1,nlstate);
3093:
3094: return bprlim; /* should not reach here */
3095: }
3096:
1.126 brouard 3097: /*************** transition probabilities ***************/
3098:
3099: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3100: {
1.138 brouard 3101: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3102: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3103: model to the ncovmodel covariates (including constant and age).
3104: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3105: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3106: ncth covariate in the global vector x is given by the formula:
3107: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3108: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3109: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3110: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3111: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3112: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3113: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3114: */
3115: double s1, lnpijopii;
1.126 brouard 3116: /*double t34;*/
1.164 brouard 3117: int i,j, nc, ii, jj;
1.126 brouard 3118:
1.223 brouard 3119: for(i=1; i<= nlstate; i++){
3120: for(j=1; j<i;j++){
3121: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3122: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3123: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3124: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3125: }
3126: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3127: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3128: }
3129: for(j=i+1; j<=nlstate+ndeath;j++){
3130: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3131: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3132: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3133: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3134: }
3135: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3136: }
3137: }
1.218 brouard 3138:
1.223 brouard 3139: for(i=1; i<= nlstate; i++){
3140: s1=0;
3141: for(j=1; j<i; j++){
3142: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3143: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3144: }
3145: for(j=i+1; j<=nlstate+ndeath; j++){
3146: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3147: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3148: }
3149: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3150: ps[i][i]=1./(s1+1.);
3151: /* Computing other pijs */
3152: for(j=1; j<i; j++)
1.325 brouard 3153: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3154: for(j=i+1; j<=nlstate+ndeath; j++)
3155: ps[i][j]= exp(ps[i][j])*ps[i][i];
3156: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3157: } /* end i */
1.218 brouard 3158:
1.223 brouard 3159: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3160: for(jj=1; jj<= nlstate+ndeath; jj++){
3161: ps[ii][jj]=0;
3162: ps[ii][ii]=1;
3163: }
3164: }
1.294 brouard 3165:
3166:
1.223 brouard 3167: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3168: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3169: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3170: /* } */
3171: /* printf("\n "); */
3172: /* } */
3173: /* printf("\n ");printf("%lf ",cov[2]);*/
3174: /*
3175: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3176: goto end;*/
1.266 brouard 3177: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3178: }
3179:
1.218 brouard 3180: /*************** backward transition probabilities ***************/
3181:
3182: /* 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 ) */
3183: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3184: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3185: {
1.302 brouard 3186: /* 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 3187: * 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 3188: */
1.218 brouard 3189: int i, ii, j,k;
1.222 brouard 3190:
3191: double **out, **pmij();
3192: double sumnew=0.;
1.218 brouard 3193: double agefin;
1.292 brouard 3194: 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 3195: double **dnewm, **dsavm, **doldm;
3196: double **bbmij;
3197:
1.218 brouard 3198: doldm=ddoldms; /* global pointers */
1.222 brouard 3199: dnewm=ddnewms;
3200: dsavm=ddsavms;
1.318 brouard 3201:
3202: /* Debug */
3203: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3204: agefin=cov[2];
1.268 brouard 3205: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3206: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3207: the observed prevalence (with this covariate ij) at beginning of transition */
3208: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3209:
3210: /* P_x */
1.325 brouard 3211: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3212: /* outputs pmmij which is a stochastic matrix in row */
3213:
3214: /* Diag(w_x) */
1.292 brouard 3215: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3216: sumnew=0.;
1.269 brouard 3217: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3218: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3219: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3220: sumnew+=prevacurrent[(int)agefin][ii][ij];
3221: }
3222: if(sumnew >0.01){ /* At least some value in the prevalence */
3223: for (ii=1;ii<=nlstate+ndeath;ii++){
3224: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3225: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3226: }
3227: }else{
3228: for (ii=1;ii<=nlstate+ndeath;ii++){
3229: for (j=1;j<=nlstate+ndeath;j++)
3230: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3231: }
3232: /* if(sumnew <0.9){ */
3233: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3234: /* } */
3235: }
3236: k3=0.0; /* We put the last diagonal to 0 */
3237: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3238: doldm[ii][ii]= k3;
3239: }
3240: /* End doldm, At the end doldm is diag[(w_i)] */
3241:
1.292 brouard 3242: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3243: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3244:
1.292 brouard 3245: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3246: /* 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 3247: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3248: sumnew=0.;
1.222 brouard 3249: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3250: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3251: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3252: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3253: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3254: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3255: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3256: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3257: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3258: /* }else */
1.268 brouard 3259: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3260: } /*End ii */
3261: } /* 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 */
3262:
1.292 brouard 3263: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3264: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3265: /* end bmij */
1.266 brouard 3266: return ps; /*pointer is unchanged */
1.218 brouard 3267: }
1.217 brouard 3268: /*************** transition probabilities ***************/
3269:
1.218 brouard 3270: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3271: {
3272: /* According to parameters values stored in x and the covariate's values stored in cov,
3273: computes the probability to be observed in state j being in state i by appying the
3274: model to the ncovmodel covariates (including constant and age).
3275: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3276: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3277: ncth covariate in the global vector x is given by the formula:
3278: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3279: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3280: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3281: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3282: Outputs ps[i][j] the probability to be observed in j being in j according to
3283: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3284: */
3285: double s1, lnpijopii;
3286: /*double t34;*/
3287: int i,j, nc, ii, jj;
3288:
1.234 brouard 3289: for(i=1; i<= nlstate; i++){
3290: for(j=1; j<i;j++){
3291: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3292: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3293: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3294: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3295: }
3296: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3297: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3298: }
3299: for(j=i+1; j<=nlstate+ndeath;j++){
3300: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3301: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3302: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3303: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3304: }
3305: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3306: }
3307: }
3308:
3309: for(i=1; i<= nlstate; i++){
3310: s1=0;
3311: for(j=1; j<i; j++){
3312: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3313: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3314: }
3315: for(j=i+1; j<=nlstate+ndeath; j++){
3316: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3317: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3318: }
3319: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3320: ps[i][i]=1./(s1+1.);
3321: /* Computing other pijs */
3322: for(j=1; j<i; j++)
3323: ps[i][j]= exp(ps[i][j])*ps[i][i];
3324: for(j=i+1; j<=nlstate+ndeath; j++)
3325: ps[i][j]= exp(ps[i][j])*ps[i][i];
3326: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3327: } /* end i */
3328:
3329: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3330: for(jj=1; jj<= nlstate+ndeath; jj++){
3331: ps[ii][jj]=0;
3332: ps[ii][ii]=1;
3333: }
3334: }
1.296 brouard 3335: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3336: for(jj=1; jj<= nlstate+ndeath; jj++){
3337: s1=0.;
3338: for(ii=1; ii<= nlstate+ndeath; ii++){
3339: s1+=ps[ii][jj];
3340: }
3341: for(ii=1; ii<= nlstate; ii++){
3342: ps[ii][jj]=ps[ii][jj]/s1;
3343: }
3344: }
3345: /* Transposition */
3346: for(jj=1; jj<= nlstate+ndeath; jj++){
3347: for(ii=jj; ii<= nlstate+ndeath; ii++){
3348: s1=ps[ii][jj];
3349: ps[ii][jj]=ps[jj][ii];
3350: ps[jj][ii]=s1;
3351: }
3352: }
3353: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3354: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3355: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3356: /* } */
3357: /* printf("\n "); */
3358: /* } */
3359: /* printf("\n ");printf("%lf ",cov[2]);*/
3360: /*
3361: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3362: goto end;*/
3363: return ps;
1.217 brouard 3364: }
3365:
3366:
1.126 brouard 3367: /**************** Product of 2 matrices ******************/
3368:
1.145 brouard 3369: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3370: {
3371: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3372: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3373: /* in, b, out are matrice of pointers which should have been initialized
3374: before: only the contents of out is modified. The function returns
3375: a pointer to pointers identical to out */
1.145 brouard 3376: int i, j, k;
1.126 brouard 3377: for(i=nrl; i<= nrh; i++)
1.145 brouard 3378: for(k=ncolol; k<=ncoloh; k++){
3379: out[i][k]=0.;
3380: for(j=ncl; j<=nch; j++)
3381: out[i][k] +=in[i][j]*b[j][k];
3382: }
1.126 brouard 3383: return out;
3384: }
3385:
3386:
3387: /************* Higher Matrix Product ***************/
3388:
1.235 brouard 3389: 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 3390: {
1.218 brouard 3391: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3392: 'nhstepm*hstepm*stepm' months (i.e. until
3393: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3394: nhstepm*hstepm matrices.
3395: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3396: (typically every 2 years instead of every month which is too big
3397: for the memory).
3398: Model is determined by parameters x and covariates have to be
3399: included manually here.
3400:
3401: */
3402:
3403: int i, j, d, h, k;
1.131 brouard 3404: double **out, cov[NCOVMAX+1];
1.126 brouard 3405: double **newm;
1.187 brouard 3406: double agexact;
1.214 brouard 3407: double agebegin, ageend;
1.126 brouard 3408:
3409: /* Hstepm could be zero and should return the unit matrix */
3410: for (i=1;i<=nlstate+ndeath;i++)
3411: for (j=1;j<=nlstate+ndeath;j++){
3412: oldm[i][j]=(i==j ? 1.0 : 0.0);
3413: po[i][j][0]=(i==j ? 1.0 : 0.0);
3414: }
3415: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3416: for(h=1; h <=nhstepm; h++){
3417: for(d=1; d <=hstepm; d++){
3418: newm=savm;
3419: /* Covariates have to be included here again */
3420: cov[1]=1.;
1.214 brouard 3421: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3422: cov[2]=agexact;
1.319 brouard 3423: if(nagesqr==1){
1.227 brouard 3424: cov[3]= agexact*agexact;
1.319 brouard 3425: }
1.235 brouard 3426: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
1.319 brouard 3427: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3428: /* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 */
3429: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3430: /* k 1 2 3 4 5 6 7 8 9 */
3431: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
3432: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
3433: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
3434: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1.235 brouard 3435: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3436: /* printf("hpxij Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,Tvar[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); */
3437: }
3438: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3439: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.319 brouard 3440: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
1.235 brouard 3441: /* 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]); */
3442: }
1.319 brouard 3443: for (k=1; k<=cptcovage;k++){ /* For product with age V1+V1*age +V4 +age*V3 */
3444: /* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*/
3445: /* */
3446: if(Dummy[Tage[k]]== 2){ /* dummy with age */
3447: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age *\/ */
1.235 brouard 3448: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 3449: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
3450: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.235 brouard 3451: }
3452: /* 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]); */
3453: }
1.319 brouard 3454: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 3455: /* printf("hPxij Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2]); */
1.319 brouard 3456: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
3457: if(Dummy[Tvard[k][1]==0]){
3458: if(Dummy[Tvard[k][2]==0]){
3459: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3460: }else{
3461: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
3462: }
3463: }else{
3464: if(Dummy[Tvard[k][2]==0]){
3465: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
3466: }else{
3467: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3468: }
3469: }
1.235 brouard 3470: }
3471: /* for (k=1; k<=cptcovn;k++) */
3472: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3473: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3474: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3475: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3476: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3477:
3478:
1.126 brouard 3479: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3480: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3481: /* right multiplication of oldm by the current matrix */
1.126 brouard 3482: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3483: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3484: /* if((int)age == 70){ */
3485: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3486: /* for(i=1; i<=nlstate+ndeath; i++) { */
3487: /* printf("%d pmmij ",i); */
3488: /* for(j=1;j<=nlstate+ndeath;j++) { */
3489: /* printf("%f ",pmmij[i][j]); */
3490: /* } */
3491: /* printf(" oldm "); */
3492: /* for(j=1;j<=nlstate+ndeath;j++) { */
3493: /* printf("%f ",oldm[i][j]); */
3494: /* } */
3495: /* printf("\n"); */
3496: /* } */
3497: /* } */
1.126 brouard 3498: savm=oldm;
3499: oldm=newm;
3500: }
3501: for(i=1; i<=nlstate+ndeath; i++)
3502: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3503: po[i][j][h]=newm[i][j];
3504: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3505: }
1.128 brouard 3506: /*printf("h=%d ",h);*/
1.126 brouard 3507: } /* end h */
1.267 brouard 3508: /* printf("\n H=%d \n",h); */
1.126 brouard 3509: return po;
3510: }
3511:
1.217 brouard 3512: /************* Higher Back Matrix Product ***************/
1.218 brouard 3513: /* 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 3514: 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 3515: {
1.266 brouard 3516: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3517: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3518: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3519: nhstepm*hstepm matrices.
3520: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3521: (typically every 2 years instead of every month which is too big
1.217 brouard 3522: for the memory).
1.218 brouard 3523: Model is determined by parameters x and covariates have to be
1.266 brouard 3524: included manually here. Then we use a call to bmij(x and cov)
3525: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3526: */
1.217 brouard 3527:
3528: int i, j, d, h, k;
1.266 brouard 3529: double **out, cov[NCOVMAX+1], **bmij();
3530: double **newm, ***newmm;
1.217 brouard 3531: double agexact;
3532: double agebegin, ageend;
1.222 brouard 3533: double **oldm, **savm;
1.217 brouard 3534:
1.266 brouard 3535: newmm=po; /* To be saved */
3536: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3537: /* Hstepm could be zero and should return the unit matrix */
3538: for (i=1;i<=nlstate+ndeath;i++)
3539: for (j=1;j<=nlstate+ndeath;j++){
3540: oldm[i][j]=(i==j ? 1.0 : 0.0);
3541: po[i][j][0]=(i==j ? 1.0 : 0.0);
3542: }
3543: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3544: for(h=1; h <=nhstepm; h++){
3545: for(d=1; d <=hstepm; d++){
3546: newm=savm;
3547: /* Covariates have to be included here again */
3548: cov[1]=1.;
1.271 brouard 3549: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3550: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3551: /* Debug */
3552: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3553: cov[2]=agexact;
3554: if(nagesqr==1)
1.222 brouard 3555: cov[3]= agexact*agexact;
1.325 brouard 3556: for (k=1; k<=nsd;k++){ /* For single dummy covariates only *//* cptcovn error */
1.266 brouard 3557: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3558: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
1.325 brouard 3559: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];/* Bug valgrind */
1.266 brouard 3560: /* 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)); */
3561: }
1.267 brouard 3562: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3563: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3564: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3565: /* 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]); */
3566: }
1.319 brouard 3567: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 *//* For product with age */
3568: /* if(Dummy[Tvar[Tage[k]]]== 2){ /\* dummy with age error!!!*\/ */
3569: if(Dummy[Tage[k]]== 2){ /* dummy with age */
1.267 brouard 3570: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
1.319 brouard 3571: } else if(Dummy[Tage[k]]== 3){ /* quantitative with age */
1.267 brouard 3572: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3573: }
3574: /* 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]); */
3575: }
3576: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3577: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.325 brouard 3578: if(Dummy[Tvard[k][1]==0]){
3579: if(Dummy[Tvard[k][2]==0]){
3580: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3581: }else{
3582: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
3583: }
3584: }else{
3585: if(Dummy[Tvard[k][2]==0]){
3586: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
3587: }else{
3588: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
3589: }
3590: }
1.267 brouard 3591: }
1.217 brouard 3592: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3593: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3594:
1.218 brouard 3595: /* Careful transposed matrix */
1.266 brouard 3596: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3597: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3598: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3599: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3600: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3601: /* if((int)age == 70){ */
3602: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3603: /* for(i=1; i<=nlstate+ndeath; i++) { */
3604: /* printf("%d pmmij ",i); */
3605: /* for(j=1;j<=nlstate+ndeath;j++) { */
3606: /* printf("%f ",pmmij[i][j]); */
3607: /* } */
3608: /* printf(" oldm "); */
3609: /* for(j=1;j<=nlstate+ndeath;j++) { */
3610: /* printf("%f ",oldm[i][j]); */
3611: /* } */
3612: /* printf("\n"); */
3613: /* } */
3614: /* } */
3615: savm=oldm;
3616: oldm=newm;
3617: }
3618: for(i=1; i<=nlstate+ndeath; i++)
3619: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3620: po[i][j][h]=newm[i][j];
1.268 brouard 3621: /* if(h==nhstepm) */
3622: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3623: }
1.268 brouard 3624: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3625: } /* end h */
1.268 brouard 3626: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3627: return po;
3628: }
3629:
3630:
1.162 brouard 3631: #ifdef NLOPT
3632: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3633: double fret;
3634: double *xt;
3635: int j;
3636: myfunc_data *d2 = (myfunc_data *) pd;
3637: /* xt = (p1-1); */
3638: xt=vector(1,n);
3639: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3640:
3641: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3642: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3643: printf("Function = %.12lf ",fret);
3644: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3645: printf("\n");
3646: free_vector(xt,1,n);
3647: return fret;
3648: }
3649: #endif
1.126 brouard 3650:
3651: /*************** log-likelihood *************/
3652: double func( double *x)
3653: {
1.226 brouard 3654: int i, ii, j, k, mi, d, kk;
3655: int ioffset=0;
3656: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3657: double **out;
3658: double lli; /* Individual log likelihood */
3659: int s1, s2;
1.228 brouard 3660: 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 3661: double bbh, survp;
3662: long ipmx;
3663: double agexact;
3664: /*extern weight */
3665: /* We are differentiating ll according to initial status */
3666: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3667: /*for(i=1;i<imx;i++)
3668: printf(" %d\n",s[4][i]);
3669: */
1.162 brouard 3670:
1.226 brouard 3671: ++countcallfunc;
1.162 brouard 3672:
1.226 brouard 3673: cov[1]=1.;
1.126 brouard 3674:
1.226 brouard 3675: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3676: ioffset=0;
1.226 brouard 3677: if(mle==1){
3678: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3679: /* Computes the values of the ncovmodel covariates of the model
3680: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3681: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3682: to be observed in j being in i according to the model.
3683: */
1.243 brouard 3684: ioffset=2+nagesqr ;
1.233 brouard 3685: /* Fixed */
1.319 brouard 3686: for (k=1; k<=ncovf;k++){ /* For each fixed covariate dummu or quant or prod */
3687: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3688: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3689: /* 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 3690: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.319 brouard 3691: 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)*/
3692: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3693: }
1.226 brouard 3694: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3695: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3696: has been calculated etc */
3697: /* For an individual i, wav[i] gives the number of effective waves */
3698: /* We compute the contribution to Likelihood of each effective transition
3699: mw[mi][i] is real wave of the mi th effectve wave */
3700: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3701: s2=s[mw[mi+1][i]][i];
3702: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3703: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3704: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3705: */
3706: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3707: 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*/
3708: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? */
1.242 brouard 3709: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3710: }
3711: for (ii=1;ii<=nlstate+ndeath;ii++)
3712: for (j=1;j<=nlstate+ndeath;j++){
3713: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3714: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3715: }
3716: for(d=0; d<dh[mi][i]; d++){
3717: newm=savm;
3718: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3719: cov[2]=agexact;
3720: if(nagesqr==1)
3721: cov[3]= agexact*agexact; /* Should be changed here */
3722: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3723: if(!FixedV[Tvar[Tage[kk]]])
3724: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3725: else
3726: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3727: }
3728: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3729: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3730: savm=oldm;
3731: oldm=newm;
3732: } /* end mult */
3733:
3734: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3735: /* But now since version 0.9 we anticipate for bias at large stepm.
3736: * If stepm is larger than one month (smallest stepm) and if the exact delay
3737: * (in months) between two waves is not a multiple of stepm, we rounded to
3738: * the nearest (and in case of equal distance, to the lowest) interval but now
3739: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3740: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3741: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3742: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3743: * -stepm/2 to stepm/2 .
3744: * For stepm=1 the results are the same as for previous versions of Imach.
3745: * For stepm > 1 the results are less biased than in previous versions.
3746: */
1.234 brouard 3747: s1=s[mw[mi][i]][i];
3748: s2=s[mw[mi+1][i]][i];
3749: bbh=(double)bh[mi][i]/(double)stepm;
3750: /* bias bh is positive if real duration
3751: * is higher than the multiple of stepm and negative otherwise.
3752: */
3753: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3754: if( s2 > nlstate){
3755: /* i.e. if s2 is a death state and if the date of death is known
3756: then the contribution to the likelihood is the probability to
3757: die between last step unit time and current step unit time,
3758: which is also equal to probability to die before dh
3759: minus probability to die before dh-stepm .
3760: In version up to 0.92 likelihood was computed
3761: as if date of death was unknown. Death was treated as any other
3762: health state: the date of the interview describes the actual state
3763: and not the date of a change in health state. The former idea was
3764: to consider that at each interview the state was recorded
3765: (healthy, disable or death) and IMaCh was corrected; but when we
3766: introduced the exact date of death then we should have modified
3767: the contribution of an exact death to the likelihood. This new
3768: contribution is smaller and very dependent of the step unit
3769: stepm. It is no more the probability to die between last interview
3770: and month of death but the probability to survive from last
3771: interview up to one month before death multiplied by the
3772: probability to die within a month. Thanks to Chris
3773: Jackson for correcting this bug. Former versions increased
3774: mortality artificially. The bad side is that we add another loop
3775: which slows down the processing. The difference can be up to 10%
3776: lower mortality.
3777: */
3778: /* If, at the beginning of the maximization mostly, the
3779: cumulative probability or probability to be dead is
3780: constant (ie = 1) over time d, the difference is equal to
3781: 0. out[s1][3] = savm[s1][3]: probability, being at state
3782: s1 at precedent wave, to be dead a month before current
3783: wave is equal to probability, being at state s1 at
3784: precedent wave, to be dead at mont of the current
3785: wave. Then the observed probability (that this person died)
3786: is null according to current estimated parameter. In fact,
3787: it should be very low but not zero otherwise the log go to
3788: infinity.
3789: */
1.183 brouard 3790: /* #ifdef INFINITYORIGINAL */
3791: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3792: /* #else */
3793: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3794: /* lli=log(mytinydouble); */
3795: /* else */
3796: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3797: /* #endif */
1.226 brouard 3798: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3799:
1.226 brouard 3800: } else if ( s2==-1 ) { /* alive */
3801: for (j=1,survp=0. ; j<=nlstate; j++)
3802: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3803: /*survp += out[s1][j]; */
3804: lli= log(survp);
3805: }
3806: else if (s2==-4) {
3807: for (j=3,survp=0. ; j<=nlstate; j++)
3808: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3809: lli= log(survp);
3810: }
3811: else if (s2==-5) {
3812: for (j=1,survp=0. ; j<=2; j++)
3813: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3814: lli= log(survp);
3815: }
3816: else{
3817: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3818: /* 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 */
3819: }
3820: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3821: /*if(lli ==000.0)*/
3822: /*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); */
3823: ipmx +=1;
3824: sw += weight[i];
3825: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3826: /* if (lli < log(mytinydouble)){ */
3827: /* 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); */
3828: /* 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]); */
3829: /* } */
3830: } /* end of wave */
3831: } /* end of individual */
3832: } else if(mle==2){
3833: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 3834: ioffset=2+nagesqr ;
3835: for (k=1; k<=ncovf;k++)
3836: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 3837: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3838: for(k=1; k <= ncovv ; k++){
3839: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3840: }
1.226 brouard 3841: for (ii=1;ii<=nlstate+ndeath;ii++)
3842: for (j=1;j<=nlstate+ndeath;j++){
3843: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3844: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3845: }
3846: for(d=0; d<=dh[mi][i]; d++){
3847: newm=savm;
3848: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3849: cov[2]=agexact;
3850: if(nagesqr==1)
3851: cov[3]= agexact*agexact;
3852: for (kk=1; kk<=cptcovage;kk++) {
3853: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3854: }
3855: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3856: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3857: savm=oldm;
3858: oldm=newm;
3859: } /* end mult */
3860:
3861: s1=s[mw[mi][i]][i];
3862: s2=s[mw[mi+1][i]][i];
3863: bbh=(double)bh[mi][i]/(double)stepm;
3864: 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 */
3865: ipmx +=1;
3866: sw += weight[i];
3867: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3868: } /* end of wave */
3869: } /* end of individual */
3870: } else if(mle==3){ /* exponential inter-extrapolation */
3871: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3872: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3873: for(mi=1; mi<= wav[i]-1; mi++){
3874: for (ii=1;ii<=nlstate+ndeath;ii++)
3875: for (j=1;j<=nlstate+ndeath;j++){
3876: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3877: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3878: }
3879: for(d=0; d<dh[mi][i]; d++){
3880: newm=savm;
3881: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3882: cov[2]=agexact;
3883: if(nagesqr==1)
3884: cov[3]= agexact*agexact;
3885: for (kk=1; kk<=cptcovage;kk++) {
3886: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3887: }
3888: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3889: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3890: savm=oldm;
3891: oldm=newm;
3892: } /* end mult */
3893:
3894: s1=s[mw[mi][i]][i];
3895: s2=s[mw[mi+1][i]][i];
3896: bbh=(double)bh[mi][i]/(double)stepm;
3897: 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 */
3898: ipmx +=1;
3899: sw += weight[i];
3900: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3901: } /* end of wave */
3902: } /* end of individual */
3903: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3904: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3905: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3906: for(mi=1; mi<= wav[i]-1; mi++){
3907: for (ii=1;ii<=nlstate+ndeath;ii++)
3908: for (j=1;j<=nlstate+ndeath;j++){
3909: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3910: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3911: }
3912: for(d=0; d<dh[mi][i]; d++){
3913: newm=savm;
3914: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3915: cov[2]=agexact;
3916: if(nagesqr==1)
3917: cov[3]= agexact*agexact;
3918: for (kk=1; kk<=cptcovage;kk++) {
3919: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3920: }
1.126 brouard 3921:
1.226 brouard 3922: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3923: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3924: savm=oldm;
3925: oldm=newm;
3926: } /* end mult */
3927:
3928: s1=s[mw[mi][i]][i];
3929: s2=s[mw[mi+1][i]][i];
3930: if( s2 > nlstate){
3931: lli=log(out[s1][s2] - savm[s1][s2]);
3932: } else if ( s2==-1 ) { /* alive */
3933: for (j=1,survp=0. ; j<=nlstate; j++)
3934: survp += out[s1][j];
3935: lli= log(survp);
3936: }else{
3937: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3938: }
3939: ipmx +=1;
3940: sw += weight[i];
3941: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3942: /* 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 3943: } /* end of wave */
3944: } /* end of individual */
3945: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3946: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3947: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3948: for(mi=1; mi<= wav[i]-1; mi++){
3949: for (ii=1;ii<=nlstate+ndeath;ii++)
3950: for (j=1;j<=nlstate+ndeath;j++){
3951: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3952: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3953: }
3954: for(d=0; d<dh[mi][i]; d++){
3955: newm=savm;
3956: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3957: cov[2]=agexact;
3958: if(nagesqr==1)
3959: cov[3]= agexact*agexact;
3960: for (kk=1; kk<=cptcovage;kk++) {
3961: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3962: }
1.126 brouard 3963:
1.226 brouard 3964: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3965: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3966: savm=oldm;
3967: oldm=newm;
3968: } /* end mult */
3969:
3970: s1=s[mw[mi][i]][i];
3971: s2=s[mw[mi+1][i]][i];
3972: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3973: ipmx +=1;
3974: sw += weight[i];
3975: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3976: /*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]);*/
3977: } /* end of wave */
3978: } /* end of individual */
3979: } /* End of if */
3980: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3981: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3982: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3983: return -l;
1.126 brouard 3984: }
3985:
3986: /*************** log-likelihood *************/
3987: double funcone( double *x)
3988: {
1.228 brouard 3989: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3990: int i, ii, j, k, mi, d, kk;
1.228 brouard 3991: int ioffset=0;
1.131 brouard 3992: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3993: double **out;
3994: double lli; /* Individual log likelihood */
3995: double llt;
3996: int s1, s2;
1.228 brouard 3997: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3998:
1.126 brouard 3999: double bbh, survp;
1.187 brouard 4000: double agexact;
1.214 brouard 4001: double agebegin, ageend;
1.126 brouard 4002: /*extern weight */
4003: /* We are differentiating ll according to initial status */
4004: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4005: /*for(i=1;i<imx;i++)
4006: printf(" %d\n",s[4][i]);
4007: */
4008: cov[1]=1.;
4009:
4010: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4011: ioffset=0;
4012: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 4013: /* ioffset=2+nagesqr+cptcovage; */
4014: ioffset=2+nagesqr;
1.232 brouard 4015: /* Fixed */
1.224 brouard 4016: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4017: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.311 brouard 4018: 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 4019: 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)*/
4020: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4021: /* cov[2+6]=covar[Tvar[6]][i]; */
4022: /* cov[2+6]=covar[2][i]; V2 */
4023: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4024: /* cov[2+7]=covar[Tvar[7]][i]; */
4025: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4026: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4027: /* cov[2+9]=covar[Tvar[9]][i]; */
4028: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4029: }
1.232 brouard 4030: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4031: /* 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?)*\/ */
4032: /* } */
1.231 brouard 4033: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4034: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4035: /* } */
1.225 brouard 4036:
1.233 brouard 4037:
4038: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 4039: /* Wave varying (but not age varying) */
4040: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 4041: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
4042: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4043: }
1.232 brouard 4044: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 4045: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4046: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4047: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4048: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4049: /* 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 4050: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4051: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4052: /* /\* 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]); *\/ */
4053: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4054: /* } */
1.126 brouard 4055: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4056: for (j=1;j<=nlstate+ndeath;j++){
4057: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4058: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4059: }
1.214 brouard 4060:
4061: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4062: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4063: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4064: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4065: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4066: and mw[mi+1][i]. dh depends on stepm.*/
4067: newm=savm;
1.247 brouard 4068: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4069: cov[2]=agexact;
4070: if(nagesqr==1)
4071: cov[3]= agexact*agexact;
4072: for (kk=1; kk<=cptcovage;kk++) {
4073: if(!FixedV[Tvar[Tage[kk]]])
4074: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4075: else
4076: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
4077: }
4078: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4079: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4080: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4081: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4082: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4083: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4084: savm=oldm;
4085: oldm=newm;
1.126 brouard 4086: } /* end mult */
4087:
4088: s1=s[mw[mi][i]][i];
4089: s2=s[mw[mi+1][i]][i];
1.217 brouard 4090: /* if(s2==-1){ */
1.268 brouard 4091: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4092: /* /\* exit(1); *\/ */
4093: /* } */
1.126 brouard 4094: bbh=(double)bh[mi][i]/(double)stepm;
4095: /* bias is positive if real duration
4096: * is higher than the multiple of stepm and negative otherwise.
4097: */
4098: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4099: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4100: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4101: for (j=1,survp=0. ; j<=nlstate; j++)
4102: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4103: lli= log(survp);
1.126 brouard 4104: }else if (mle==1){
1.242 brouard 4105: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4106: } else if(mle==2){
1.242 brouard 4107: 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 4108: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4109: 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 4110: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4111: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4112: } else{ /* mle=0 back to 1 */
1.242 brouard 4113: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4114: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4115: } /* End of if */
4116: ipmx +=1;
4117: sw += weight[i];
4118: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 4119: /*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 4120: if(globpr){
1.246 brouard 4121: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4122: %11.6f %11.6f %11.6f ", \
1.242 brouard 4123: 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 4124: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 4125: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4126: llt +=ll[k]*gipmx/gsw;
4127: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
4128: }
4129: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 4130: }
1.232 brouard 4131: } /* end of wave */
4132: } /* end of individual */
4133: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4134: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4135: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4136: if(globpr==0){ /* First time we count the contributions and weights */
4137: gipmx=ipmx;
4138: gsw=sw;
4139: }
4140: return -l;
1.126 brouard 4141: }
4142:
4143:
4144: /*************** function likelione ***********/
1.292 brouard 4145: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4146: {
4147: /* This routine should help understanding what is done with
4148: the selection of individuals/waves and
4149: to check the exact contribution to the likelihood.
4150: Plotting could be done.
4151: */
4152: int k;
4153:
4154: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4155: strcpy(fileresilk,"ILK_");
1.202 brouard 4156: strcat(fileresilk,fileresu);
1.126 brouard 4157: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4158: printf("Problem with resultfile: %s\n", fileresilk);
4159: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4160: }
1.214 brouard 4161: 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");
4162: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4163: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4164: for(k=1; k<=nlstate; k++)
4165: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4166: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4167: }
4168:
1.292 brouard 4169: *fretone=(*func)(p);
1.126 brouard 4170: if(*globpri !=0){
4171: fclose(ficresilk);
1.205 brouard 4172: if (mle ==0)
4173: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4174: else if(mle >=1)
4175: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4176: 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 4177: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4178:
4179: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4180: 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 4181: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4182: }
1.207 brouard 4183: 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 4184: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4185: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4186: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4187: fflush(fichtm);
1.205 brouard 4188: }
1.126 brouard 4189: return;
4190: }
4191:
4192:
4193: /*********** Maximum Likelihood Estimation ***************/
4194:
4195: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4196: {
1.319 brouard 4197: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4198: double **xi;
4199: double fret;
4200: double fretone; /* Only one call to likelihood */
4201: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4202:
4203: #ifdef NLOPT
4204: int creturn;
4205: nlopt_opt opt;
4206: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4207: double *lb;
4208: double minf; /* the minimum objective value, upon return */
4209: double * p1; /* Shifted parameters from 0 instead of 1 */
4210: myfunc_data dinst, *d = &dinst;
4211: #endif
4212:
4213:
1.126 brouard 4214: xi=matrix(1,npar,1,npar);
4215: for (i=1;i<=npar;i++)
4216: for (j=1;j<=npar;j++)
4217: xi[i][j]=(i==j ? 1.0 : 0.0);
4218: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4219: strcpy(filerespow,"POW_");
1.126 brouard 4220: strcat(filerespow,fileres);
4221: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4222: printf("Problem with resultfile: %s\n", filerespow);
4223: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4224: }
4225: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4226: for (i=1;i<=nlstate;i++)
4227: for(j=1;j<=nlstate+ndeath;j++)
4228: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4229: fprintf(ficrespow,"\n");
1.162 brouard 4230: #ifdef POWELL
1.319 brouard 4231: #ifdef LINMINORIGINAL
4232: #else /* LINMINORIGINAL */
4233:
4234: flatdir=ivector(1,npar);
4235: for (j=1;j<=npar;j++) flatdir[j]=0;
4236: #endif /*LINMINORIGINAL */
4237:
4238: #ifdef FLATSUP
4239: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4240: /* reorganizing p by suppressing flat directions */
4241: for(i=1, jk=1; i <=nlstate; i++){
4242: for(k=1; k <=(nlstate+ndeath); k++){
4243: if (k != i) {
4244: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4245: if(flatdir[jk]==1){
4246: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4247: }
4248: for(j=1; j <=ncovmodel; j++){
4249: printf("%12.7f ",p[jk]);
4250: jk++;
4251: }
4252: printf("\n");
4253: }
4254: }
4255: }
4256: /* skipping */
4257: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4258: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4259: for(k=1; k <=(nlstate+ndeath); k++){
4260: if (k != i) {
4261: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4262: if(flatdir[jk]==1){
4263: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4264: for(j=1; j <=ncovmodel; jk++,j++){
4265: printf(" p[%d]=%12.7f",jk, p[jk]);
4266: /*q[jjk]=p[jk];*/
4267: }
4268: }else{
4269: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4270: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4271: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4272: /*q[jjk]=p[jk];*/
4273: }
4274: }
4275: printf("\n");
4276: }
4277: fflush(stdout);
4278: }
4279: }
4280: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4281: #else /* FLATSUP */
1.126 brouard 4282: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4283: #endif /* FLATSUP */
4284:
4285: #ifdef LINMINORIGINAL
4286: #else
4287: free_ivector(flatdir,1,npar);
4288: #endif /* LINMINORIGINAL*/
4289: #endif /* POWELL */
1.126 brouard 4290:
1.162 brouard 4291: #ifdef NLOPT
4292: #ifdef NEWUOA
4293: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4294: #else
4295: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4296: #endif
4297: lb=vector(0,npar-1);
4298: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4299: nlopt_set_lower_bounds(opt, lb);
4300: nlopt_set_initial_step1(opt, 0.1);
4301:
4302: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4303: d->function = func;
4304: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4305: nlopt_set_min_objective(opt, myfunc, d);
4306: nlopt_set_xtol_rel(opt, ftol);
4307: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4308: printf("nlopt failed! %d\n",creturn);
4309: }
4310: else {
4311: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4312: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4313: iter=1; /* not equal */
4314: }
4315: nlopt_destroy(opt);
4316: #endif
1.319 brouard 4317: #ifdef FLATSUP
4318: /* npared = npar -flatd/ncovmodel; */
4319: /* xired= matrix(1,npared,1,npared); */
4320: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4321: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4322: /* free_matrix(xire,1,npared,1,npared); */
4323: #else /* FLATSUP */
4324: #endif /* FLATSUP */
1.126 brouard 4325: free_matrix(xi,1,npar,1,npar);
4326: fclose(ficrespow);
1.203 brouard 4327: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4328: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4329: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4330:
4331: }
4332:
4333: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4334: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4335: {
4336: double **a,**y,*x,pd;
1.203 brouard 4337: /* double **hess; */
1.164 brouard 4338: int i, j;
1.126 brouard 4339: int *indx;
4340:
4341: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4342: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4343: void lubksb(double **a, int npar, int *indx, double b[]) ;
4344: void ludcmp(double **a, int npar, int *indx, double *d) ;
4345: double gompertz(double p[]);
1.203 brouard 4346: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4347:
4348: printf("\nCalculation of the hessian matrix. Wait...\n");
4349: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4350: for (i=1;i<=npar;i++){
1.203 brouard 4351: printf("%d-",i);fflush(stdout);
4352: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4353:
4354: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4355:
4356: /* printf(" %f ",p[i]);
4357: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4358: }
4359:
4360: for (i=1;i<=npar;i++) {
4361: for (j=1;j<=npar;j++) {
4362: if (j>i) {
1.203 brouard 4363: printf(".%d-%d",i,j);fflush(stdout);
4364: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4365: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4366:
4367: hess[j][i]=hess[i][j];
4368: /*printf(" %lf ",hess[i][j]);*/
4369: }
4370: }
4371: }
4372: printf("\n");
4373: fprintf(ficlog,"\n");
4374:
4375: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4376: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4377:
4378: a=matrix(1,npar,1,npar);
4379: y=matrix(1,npar,1,npar);
4380: x=vector(1,npar);
4381: indx=ivector(1,npar);
4382: for (i=1;i<=npar;i++)
4383: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4384: ludcmp(a,npar,indx,&pd);
4385:
4386: for (j=1;j<=npar;j++) {
4387: for (i=1;i<=npar;i++) x[i]=0;
4388: x[j]=1;
4389: lubksb(a,npar,indx,x);
4390: for (i=1;i<=npar;i++){
4391: matcov[i][j]=x[i];
4392: }
4393: }
4394:
4395: printf("\n#Hessian matrix#\n");
4396: fprintf(ficlog,"\n#Hessian matrix#\n");
4397: for (i=1;i<=npar;i++) {
4398: for (j=1;j<=npar;j++) {
1.203 brouard 4399: printf("%.6e ",hess[i][j]);
4400: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4401: }
4402: printf("\n");
4403: fprintf(ficlog,"\n");
4404: }
4405:
1.203 brouard 4406: /* printf("\n#Covariance matrix#\n"); */
4407: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4408: /* for (i=1;i<=npar;i++) { */
4409: /* for (j=1;j<=npar;j++) { */
4410: /* printf("%.6e ",matcov[i][j]); */
4411: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4412: /* } */
4413: /* printf("\n"); */
4414: /* fprintf(ficlog,"\n"); */
4415: /* } */
4416:
1.126 brouard 4417: /* Recompute Inverse */
1.203 brouard 4418: /* for (i=1;i<=npar;i++) */
4419: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4420: /* ludcmp(a,npar,indx,&pd); */
4421:
4422: /* printf("\n#Hessian matrix recomputed#\n"); */
4423:
4424: /* for (j=1;j<=npar;j++) { */
4425: /* for (i=1;i<=npar;i++) x[i]=0; */
4426: /* x[j]=1; */
4427: /* lubksb(a,npar,indx,x); */
4428: /* for (i=1;i<=npar;i++){ */
4429: /* y[i][j]=x[i]; */
4430: /* printf("%.3e ",y[i][j]); */
4431: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4432: /* } */
4433: /* printf("\n"); */
4434: /* fprintf(ficlog,"\n"); */
4435: /* } */
4436:
4437: /* Verifying the inverse matrix */
4438: #ifdef DEBUGHESS
4439: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4440:
1.203 brouard 4441: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4442: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4443:
4444: for (j=1;j<=npar;j++) {
4445: for (i=1;i<=npar;i++){
1.203 brouard 4446: printf("%.2f ",y[i][j]);
4447: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4448: }
4449: printf("\n");
4450: fprintf(ficlog,"\n");
4451: }
1.203 brouard 4452: #endif
1.126 brouard 4453:
4454: free_matrix(a,1,npar,1,npar);
4455: free_matrix(y,1,npar,1,npar);
4456: free_vector(x,1,npar);
4457: free_ivector(indx,1,npar);
1.203 brouard 4458: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4459:
4460:
4461: }
4462:
4463: /*************** hessian matrix ****************/
4464: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4465: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4466: int i;
4467: int l=1, lmax=20;
1.203 brouard 4468: double k1,k2, res, fx;
1.132 brouard 4469: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4470: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4471: int k=0,kmax=10;
4472: double l1;
4473:
4474: fx=func(x);
4475: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4476: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4477: l1=pow(10,l);
4478: delts=delt;
4479: for(k=1 ; k <kmax; k=k+1){
4480: delt = delta*(l1*k);
4481: p2[theta]=x[theta] +delt;
1.145 brouard 4482: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4483: p2[theta]=x[theta]-delt;
4484: k2=func(p2)-fx;
4485: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4486: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4487:
1.203 brouard 4488: #ifdef DEBUGHESSII
1.126 brouard 4489: 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);
4490: 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);
4491: #endif
4492: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4493: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4494: k=kmax;
4495: }
4496: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4497: k=kmax; l=lmax*10;
1.126 brouard 4498: }
4499: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4500: delts=delt;
4501: }
1.203 brouard 4502: } /* End loop k */
1.126 brouard 4503: }
4504: delti[theta]=delts;
4505: return res;
4506:
4507: }
4508:
1.203 brouard 4509: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4510: {
4511: int i;
1.164 brouard 4512: int l=1, lmax=20;
1.126 brouard 4513: double k1,k2,k3,k4,res,fx;
1.132 brouard 4514: double p2[MAXPARM+1];
1.203 brouard 4515: int k, kmax=1;
4516: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4517:
4518: int firstime=0;
1.203 brouard 4519:
1.126 brouard 4520: fx=func(x);
1.203 brouard 4521: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4522: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4523: p2[thetai]=x[thetai]+delti[thetai]*k;
4524: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4525: k1=func(p2)-fx;
4526:
1.203 brouard 4527: p2[thetai]=x[thetai]+delti[thetai]*k;
4528: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4529: k2=func(p2)-fx;
4530:
1.203 brouard 4531: p2[thetai]=x[thetai]-delti[thetai]*k;
4532: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4533: k3=func(p2)-fx;
4534:
1.203 brouard 4535: p2[thetai]=x[thetai]-delti[thetai]*k;
4536: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4537: k4=func(p2)-fx;
1.203 brouard 4538: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4539: if(k1*k2*k3*k4 <0.){
1.208 brouard 4540: firstime=1;
1.203 brouard 4541: kmax=kmax+10;
1.208 brouard 4542: }
4543: if(kmax >=10 || firstime ==1){
1.246 brouard 4544: 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);
4545: 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 4546: 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);
4547: 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);
4548: }
4549: #ifdef DEBUGHESSIJ
4550: v1=hess[thetai][thetai];
4551: v2=hess[thetaj][thetaj];
4552: cv12=res;
4553: /* Computing eigen value of Hessian matrix */
4554: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4555: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4556: if ((lc2 <0) || (lc1 <0) ){
4557: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4558: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4559: 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);
4560: 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);
4561: }
1.126 brouard 4562: #endif
4563: }
4564: return res;
4565: }
4566:
1.203 brouard 4567: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4568: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4569: /* { */
4570: /* int i; */
4571: /* int l=1, lmax=20; */
4572: /* double k1,k2,k3,k4,res,fx; */
4573: /* double p2[MAXPARM+1]; */
4574: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4575: /* int k=0,kmax=10; */
4576: /* double l1; */
4577:
4578: /* fx=func(x); */
4579: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4580: /* l1=pow(10,l); */
4581: /* delts=delt; */
4582: /* for(k=1 ; k <kmax; k=k+1){ */
4583: /* delt = delti*(l1*k); */
4584: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4585: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4586: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4587: /* k1=func(p2)-fx; */
4588:
4589: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4590: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4591: /* k2=func(p2)-fx; */
4592:
4593: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4594: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4595: /* k3=func(p2)-fx; */
4596:
4597: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4598: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4599: /* k4=func(p2)-fx; */
4600: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4601: /* #ifdef DEBUGHESSIJ */
4602: /* 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); */
4603: /* 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); */
4604: /* #endif */
4605: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4606: /* k=kmax; */
4607: /* } */
4608: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4609: /* k=kmax; l=lmax*10; */
4610: /* } */
4611: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4612: /* delts=delt; */
4613: /* } */
4614: /* } /\* End loop k *\/ */
4615: /* } */
4616: /* delti[theta]=delts; */
4617: /* return res; */
4618: /* } */
4619:
4620:
1.126 brouard 4621: /************** Inverse of matrix **************/
4622: void ludcmp(double **a, int n, int *indx, double *d)
4623: {
4624: int i,imax,j,k;
4625: double big,dum,sum,temp;
4626: double *vv;
4627:
4628: vv=vector(1,n);
4629: *d=1.0;
4630: for (i=1;i<=n;i++) {
4631: big=0.0;
4632: for (j=1;j<=n;j++)
4633: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4634: if (big == 0.0){
4635: printf(" Singular Hessian matrix at row %d:\n",i);
4636: for (j=1;j<=n;j++) {
4637: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4638: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4639: }
4640: fflush(ficlog);
4641: fclose(ficlog);
4642: nrerror("Singular matrix in routine ludcmp");
4643: }
1.126 brouard 4644: vv[i]=1.0/big;
4645: }
4646: for (j=1;j<=n;j++) {
4647: for (i=1;i<j;i++) {
4648: sum=a[i][j];
4649: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4650: a[i][j]=sum;
4651: }
4652: big=0.0;
4653: for (i=j;i<=n;i++) {
4654: sum=a[i][j];
4655: for (k=1;k<j;k++)
4656: sum -= a[i][k]*a[k][j];
4657: a[i][j]=sum;
4658: if ( (dum=vv[i]*fabs(sum)) >= big) {
4659: big=dum;
4660: imax=i;
4661: }
4662: }
4663: if (j != imax) {
4664: for (k=1;k<=n;k++) {
4665: dum=a[imax][k];
4666: a[imax][k]=a[j][k];
4667: a[j][k]=dum;
4668: }
4669: *d = -(*d);
4670: vv[imax]=vv[j];
4671: }
4672: indx[j]=imax;
4673: if (a[j][j] == 0.0) a[j][j]=TINY;
4674: if (j != n) {
4675: dum=1.0/(a[j][j]);
4676: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4677: }
4678: }
4679: free_vector(vv,1,n); /* Doesn't work */
4680: ;
4681: }
4682:
4683: void lubksb(double **a, int n, int *indx, double b[])
4684: {
4685: int i,ii=0,ip,j;
4686: double sum;
4687:
4688: for (i=1;i<=n;i++) {
4689: ip=indx[i];
4690: sum=b[ip];
4691: b[ip]=b[i];
4692: if (ii)
4693: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4694: else if (sum) ii=i;
4695: b[i]=sum;
4696: }
4697: for (i=n;i>=1;i--) {
4698: sum=b[i];
4699: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4700: b[i]=sum/a[i][i];
4701: }
4702: }
4703:
4704: void pstamp(FILE *fichier)
4705: {
1.196 brouard 4706: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4707: }
4708:
1.297 brouard 4709: void date2dmy(double date,double *day, double *month, double *year){
4710: double yp=0., yp1=0., yp2=0.;
4711:
4712: yp1=modf(date,&yp);/* extracts integral of date in yp and
4713: fractional in yp1 */
4714: *year=yp;
4715: yp2=modf((yp1*12),&yp);
4716: *month=yp;
4717: yp1=modf((yp2*30.5),&yp);
4718: *day=yp;
4719: if(*day==0) *day=1;
4720: if(*month==0) *month=1;
4721: }
4722:
1.253 brouard 4723:
4724:
1.126 brouard 4725: /************ Frequencies ********************/
1.251 brouard 4726: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4727: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4728: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4729: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4730:
1.265 brouard 4731: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4732: int iind=0, iage=0;
4733: int mi; /* Effective wave */
4734: int first;
4735: double ***freq; /* Frequencies */
1.268 brouard 4736: 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 */
4737: 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 4738: double *meanq, *stdq, *idq;
1.226 brouard 4739: double **meanqt;
4740: double *pp, **prop, *posprop, *pospropt;
4741: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4742: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4743: double agebegin, ageend;
4744:
4745: pp=vector(1,nlstate);
1.251 brouard 4746: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4747: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4748: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4749: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4750: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4751: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4752: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4753: meanqt=matrix(1,lastpass,1,nqtveff);
4754: strcpy(fileresp,"P_");
4755: strcat(fileresp,fileresu);
4756: /*strcat(fileresphtm,fileresu);*/
4757: if((ficresp=fopen(fileresp,"w"))==NULL) {
4758: printf("Problem with prevalence resultfile: %s\n", fileresp);
4759: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4760: exit(0);
4761: }
1.240 brouard 4762:
1.226 brouard 4763: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4764: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4765: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4766: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4767: fflush(ficlog);
4768: exit(70);
4769: }
4770: else{
4771: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4772: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4773: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4774: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4775: }
1.319 brouard 4776: 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 4777:
1.226 brouard 4778: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4779: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4780: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4781: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4782: fflush(ficlog);
4783: exit(70);
1.240 brouard 4784: } else{
1.226 brouard 4785: 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 4786: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4787: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4788: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4789: }
1.319 brouard 4790: 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 4791:
1.253 brouard 4792: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4793: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4794: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4795: j1=0;
1.126 brouard 4796:
1.227 brouard 4797: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4798: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4799: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4800:
4801:
1.226 brouard 4802: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4803: reference=low_education V1=0,V2=0
4804: med_educ V1=1 V2=0,
4805: high_educ V1=0 V2=1
4806: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4807: */
1.249 brouard 4808: dateintsum=0;
4809: k2cpt=0;
4810:
1.253 brouard 4811: if(cptcoveff == 0 )
1.265 brouard 4812: nl=1; /* Constant and age model only */
1.253 brouard 4813: else
4814: nl=2;
1.265 brouard 4815:
4816: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4817: /* Loop on nj=1 or 2 if dummy covariates j!=0
4818: * Loop on j1(1 to 2**cptcoveff) covariate combination
4819: * freq[s1][s2][iage] =0.
4820: * Loop on iind
4821: * ++freq[s1][s2][iage] weighted
4822: * end iind
4823: * if covariate and j!0
4824: * headers Variable on one line
4825: * endif cov j!=0
4826: * header of frequency table by age
4827: * Loop on age
4828: * pp[s1]+=freq[s1][s2][iage] weighted
4829: * pos+=freq[s1][s2][iage] weighted
4830: * Loop on s1 initial state
4831: * fprintf(ficresp
4832: * end s1
4833: * end age
4834: * if j!=0 computes starting values
4835: * end compute starting values
4836: * end j1
4837: * end nl
4838: */
1.253 brouard 4839: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4840: if(nj==1)
4841: j=0; /* First pass for the constant */
1.265 brouard 4842: else{
1.253 brouard 4843: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4844: }
1.251 brouard 4845: first=1;
1.265 brouard 4846: 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 4847: posproptt=0.;
4848: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4849: scanf("%d", i);*/
4850: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4851: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4852: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4853: freq[i][s2][m]=0;
1.251 brouard 4854:
4855: for (i=1; i<=nlstate; i++) {
1.240 brouard 4856: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4857: prop[i][m]=0;
4858: posprop[i]=0;
4859: pospropt[i]=0;
4860: }
1.283 brouard 4861: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4862: idq[z1]=0.;
4863: meanq[z1]=0.;
4864: stdq[z1]=0.;
1.283 brouard 4865: }
4866: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4867: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4868: /* meanqt[m][z1]=0.; */
4869: /* } */
4870: /* } */
1.251 brouard 4871: /* dateintsum=0; */
4872: /* k2cpt=0; */
4873:
1.265 brouard 4874: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4875: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4876: bool=1;
4877: if(j !=0){
4878: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4879: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4880: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4881: /* if(Tvaraff[z1] ==-20){ */
4882: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4883: /* }else if(Tvaraff[z1] ==-10){ */
4884: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4885: /* }else */
4886: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4887: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4888: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4889: /* 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",
4890: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4891: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4892: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4893: } /* Onlyf fixed */
4894: } /* end z1 */
4895: } /* cptcovn > 0 */
4896: } /* end any */
4897: }/* end j==0 */
1.265 brouard 4898: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4899: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4900: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4901: m=mw[mi][iind];
4902: if(j!=0){
4903: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4904: for (z1=1; z1<=cptcoveff; z1++) {
4905: if( Fixed[Tmodelind[z1]]==1){
4906: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4907: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4908: value is -1, we don't select. It differs from the
4909: constant and age model which counts them. */
4910: bool=0; /* not selected */
4911: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4912: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4913: bool=0;
4914: }
4915: }
4916: }
4917: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4918: } /* end j==0 */
4919: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4920: if(bool==1){ /*Selected */
1.251 brouard 4921: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4922: and mw[mi+1][iind]. dh depends on stepm. */
4923: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4924: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4925: if(m >=firstpass && m <=lastpass){
4926: k2=anint[m][iind]+(mint[m][iind]/12.);
4927: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4928: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4929: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4930: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4931: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4932: if (m<lastpass) {
4933: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4934: /* 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]); */
4935: if(s[m][iind]==-1)
4936: 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.));
4937: 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 4938: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
4939: if(!isnan(covar[ncovcol+z1][iind])){
4940: idq[z1]=idq[z1]+weight[iind];
4941: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4942: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
4943: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4944: }
1.284 brouard 4945: }
1.251 brouard 4946: /* if((int)agev[m][iind] == 55) */
4947: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4948: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4949: 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 4950: }
1.251 brouard 4951: } /* end if between passes */
4952: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4953: dateintsum=dateintsum+k2; /* on all covariates ?*/
4954: k2cpt++;
4955: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4956: }
1.251 brouard 4957: }else{
4958: bool=1;
4959: }/* end bool 2 */
4960: } /* end m */
1.284 brouard 4961: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4962: /* idq[z1]=idq[z1]+weight[iind]; */
4963: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4964: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4965: /* } */
1.251 brouard 4966: } /* end bool */
4967: } /* end iind = 1 to imx */
1.319 brouard 4968: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 4969: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4970:
4971:
4972: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4973: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4974: pstamp(ficresp);
1.251 brouard 4975: if (cptcoveff>0 && j!=0){
1.265 brouard 4976: pstamp(ficresp);
1.251 brouard 4977: printf( "\n#********** Variable ");
4978: fprintf(ficresp, "\n#********** Variable ");
4979: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4980: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4981: fprintf(ficlog, "\n#********** Variable ");
4982: for (z1=1; z1<=cptcoveff; z1++){
4983: if(!FixedV[Tvaraff[z1]]){
4984: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4985: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4986: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4987: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4988: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4989: }else{
1.251 brouard 4990: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4991: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4992: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4993: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4994: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4995: }
4996: }
4997: printf( "**********\n#");
4998: fprintf(ficresp, "**********\n#");
4999: fprintf(ficresphtm, "**********</h3>\n");
5000: fprintf(ficresphtmfr, "**********</h3>\n");
5001: fprintf(ficlog, "**********\n");
5002: }
1.284 brouard 5003: /*
5004: Printing means of quantitative variables if any
5005: */
5006: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5007: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5008: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5009: if(weightopt==1){
5010: printf(" Weighted mean and standard deviation of");
5011: fprintf(ficlog," Weighted mean and standard deviation of");
5012: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5013: }
1.311 brouard 5014: /* mu = \frac{w x}{\sum w}
5015: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5016: */
5017: 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]));
5018: 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]));
5019: 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 5020: }
5021: /* for (z1=1; z1<= nqtveff; z1++) { */
5022: /* for(m=1;m<=lastpass;m++){ */
5023: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5024: /* } */
5025: /* } */
1.283 brouard 5026:
1.251 brouard 5027: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 5028: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
5029: fprintf(ficresp, " Age");
5030: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.251 brouard 5031: for(i=1; i<=nlstate;i++) {
1.265 brouard 5032: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5033: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5034: }
1.265 brouard 5035: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5036: fprintf(ficresphtm, "\n");
5037:
5038: /* Header of frequency table by age */
5039: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5040: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5041: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5042: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5043: if(s2!=0 && m!=0)
5044: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5045: }
1.226 brouard 5046: }
1.251 brouard 5047: fprintf(ficresphtmfr, "\n");
5048:
5049: /* For each age */
5050: for(iage=iagemin; iage <= iagemax+3; iage++){
5051: fprintf(ficresphtm,"<tr>");
5052: if(iage==iagemax+1){
5053: fprintf(ficlog,"1");
5054: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5055: }else if(iage==iagemax+2){
5056: fprintf(ficlog,"0");
5057: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5058: }else if(iage==iagemax+3){
5059: fprintf(ficlog,"Total");
5060: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5061: }else{
1.240 brouard 5062: if(first==1){
1.251 brouard 5063: first=0;
5064: printf("See log file for details...\n");
5065: }
5066: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5067: fprintf(ficlog,"Age %d", iage);
5068: }
1.265 brouard 5069: for(s1=1; s1 <=nlstate ; s1++){
5070: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5071: pp[s1] += freq[s1][m][iage];
1.251 brouard 5072: }
1.265 brouard 5073: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5074: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5075: pos += freq[s1][m][iage];
5076: if(pp[s1]>=1.e-10){
1.251 brouard 5077: if(first==1){
1.265 brouard 5078: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5079: }
1.265 brouard 5080: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5081: }else{
5082: if(first==1)
1.265 brouard 5083: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5084: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5085: }
5086: }
5087:
1.265 brouard 5088: for(s1=1; s1 <=nlstate ; s1++){
5089: /* posprop[s1]=0; */
5090: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5091: pp[s1] += freq[s1][m][iage];
5092: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5093:
5094: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5095: pos += pp[s1]; /* pos is the total number of transitions until this age */
5096: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5097: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5098: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5099: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5100: }
5101:
5102: /* Writing ficresp */
5103: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
5104: if( iage <= iagemax){
5105: fprintf(ficresp," %d",iage);
5106: }
5107: }else if( nj==2){
5108: if( iage <= iagemax){
5109: fprintf(ficresp," %d",iage);
5110: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
5111: }
1.240 brouard 5112: }
1.265 brouard 5113: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5114: if(pos>=1.e-5){
1.251 brouard 5115: if(first==1)
1.265 brouard 5116: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5117: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5118: }else{
5119: if(first==1)
1.265 brouard 5120: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5121: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5122: }
5123: if( iage <= iagemax){
5124: if(pos>=1.e-5){
1.265 brouard 5125: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
5126: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5127: }else if( nj==2){
5128: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5129: }
5130: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5131: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5132: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5133: } else{
5134: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
5135: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5136: }
1.240 brouard 5137: }
1.265 brouard 5138: pospropt[s1] +=posprop[s1];
5139: } /* end loop s1 */
1.251 brouard 5140: /* pospropt=0.; */
1.265 brouard 5141: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5142: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5143: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5144: if(first==1){
1.265 brouard 5145: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5146: }
1.265 brouard 5147: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5148: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5149: }
1.265 brouard 5150: if(s1!=0 && m!=0)
5151: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5152: }
1.265 brouard 5153: } /* end loop s1 */
1.251 brouard 5154: posproptt=0.;
1.265 brouard 5155: for(s1=1; s1 <=nlstate; s1++){
5156: posproptt += pospropt[s1];
1.251 brouard 5157: }
5158: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5159: fprintf(ficresphtm,"</tr>\n");
5160: if((cptcoveff==0 && nj==1)|| nj==2 ) {
5161: if(iage <= iagemax)
5162: fprintf(ficresp,"\n");
1.240 brouard 5163: }
1.251 brouard 5164: if(first==1)
5165: printf("Others in log...\n");
5166: fprintf(ficlog,"\n");
5167: } /* end loop age iage */
1.265 brouard 5168:
1.251 brouard 5169: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5170: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5171: if(posproptt < 1.e-5){
1.265 brouard 5172: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5173: }else{
1.265 brouard 5174: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5175: }
1.226 brouard 5176: }
1.251 brouard 5177: fprintf(ficresphtm,"</tr>\n");
5178: fprintf(ficresphtm,"</table>\n");
5179: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5180: if(posproptt < 1.e-5){
1.251 brouard 5181: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5182: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5183: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5184: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5185: invalidvarcomb[j1]=1;
1.226 brouard 5186: }else{
1.251 brouard 5187: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
5188: invalidvarcomb[j1]=0;
1.226 brouard 5189: }
1.251 brouard 5190: fprintf(ficresphtmfr,"</table>\n");
5191: fprintf(ficlog,"\n");
5192: if(j!=0){
5193: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5194: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5195: for(k=1; k <=(nlstate+ndeath); k++){
5196: if (k != i) {
1.265 brouard 5197: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5198: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5199: if(j1==1){ /* All dummy covariates to zero */
5200: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5201: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5202: printf("%d%d ",i,k);
5203: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5204: 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]));
5205: 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]));
5206: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5207: }
1.253 brouard 5208: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5209: for(iage=iagemin; iage <= iagemax+3; iage++){
5210: x[iage]= (double)iage;
5211: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5212: /* 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 5213: }
1.268 brouard 5214: /* Some are not finite, but linreg will ignore these ages */
5215: no=0;
1.253 brouard 5216: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5217: pstart[s1]=b;
5218: pstart[s1-1]=a;
1.252 brouard 5219: }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 */
5220: 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]);
5221: 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 5222: 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 5223: printf("%d%d ",i,k);
5224: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5225: 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 5226: }else{ /* Other cases, like quantitative fixed or varying covariates */
5227: ;
5228: }
5229: /* printf("%12.7f )", param[i][jj][k]); */
5230: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5231: s1++;
1.251 brouard 5232: } /* end jj */
5233: } /* end k!= i */
5234: } /* end k */
1.265 brouard 5235: } /* end i, s1 */
1.251 brouard 5236: } /* end j !=0 */
5237: } /* end selected combination of covariate j1 */
5238: if(j==0){ /* We can estimate starting values from the occurences in each case */
5239: printf("#Freqsummary: Starting values for the constants:\n");
5240: fprintf(ficlog,"\n");
1.265 brouard 5241: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5242: for(k=1; k <=(nlstate+ndeath); k++){
5243: if (k != i) {
5244: printf("%d%d ",i,k);
5245: fprintf(ficlog,"%d%d ",i,k);
5246: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5247: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5248: if(jj==1){ /* Age has to be done */
1.265 brouard 5249: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5250: 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]));
5251: 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 5252: }
5253: /* printf("%12.7f )", param[i][jj][k]); */
5254: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5255: s1++;
1.250 brouard 5256: }
1.251 brouard 5257: printf("\n");
5258: fprintf(ficlog,"\n");
1.250 brouard 5259: }
5260: }
1.284 brouard 5261: } /* end of state i */
1.251 brouard 5262: printf("#Freqsummary\n");
5263: fprintf(ficlog,"\n");
1.265 brouard 5264: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5265: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5266: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5267: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5268: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5269: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5270: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5271: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5272: /* } */
5273: }
1.265 brouard 5274: } /* end loop s1 */
1.251 brouard 5275:
5276: printf("\n");
5277: fprintf(ficlog,"\n");
5278: } /* end j=0 */
1.249 brouard 5279: } /* end j */
1.252 brouard 5280:
1.253 brouard 5281: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5282: for(i=1, jk=1; i <=nlstate; i++){
5283: for(j=1; j <=nlstate+ndeath; j++){
5284: if(j!=i){
5285: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5286: printf("%1d%1d",i,j);
5287: fprintf(ficparo,"%1d%1d",i,j);
5288: for(k=1; k<=ncovmodel;k++){
5289: /* printf(" %lf",param[i][j][k]); */
5290: /* fprintf(ficparo," %lf",param[i][j][k]); */
5291: p[jk]=pstart[jk];
5292: printf(" %f ",pstart[jk]);
5293: fprintf(ficparo," %f ",pstart[jk]);
5294: jk++;
5295: }
5296: printf("\n");
5297: fprintf(ficparo,"\n");
5298: }
5299: }
5300: }
5301: } /* end mle=-2 */
1.226 brouard 5302: dateintmean=dateintsum/k2cpt;
1.296 brouard 5303: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5304:
1.226 brouard 5305: fclose(ficresp);
5306: fclose(ficresphtm);
5307: fclose(ficresphtmfr);
1.283 brouard 5308: free_vector(idq,1,nqfveff);
1.226 brouard 5309: free_vector(meanq,1,nqfveff);
1.284 brouard 5310: free_vector(stdq,1,nqfveff);
1.226 brouard 5311: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5312: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5313: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5314: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5315: free_vector(pospropt,1,nlstate);
5316: free_vector(posprop,1,nlstate);
1.251 brouard 5317: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5318: free_vector(pp,1,nlstate);
5319: /* End of freqsummary */
5320: }
1.126 brouard 5321:
1.268 brouard 5322: /* Simple linear regression */
5323: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5324:
5325: /* y=a+bx regression */
5326: double sumx = 0.0; /* sum of x */
5327: double sumx2 = 0.0; /* sum of x**2 */
5328: double sumxy = 0.0; /* sum of x * y */
5329: double sumy = 0.0; /* sum of y */
5330: double sumy2 = 0.0; /* sum of y**2 */
5331: double sume2 = 0.0; /* sum of square or residuals */
5332: double yhat;
5333:
5334: double denom=0;
5335: int i;
5336: int ne=*no;
5337:
5338: for ( i=ifi, ne=0;i<=ila;i++) {
5339: if(!isfinite(x[i]) || !isfinite(y[i])){
5340: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5341: continue;
5342: }
5343: ne=ne+1;
5344: sumx += x[i];
5345: sumx2 += x[i]*x[i];
5346: sumxy += x[i] * y[i];
5347: sumy += y[i];
5348: sumy2 += y[i]*y[i];
5349: denom = (ne * sumx2 - sumx*sumx);
5350: /* 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); */
5351: }
5352:
5353: denom = (ne * sumx2 - sumx*sumx);
5354: if (denom == 0) {
5355: // vertical, slope m is infinity
5356: *b = INFINITY;
5357: *a = 0;
5358: if (r) *r = 0;
5359: return 1;
5360: }
5361:
5362: *b = (ne * sumxy - sumx * sumy) / denom;
5363: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5364: if (r!=NULL) {
5365: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5366: sqrt((sumx2 - sumx*sumx/ne) *
5367: (sumy2 - sumy*sumy/ne));
5368: }
5369: *no=ne;
5370: for ( i=ifi, ne=0;i<=ila;i++) {
5371: if(!isfinite(x[i]) || !isfinite(y[i])){
5372: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5373: continue;
5374: }
5375: ne=ne+1;
5376: yhat = y[i] - *a -*b* x[i];
5377: sume2 += yhat * yhat ;
5378:
5379: denom = (ne * sumx2 - sumx*sumx);
5380: /* 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); */
5381: }
5382: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5383: *sa= *sb * sqrt(sumx2/ne);
5384:
5385: return 0;
5386: }
5387:
1.126 brouard 5388: /************ Prevalence ********************/
1.227 brouard 5389: 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)
5390: {
5391: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5392: in each health status at the date of interview (if between dateprev1 and dateprev2).
5393: We still use firstpass and lastpass as another selection.
5394: */
1.126 brouard 5395:
1.227 brouard 5396: int i, m, jk, j1, bool, z1,j, iv;
5397: int mi; /* Effective wave */
5398: int iage;
5399: double agebegin, ageend;
5400:
5401: double **prop;
5402: double posprop;
5403: double y2; /* in fractional years */
5404: int iagemin, iagemax;
5405: int first; /** to stop verbosity which is redirected to log file */
5406:
5407: iagemin= (int) agemin;
5408: iagemax= (int) agemax;
5409: /*pp=vector(1,nlstate);*/
1.251 brouard 5410: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5411: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5412: j1=0;
1.222 brouard 5413:
1.227 brouard 5414: /*j=cptcoveff;*/
5415: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5416:
1.288 brouard 5417: first=0;
1.227 brouard 5418: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5419: for (i=1; i<=nlstate; i++)
1.251 brouard 5420: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5421: prop[i][iage]=0.0;
5422: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5423: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5424: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5425:
5426: for (i=1; i<=imx; i++) { /* Each individual */
5427: bool=1;
5428: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5429: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5430: m=mw[mi][i];
5431: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5432: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5433: for (z1=1; z1<=cptcoveff; z1++){
5434: if( Fixed[Tmodelind[z1]]==1){
5435: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5436: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5437: bool=0;
5438: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5439: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5440: bool=0;
5441: }
5442: }
5443: if(bool==1){ /* Otherwise we skip that wave/person */
5444: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5445: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5446: if(m >=firstpass && m <=lastpass){
5447: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5448: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5449: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5450: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5451: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5452: 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);
5453: exit(1);
5454: }
5455: if (s[m][i]>0 && s[m][i]<=nlstate) {
5456: /*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]]);*/
5457: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5458: prop[s[m][i]][iagemax+3] += weight[i];
5459: } /* end valid statuses */
5460: } /* end selection of dates */
5461: } /* end selection of waves */
5462: } /* end bool */
5463: } /* end wave */
5464: } /* end individual */
5465: for(i=iagemin; i <= iagemax+3; i++){
5466: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5467: posprop += prop[jk][i];
5468: }
5469:
5470: for(jk=1; jk <=nlstate ; jk++){
5471: if( i <= iagemax){
5472: if(posprop>=1.e-5){
5473: probs[i][jk][j1]= prop[jk][i]/posprop;
5474: } else{
1.288 brouard 5475: if(!first){
5476: first=1;
1.266 brouard 5477: 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]);
5478: }else{
1.288 brouard 5479: 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 5480: }
5481: }
5482: }
5483: }/* end jk */
5484: }/* end i */
1.222 brouard 5485: /*} *//* end i1 */
1.227 brouard 5486: } /* end j1 */
1.222 brouard 5487:
1.227 brouard 5488: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5489: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5490: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5491: } /* End of prevalence */
1.126 brouard 5492:
5493: /************* Waves Concatenation ***************/
5494:
5495: 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)
5496: {
1.298 brouard 5497: /* 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 5498: Death is a valid wave (if date is known).
5499: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5500: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5501: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5502: */
1.126 brouard 5503:
1.224 brouard 5504: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5505: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5506: double sum=0., jmean=0.;*/
1.224 brouard 5507: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5508: int j, k=0,jk, ju, jl;
5509: double sum=0.;
5510: first=0;
1.214 brouard 5511: firstwo=0;
1.217 brouard 5512: firsthree=0;
1.218 brouard 5513: firstfour=0;
1.164 brouard 5514: jmin=100000;
1.126 brouard 5515: jmax=-1;
5516: jmean=0.;
1.224 brouard 5517:
5518: /* Treating live states */
1.214 brouard 5519: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5520: mi=0; /* First valid wave */
1.227 brouard 5521: mli=0; /* Last valid wave */
1.309 brouard 5522: m=firstpass; /* Loop on waves */
5523: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5524: 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 */
5525: mli=m-1;/* mw[++mi][i]=m-1; */
5526: }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 5527: 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 5528: mli=m;
1.224 brouard 5529: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5530: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5531: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5532: }
1.309 brouard 5533: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5534: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5535: break;
1.224 brouard 5536: #else
1.317 brouard 5537: 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 5538: if(firsthree == 0){
1.302 brouard 5539: 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 5540: firsthree=1;
1.317 brouard 5541: }else if(firsthree >=1 && firsthree < 10){
5542: 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);
5543: firsthree++;
5544: }else if(firsthree == 10){
5545: printf("Information, too many Information flags: no more reported to log either\n");
5546: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5547: firsthree++;
5548: }else{
5549: firsthree++;
1.227 brouard 5550: }
1.309 brouard 5551: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5552: mli=m;
5553: }
5554: if(s[m][i]==-2){ /* Vital status is really unknown */
5555: nbwarn++;
1.309 brouard 5556: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5557: 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);
5558: 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);
5559: }
5560: break;
5561: }
5562: break;
1.224 brouard 5563: #endif
1.227 brouard 5564: }/* End m >= lastpass */
1.126 brouard 5565: }/* end while */
1.224 brouard 5566:
1.227 brouard 5567: /* 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 5568: /* After last pass */
1.224 brouard 5569: /* Treating death states */
1.214 brouard 5570: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5571: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5572: /* } */
1.126 brouard 5573: mi++; /* Death is another wave */
5574: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5575: /* Only death is a correct wave */
1.126 brouard 5576: mw[mi][i]=m;
1.257 brouard 5577: } /* else not in a death state */
1.224 brouard 5578: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5579: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5580: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5581: 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 5582: nbwarn++;
5583: if(firstfiv==0){
1.309 brouard 5584: 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 5585: firstfiv=1;
5586: }else{
1.309 brouard 5587: 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 5588: }
1.309 brouard 5589: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5590: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5591: nberr++;
5592: if(firstwo==0){
1.309 brouard 5593: 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 5594: firstwo=1;
5595: }
1.309 brouard 5596: 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 5597: }
1.257 brouard 5598: }else{ /* if date of interview is unknown */
1.227 brouard 5599: /* death is known but not confirmed by death status at any wave */
5600: if(firstfour==0){
1.309 brouard 5601: 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 5602: firstfour=1;
5603: }
1.309 brouard 5604: 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 5605: }
1.224 brouard 5606: } /* end if date of death is known */
5607: #endif
1.309 brouard 5608: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5609: /* wav[i]=mw[mi][i]; */
1.126 brouard 5610: if(mi==0){
5611: nbwarn++;
5612: if(first==0){
1.227 brouard 5613: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5614: first=1;
1.126 brouard 5615: }
5616: if(first==1){
1.227 brouard 5617: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5618: }
5619: } /* end mi==0 */
5620: } /* End individuals */
1.214 brouard 5621: /* wav and mw are no more changed */
1.223 brouard 5622:
1.317 brouard 5623: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
5624: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
5625:
5626:
1.126 brouard 5627: for(i=1; i<=imx; i++){
5628: for(mi=1; mi<wav[i];mi++){
5629: if (stepm <=0)
1.227 brouard 5630: dh[mi][i]=1;
1.126 brouard 5631: else{
1.260 brouard 5632: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5633: if (agedc[i] < 2*AGESUP) {
5634: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5635: if(j==0) j=1; /* Survives at least one month after exam */
5636: else if(j<0){
5637: nberr++;
5638: 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]);
5639: j=1; /* Temporary Dangerous patch */
5640: 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);
5641: 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]);
5642: 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);
5643: }
5644: k=k+1;
5645: if (j >= jmax){
5646: jmax=j;
5647: ijmax=i;
5648: }
5649: if (j <= jmin){
5650: jmin=j;
5651: ijmin=i;
5652: }
5653: sum=sum+j;
5654: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5655: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5656: }
5657: }
5658: else{
5659: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5660: /* 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 5661:
1.227 brouard 5662: k=k+1;
5663: if (j >= jmax) {
5664: jmax=j;
5665: ijmax=i;
5666: }
5667: else if (j <= jmin){
5668: jmin=j;
5669: ijmin=i;
5670: }
5671: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5672: /*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]);*/
5673: if(j<0){
5674: nberr++;
5675: 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]);
5676: 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]);
5677: }
5678: sum=sum+j;
5679: }
5680: jk= j/stepm;
5681: jl= j -jk*stepm;
5682: ju= j -(jk+1)*stepm;
5683: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5684: if(jl==0){
5685: dh[mi][i]=jk;
5686: bh[mi][i]=0;
5687: }else{ /* We want a negative bias in order to only have interpolation ie
5688: * to avoid the price of an extra matrix product in likelihood */
5689: dh[mi][i]=jk+1;
5690: bh[mi][i]=ju;
5691: }
5692: }else{
5693: if(jl <= -ju){
5694: dh[mi][i]=jk;
5695: bh[mi][i]=jl; /* bias is positive if real duration
5696: * is higher than the multiple of stepm and negative otherwise.
5697: */
5698: }
5699: else{
5700: dh[mi][i]=jk+1;
5701: bh[mi][i]=ju;
5702: }
5703: if(dh[mi][i]==0){
5704: dh[mi][i]=1; /* At least one step */
5705: bh[mi][i]=ju; /* At least one step */
5706: /* 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);*/
5707: }
5708: } /* end if mle */
1.126 brouard 5709: }
5710: } /* end wave */
5711: }
5712: jmean=sum/k;
5713: 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 5714: 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 5715: }
1.126 brouard 5716:
5717: /*********** Tricode ****************************/
1.220 brouard 5718: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5719: {
5720: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5721: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5722: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5723: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5724: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5725: */
1.130 brouard 5726:
1.242 brouard 5727: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5728: int modmaxcovj=0; /* Modality max of covariates j */
5729: int cptcode=0; /* Modality max of covariates j */
5730: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5731:
5732:
1.242 brouard 5733: /* cptcoveff=0; */
5734: /* *cptcov=0; */
1.126 brouard 5735:
1.242 brouard 5736: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5737: for (k=1; k <= maxncov; k++)
5738: for(j=1; j<=2; j++)
5739: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5740:
1.242 brouard 5741: /* Loop on covariates without age and products and no quantitative variable */
5742: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5743: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5744: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5745: switch(Fixed[k]) {
5746: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5747: modmaxcovj=0;
5748: modmincovj=0;
1.242 brouard 5749: 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*/
5750: ij=(int)(covar[Tvar[k]][i]);
5751: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5752: * If product of Vn*Vm, still boolean *:
5753: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5754: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5755: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5756: modality of the nth covariate of individual i. */
5757: if (ij > modmaxcovj)
5758: modmaxcovj=ij;
5759: else if (ij < modmincovj)
5760: modmincovj=ij;
1.287 brouard 5761: if (ij <0 || ij >1 ){
1.311 brouard 5762: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5763: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5764: fflush(ficlog);
5765: exit(1);
1.287 brouard 5766: }
5767: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5768: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5769: exit(1);
5770: }else
5771: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5772: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5773: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5774: /* getting the maximum value of the modality of the covariate
5775: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5776: female ies 1, then modmaxcovj=1.
5777: */
5778: } /* end for loop on individuals i */
5779: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5780: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5781: cptcode=modmaxcovj;
5782: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5783: /*for (i=0; i<=cptcode; i++) {*/
5784: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5785: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5786: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5787: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5788: if( j != -1){
5789: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5790: covariate for which somebody answered excluding
5791: undefined. Usually 2: 0 and 1. */
5792: }
5793: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5794: covariate for which somebody answered including
5795: undefined. Usually 3: -1, 0 and 1. */
5796: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5797: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5798: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5799:
1.242 brouard 5800: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5801: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5802: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5803: /* modmincovj=3; modmaxcovj = 7; */
5804: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5805: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5806: /* defining two dummy variables: variables V1_1 and V1_2.*/
5807: /* nbcode[Tvar[j]][ij]=k; */
5808: /* nbcode[Tvar[j]][1]=0; */
5809: /* nbcode[Tvar[j]][2]=1; */
5810: /* nbcode[Tvar[j]][3]=2; */
5811: /* To be continued (not working yet). */
5812: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5813:
5814: /* 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*/
5815: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5816: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5817: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5818: /*, could be restored in the future */
5819: 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 5820: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5821: break;
5822: }
5823: ij++;
1.287 brouard 5824: 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 5825: cptcode = ij; /* New max modality for covar j */
5826: } /* end of loop on modality i=-1 to 1 or more */
5827: break;
5828: case 1: /* Testing on varying covariate, could be simple and
5829: * should look at waves or product of fixed *
5830: * varying. No time to test -1, assuming 0 and 1 only */
5831: ij=0;
5832: for(i=0; i<=1;i++){
5833: nbcode[Tvar[k]][++ij]=i;
5834: }
5835: break;
5836: default:
5837: break;
5838: } /* end switch */
5839: } /* end dummy test */
1.311 brouard 5840: if(Dummy[k]==1 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5841: 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*/
5842: if(isnan(covar[Tvar[k]][i])){
5843: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5844: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5845: fflush(ficlog);
5846: exit(1);
5847: }
5848: }
5849: }
1.287 brouard 5850: } /* 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 5851:
5852: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5853: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5854: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5855: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5856: 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 */
5857: 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 */
5858: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5859: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5860:
5861: ij=0;
5862: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5863: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5864: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5865: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5866: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5867: /* If product not in single variable we don't print results */
5868: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5869: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5870: 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*/
5871: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5872: 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 */
5873: if(Fixed[k]!=0)
5874: anyvaryingduminmodel=1;
5875: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5876: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5877: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5878: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5879: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5880: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5881: }
5882: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5883: /* ij--; */
5884: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5885: *cptcov=ij; /*Number of total real effective covariates: effective
5886: * because they can be excluded from the model and real
5887: * if in the model but excluded because missing values, but how to get k from ij?*/
5888: for(j=ij+1; j<= cptcovt; j++){
5889: Tvaraff[j]=0;
5890: Tmodelind[j]=0;
5891: }
5892: for(j=ntveff+1; j<= cptcovt; j++){
5893: TmodelInvind[j]=0;
5894: }
5895: /* To be sorted */
5896: ;
5897: }
1.126 brouard 5898:
1.145 brouard 5899:
1.126 brouard 5900: /*********** Health Expectancies ****************/
5901:
1.235 brouard 5902: 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 5903:
5904: {
5905: /* Health expectancies, no variances */
1.164 brouard 5906: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5907: int nhstepma, nstepma; /* Decreasing with age */
5908: double age, agelim, hf;
5909: double ***p3mat;
5910: double eip;
5911:
1.238 brouard 5912: /* pstamp(ficreseij); */
1.126 brouard 5913: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5914: fprintf(ficreseij,"# Age");
5915: for(i=1; i<=nlstate;i++){
5916: for(j=1; j<=nlstate;j++){
5917: fprintf(ficreseij," e%1d%1d ",i,j);
5918: }
5919: fprintf(ficreseij," e%1d. ",i);
5920: }
5921: fprintf(ficreseij,"\n");
5922:
5923:
5924: if(estepm < stepm){
5925: printf ("Problem %d lower than %d\n",estepm, stepm);
5926: }
5927: else hstepm=estepm;
5928: /* We compute the life expectancy from trapezoids spaced every estepm months
5929: * This is mainly to measure the difference between two models: for example
5930: * if stepm=24 months pijx are given only every 2 years and by summing them
5931: * we are calculating an estimate of the Life Expectancy assuming a linear
5932: * progression in between and thus overestimating or underestimating according
5933: * to the curvature of the survival function. If, for the same date, we
5934: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5935: * to compare the new estimate of Life expectancy with the same linear
5936: * hypothesis. A more precise result, taking into account a more precise
5937: * curvature will be obtained if estepm is as small as stepm. */
5938:
5939: /* For example we decided to compute the life expectancy with the smallest unit */
5940: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5941: nhstepm is the number of hstepm from age to agelim
5942: nstepm is the number of stepm from age to agelin.
1.270 brouard 5943: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5944: and note for a fixed period like estepm months */
5945: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5946: survival function given by stepm (the optimization length). Unfortunately it
5947: means that if the survival funtion is printed only each two years of age and if
5948: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5949: results. So we changed our mind and took the option of the best precision.
5950: */
5951: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5952:
5953: agelim=AGESUP;
5954: /* If stepm=6 months */
5955: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5956: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5957:
5958: /* nhstepm age range expressed in number of stepm */
5959: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5960: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5961: /* if (stepm >= YEARM) hstepm=1;*/
5962: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5963: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5964:
5965: for (age=bage; age<=fage; age ++){
5966: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5967: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5968: /* if (stepm >= YEARM) hstepm=1;*/
5969: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5970:
5971: /* If stepm=6 months */
5972: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5973: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5974:
1.235 brouard 5975: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5976:
5977: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5978:
5979: printf("%d|",(int)age);fflush(stdout);
5980: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5981:
5982: /* Computing expectancies */
5983: for(i=1; i<=nlstate;i++)
5984: for(j=1; j<=nlstate;j++)
5985: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5986: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5987:
5988: /* 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]);*/
5989:
5990: }
5991:
5992: fprintf(ficreseij,"%3.0f",age );
5993: for(i=1; i<=nlstate;i++){
5994: eip=0;
5995: for(j=1; j<=nlstate;j++){
5996: eip +=eij[i][j][(int)age];
5997: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5998: }
5999: fprintf(ficreseij,"%9.4f", eip );
6000: }
6001: fprintf(ficreseij,"\n");
6002:
6003: }
6004: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6005: printf("\n");
6006: fprintf(ficlog,"\n");
6007:
6008: }
6009:
1.235 brouard 6010: 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 6011:
6012: {
6013: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6014: to initial status i, ei. .
1.126 brouard 6015: */
6016: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6017: int nhstepma, nstepma; /* Decreasing with age */
6018: double age, agelim, hf;
6019: double ***p3matp, ***p3matm, ***varhe;
6020: double **dnewm,**doldm;
6021: double *xp, *xm;
6022: double **gp, **gm;
6023: double ***gradg, ***trgradg;
6024: int theta;
6025:
6026: double eip, vip;
6027:
6028: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6029: xp=vector(1,npar);
6030: xm=vector(1,npar);
6031: dnewm=matrix(1,nlstate*nlstate,1,npar);
6032: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6033:
6034: pstamp(ficresstdeij);
6035: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6036: fprintf(ficresstdeij,"# Age");
6037: for(i=1; i<=nlstate;i++){
6038: for(j=1; j<=nlstate;j++)
6039: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6040: fprintf(ficresstdeij," e%1d. ",i);
6041: }
6042: fprintf(ficresstdeij,"\n");
6043:
6044: pstamp(ficrescveij);
6045: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6046: fprintf(ficrescveij,"# Age");
6047: for(i=1; i<=nlstate;i++)
6048: for(j=1; j<=nlstate;j++){
6049: cptj= (j-1)*nlstate+i;
6050: for(i2=1; i2<=nlstate;i2++)
6051: for(j2=1; j2<=nlstate;j2++){
6052: cptj2= (j2-1)*nlstate+i2;
6053: if(cptj2 <= cptj)
6054: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6055: }
6056: }
6057: fprintf(ficrescveij,"\n");
6058:
6059: if(estepm < stepm){
6060: printf ("Problem %d lower than %d\n",estepm, stepm);
6061: }
6062: else hstepm=estepm;
6063: /* We compute the life expectancy from trapezoids spaced every estepm months
6064: * This is mainly to measure the difference between two models: for example
6065: * if stepm=24 months pijx are given only every 2 years and by summing them
6066: * we are calculating an estimate of the Life Expectancy assuming a linear
6067: * progression in between and thus overestimating or underestimating according
6068: * to the curvature of the survival function. If, for the same date, we
6069: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6070: * to compare the new estimate of Life expectancy with the same linear
6071: * hypothesis. A more precise result, taking into account a more precise
6072: * curvature will be obtained if estepm is as small as stepm. */
6073:
6074: /* For example we decided to compute the life expectancy with the smallest unit */
6075: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6076: nhstepm is the number of hstepm from age to agelim
6077: nstepm is the number of stepm from age to agelin.
6078: Look at hpijx to understand the reason of that which relies in memory size
6079: and note for a fixed period like estepm months */
6080: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6081: survival function given by stepm (the optimization length). Unfortunately it
6082: means that if the survival funtion is printed only each two years of age and if
6083: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6084: results. So we changed our mind and took the option of the best precision.
6085: */
6086: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6087:
6088: /* If stepm=6 months */
6089: /* nhstepm age range expressed in number of stepm */
6090: agelim=AGESUP;
6091: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6092: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6093: /* if (stepm >= YEARM) hstepm=1;*/
6094: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6095:
6096: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6097: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6098: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6099: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6100: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6101: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6102:
6103: for (age=bage; age<=fage; age ++){
6104: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6105: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6106: /* if (stepm >= YEARM) hstepm=1;*/
6107: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6108:
1.126 brouard 6109: /* If stepm=6 months */
6110: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6111: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6112:
6113: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6114:
1.126 brouard 6115: /* Computing Variances of health expectancies */
6116: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6117: decrease memory allocation */
6118: for(theta=1; theta <=npar; theta++){
6119: for(i=1; i<=npar; i++){
1.222 brouard 6120: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6121: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6122: }
1.235 brouard 6123: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6124: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6125:
1.126 brouard 6126: for(j=1; j<= nlstate; j++){
1.222 brouard 6127: for(i=1; i<=nlstate; i++){
6128: for(h=0; h<=nhstepm-1; h++){
6129: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6130: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6131: }
6132: }
1.126 brouard 6133: }
1.218 brouard 6134:
1.126 brouard 6135: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6136: for(h=0; h<=nhstepm-1; h++){
6137: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6138: }
1.126 brouard 6139: }/* End theta */
6140:
6141:
6142: for(h=0; h<=nhstepm-1; h++)
6143: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6144: for(theta=1; theta <=npar; theta++)
6145: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6146:
1.218 brouard 6147:
1.222 brouard 6148: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6149: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6150: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6151:
1.222 brouard 6152: printf("%d|",(int)age);fflush(stdout);
6153: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6154: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6155: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6156: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6157: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6158: for(ij=1;ij<=nlstate*nlstate;ij++)
6159: for(ji=1;ji<=nlstate*nlstate;ji++)
6160: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6161: }
6162: }
1.320 brouard 6163: /* if((int)age ==50){ */
6164: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6165: /* } */
1.126 brouard 6166: /* Computing expectancies */
1.235 brouard 6167: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6168: for(i=1; i<=nlstate;i++)
6169: for(j=1; j<=nlstate;j++)
1.222 brouard 6170: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6171: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6172:
1.222 brouard 6173: /* 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 6174:
1.222 brouard 6175: }
1.269 brouard 6176:
6177: /* Standard deviation of expectancies ij */
1.126 brouard 6178: fprintf(ficresstdeij,"%3.0f",age );
6179: for(i=1; i<=nlstate;i++){
6180: eip=0.;
6181: vip=0.;
6182: for(j=1; j<=nlstate;j++){
1.222 brouard 6183: eip += eij[i][j][(int)age];
6184: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6185: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6186: 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 6187: }
6188: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6189: }
6190: fprintf(ficresstdeij,"\n");
1.218 brouard 6191:
1.269 brouard 6192: /* Variance of expectancies ij */
1.126 brouard 6193: fprintf(ficrescveij,"%3.0f",age );
6194: for(i=1; i<=nlstate;i++)
6195: for(j=1; j<=nlstate;j++){
1.222 brouard 6196: cptj= (j-1)*nlstate+i;
6197: for(i2=1; i2<=nlstate;i2++)
6198: for(j2=1; j2<=nlstate;j2++){
6199: cptj2= (j2-1)*nlstate+i2;
6200: if(cptj2 <= cptj)
6201: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6202: }
1.126 brouard 6203: }
6204: fprintf(ficrescveij,"\n");
1.218 brouard 6205:
1.126 brouard 6206: }
6207: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6208: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6209: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6210: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6211: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6212: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6213: printf("\n");
6214: fprintf(ficlog,"\n");
1.218 brouard 6215:
1.126 brouard 6216: free_vector(xm,1,npar);
6217: free_vector(xp,1,npar);
6218: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6219: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6220: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6221: }
1.218 brouard 6222:
1.126 brouard 6223: /************ Variance ******************/
1.235 brouard 6224: 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 6225: {
1.279 brouard 6226: /** Variance of health expectancies
6227: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6228: * double **newm;
6229: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6230: */
1.218 brouard 6231:
6232: /* int movingaverage(); */
6233: double **dnewm,**doldm;
6234: double **dnewmp,**doldmp;
6235: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6236: int first=0;
1.218 brouard 6237: int k;
6238: double *xp;
1.279 brouard 6239: double **gp, **gm; /**< for var eij */
6240: double ***gradg, ***trgradg; /**< for var eij */
6241: double **gradgp, **trgradgp; /**< for var p point j */
6242: double *gpp, *gmp; /**< for var p point j */
6243: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6244: double ***p3mat;
6245: double age,agelim, hf;
6246: /* double ***mobaverage; */
6247: int theta;
6248: char digit[4];
6249: char digitp[25];
6250:
6251: char fileresprobmorprev[FILENAMELENGTH];
6252:
6253: if(popbased==1){
6254: if(mobilav!=0)
6255: strcpy(digitp,"-POPULBASED-MOBILAV_");
6256: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6257: }
6258: else
6259: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6260:
1.218 brouard 6261: /* if (mobilav!=0) { */
6262: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6263: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6264: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6265: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6266: /* } */
6267: /* } */
6268:
6269: strcpy(fileresprobmorprev,"PRMORPREV-");
6270: sprintf(digit,"%-d",ij);
6271: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6272: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6273: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6274: strcat(fileresprobmorprev,fileresu);
6275: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6276: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6277: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6278: }
6279: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6280: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6281: pstamp(ficresprobmorprev);
6282: 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 6283: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
6284: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
6285: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
6286: }
6287: for(j=1;j<=cptcoveff;j++)
6288: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
6289: fprintf(ficresprobmorprev,"\n");
6290:
1.218 brouard 6291: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6292: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6293: fprintf(ficresprobmorprev," p.%-d SE",j);
6294: for(i=1; i<=nlstate;i++)
6295: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6296: }
6297: fprintf(ficresprobmorprev,"\n");
6298:
6299: fprintf(ficgp,"\n# Routine varevsij");
6300: fprintf(ficgp,"\nunset title \n");
6301: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6302: 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");
6303: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6304:
1.218 brouard 6305: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6306: pstamp(ficresvij);
6307: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6308: if(popbased==1)
6309: 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);
6310: else
6311: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6312: fprintf(ficresvij,"# Age");
6313: for(i=1; i<=nlstate;i++)
6314: for(j=1; j<=nlstate;j++)
6315: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6316: fprintf(ficresvij,"\n");
6317:
6318: xp=vector(1,npar);
6319: dnewm=matrix(1,nlstate,1,npar);
6320: doldm=matrix(1,nlstate,1,nlstate);
6321: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6322: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6323:
6324: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6325: gpp=vector(nlstate+1,nlstate+ndeath);
6326: gmp=vector(nlstate+1,nlstate+ndeath);
6327: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6328:
1.218 brouard 6329: if(estepm < stepm){
6330: printf ("Problem %d lower than %d\n",estepm, stepm);
6331: }
6332: else hstepm=estepm;
6333: /* For example we decided to compute the life expectancy with the smallest unit */
6334: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6335: nhstepm is the number of hstepm from age to agelim
6336: nstepm is the number of stepm from age to agelim.
6337: Look at function hpijx to understand why because of memory size limitations,
6338: we decided (b) to get a life expectancy respecting the most precise curvature of the
6339: survival function given by stepm (the optimization length). Unfortunately it
6340: means that if the survival funtion is printed every two years of age and if
6341: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6342: results. So we changed our mind and took the option of the best precision.
6343: */
6344: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6345: agelim = AGESUP;
6346: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6347: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6348: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6349: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6350: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6351: gp=matrix(0,nhstepm,1,nlstate);
6352: gm=matrix(0,nhstepm,1,nlstate);
6353:
6354:
6355: for(theta=1; theta <=npar; theta++){
6356: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6357: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6358: }
1.279 brouard 6359: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6360: * returns into prlim .
1.288 brouard 6361: */
1.242 brouard 6362: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6363:
6364: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6365: if (popbased==1) {
6366: if(mobilav ==0){
6367: for(i=1; i<=nlstate;i++)
6368: prlim[i][i]=probs[(int)age][i][ij];
6369: }else{ /* mobilav */
6370: for(i=1; i<=nlstate;i++)
6371: prlim[i][i]=mobaverage[(int)age][i][ij];
6372: }
6373: }
1.295 brouard 6374: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6375: */
6376: 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 6377: /**< 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 6378: * at horizon h in state j including mortality.
6379: */
1.218 brouard 6380: for(j=1; j<= nlstate; j++){
6381: for(h=0; h<=nhstepm; h++){
6382: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6383: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6384: }
6385: }
1.279 brouard 6386: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6387: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6388: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6389: */
6390: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6391: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6392: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6393: }
6394:
6395: /* Again with minus shift */
1.218 brouard 6396:
6397: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6398: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6399:
1.242 brouard 6400: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6401:
6402: if (popbased==1) {
6403: if(mobilav ==0){
6404: for(i=1; i<=nlstate;i++)
6405: prlim[i][i]=probs[(int)age][i][ij];
6406: }else{ /* mobilav */
6407: for(i=1; i<=nlstate;i++)
6408: prlim[i][i]=mobaverage[(int)age][i][ij];
6409: }
6410: }
6411:
1.235 brouard 6412: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6413:
6414: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6415: for(h=0; h<=nhstepm; h++){
6416: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6417: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6418: }
6419: }
6420: /* This for computing probability of death (h=1 means
6421: computed over hstepm matrices product = hstepm*stepm months)
6422: as a weighted average of prlim.
6423: */
6424: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6425: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6426: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6427: }
1.279 brouard 6428: /* end shifting computations */
6429:
6430: /**< Computing gradient matrix at horizon h
6431: */
1.218 brouard 6432: for(j=1; j<= nlstate; j++) /* vareij */
6433: for(h=0; h<=nhstepm; h++){
6434: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6435: }
1.279 brouard 6436: /**< Gradient of overall mortality p.3 (or p.j)
6437: */
6438: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6439: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6440: }
6441:
6442: } /* End theta */
1.279 brouard 6443:
6444: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6445: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6446:
6447: for(h=0; h<=nhstepm; h++) /* veij */
6448: for(j=1; j<=nlstate;j++)
6449: for(theta=1; theta <=npar; theta++)
6450: trgradg[h][j][theta]=gradg[h][theta][j];
6451:
6452: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6453: for(theta=1; theta <=npar; theta++)
6454: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6455: /**< as well as its transposed matrix
6456: */
1.218 brouard 6457:
6458: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6459: for(i=1;i<=nlstate;i++)
6460: for(j=1;j<=nlstate;j++)
6461: vareij[i][j][(int)age] =0.;
1.279 brouard 6462:
6463: /* Computing trgradg by matcov by gradg at age and summing over h
6464: * and k (nhstepm) formula 15 of article
6465: * Lievre-Brouard-Heathcote
6466: */
6467:
1.218 brouard 6468: for(h=0;h<=nhstepm;h++){
6469: for(k=0;k<=nhstepm;k++){
6470: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6471: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6472: for(i=1;i<=nlstate;i++)
6473: for(j=1;j<=nlstate;j++)
6474: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6475: }
6476: }
6477:
1.279 brouard 6478: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6479: * p.j overall mortality formula 49 but computed directly because
6480: * we compute the grad (wix pijx) instead of grad (pijx),even if
6481: * wix is independent of theta.
6482: */
1.218 brouard 6483: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6484: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6485: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6486: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6487: varppt[j][i]=doldmp[j][i];
6488: /* end ppptj */
6489: /* x centered again */
6490:
1.242 brouard 6491: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6492:
6493: if (popbased==1) {
6494: if(mobilav ==0){
6495: for(i=1; i<=nlstate;i++)
6496: prlim[i][i]=probs[(int)age][i][ij];
6497: }else{ /* mobilav */
6498: for(i=1; i<=nlstate;i++)
6499: prlim[i][i]=mobaverage[(int)age][i][ij];
6500: }
6501: }
6502:
6503: /* This for computing probability of death (h=1 means
6504: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6505: as a weighted average of prlim.
6506: */
1.235 brouard 6507: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6508: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6509: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6510: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6511: }
6512: /* end probability of death */
6513:
6514: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6515: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6516: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6517: for(i=1; i<=nlstate;i++){
6518: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6519: }
6520: }
6521: fprintf(ficresprobmorprev,"\n");
6522:
6523: fprintf(ficresvij,"%.0f ",age );
6524: for(i=1; i<=nlstate;i++)
6525: for(j=1; j<=nlstate;j++){
6526: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6527: }
6528: fprintf(ficresvij,"\n");
6529: free_matrix(gp,0,nhstepm,1,nlstate);
6530: free_matrix(gm,0,nhstepm,1,nlstate);
6531: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6532: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6533: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6534: } /* End age */
6535: free_vector(gpp,nlstate+1,nlstate+ndeath);
6536: free_vector(gmp,nlstate+1,nlstate+ndeath);
6537: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6538: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6539: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6540: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6541: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6542: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6543: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6544: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6545: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6546: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6547: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6548: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6549: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6550: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6551: 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);
6552: /* 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 6553: */
1.218 brouard 6554: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6555: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6556:
1.218 brouard 6557: free_vector(xp,1,npar);
6558: free_matrix(doldm,1,nlstate,1,nlstate);
6559: free_matrix(dnewm,1,nlstate,1,npar);
6560: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6561: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6562: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6563: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6564: fclose(ficresprobmorprev);
6565: fflush(ficgp);
6566: fflush(fichtm);
6567: } /* end varevsij */
1.126 brouard 6568:
6569: /************ Variance of prevlim ******************/
1.269 brouard 6570: 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 6571: {
1.205 brouard 6572: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6573: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6574:
1.268 brouard 6575: double **dnewmpar,**doldm;
1.126 brouard 6576: int i, j, nhstepm, hstepm;
6577: double *xp;
6578: double *gp, *gm;
6579: double **gradg, **trgradg;
1.208 brouard 6580: double **mgm, **mgp;
1.126 brouard 6581: double age,agelim;
6582: int theta;
6583:
6584: pstamp(ficresvpl);
1.288 brouard 6585: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6586: fprintf(ficresvpl,"# Age ");
6587: if(nresult >=1)
6588: fprintf(ficresvpl," Result# ");
1.126 brouard 6589: for(i=1; i<=nlstate;i++)
6590: fprintf(ficresvpl," %1d-%1d",i,i);
6591: fprintf(ficresvpl,"\n");
6592:
6593: xp=vector(1,npar);
1.268 brouard 6594: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6595: doldm=matrix(1,nlstate,1,nlstate);
6596:
6597: hstepm=1*YEARM; /* Every year of age */
6598: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6599: agelim = AGESUP;
6600: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6601: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6602: if (stepm >= YEARM) hstepm=1;
6603: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6604: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6605: mgp=matrix(1,npar,1,nlstate);
6606: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6607: gp=vector(1,nlstate);
6608: gm=vector(1,nlstate);
6609:
6610: for(theta=1; theta <=npar; theta++){
6611: for(i=1; i<=npar; i++){ /* Computes gradient */
6612: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6613: }
1.288 brouard 6614: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6615: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6616: /* else */
6617: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6618: for(i=1;i<=nlstate;i++){
1.126 brouard 6619: gp[i] = prlim[i][i];
1.208 brouard 6620: mgp[theta][i] = prlim[i][i];
6621: }
1.126 brouard 6622: for(i=1; i<=npar; i++) /* Computes gradient */
6623: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6624: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6625: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6626: /* else */
6627: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6628: for(i=1;i<=nlstate;i++){
1.126 brouard 6629: gm[i] = prlim[i][i];
1.208 brouard 6630: mgm[theta][i] = prlim[i][i];
6631: }
1.126 brouard 6632: for(i=1;i<=nlstate;i++)
6633: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6634: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6635: } /* End theta */
6636:
6637: trgradg =matrix(1,nlstate,1,npar);
6638:
6639: for(j=1; j<=nlstate;j++)
6640: for(theta=1; theta <=npar; theta++)
6641: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6642: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6643: /* printf("\nmgm mgp %d ",(int)age); */
6644: /* for(j=1; j<=nlstate;j++){ */
6645: /* printf(" %d ",j); */
6646: /* for(theta=1; theta <=npar; theta++) */
6647: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6648: /* printf("\n "); */
6649: /* } */
6650: /* } */
6651: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6652: /* printf("\n gradg %d ",(int)age); */
6653: /* for(j=1; j<=nlstate;j++){ */
6654: /* printf("%d ",j); */
6655: /* for(theta=1; theta <=npar; theta++) */
6656: /* printf("%d %lf ",theta,gradg[theta][j]); */
6657: /* printf("\n "); */
6658: /* } */
6659: /* } */
1.126 brouard 6660:
6661: for(i=1;i<=nlstate;i++)
6662: varpl[i][(int)age] =0.;
1.209 brouard 6663: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6664: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6665: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6666: }else{
1.268 brouard 6667: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6668: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6669: }
1.126 brouard 6670: for(i=1;i<=nlstate;i++)
6671: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6672:
6673: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6674: if(nresult >=1)
6675: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6676: for(i=1; i<=nlstate;i++){
1.126 brouard 6677: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6678: /* for(j=1;j<=nlstate;j++) */
6679: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6680: }
1.126 brouard 6681: fprintf(ficresvpl,"\n");
6682: free_vector(gp,1,nlstate);
6683: free_vector(gm,1,nlstate);
1.208 brouard 6684: free_matrix(mgm,1,npar,1,nlstate);
6685: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6686: free_matrix(gradg,1,npar,1,nlstate);
6687: free_matrix(trgradg,1,nlstate,1,npar);
6688: } /* End age */
6689:
6690: free_vector(xp,1,npar);
6691: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6692: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6693:
6694: }
6695:
6696:
6697: /************ Variance of backprevalence limit ******************/
1.269 brouard 6698: 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 6699: {
6700: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6701: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6702:
6703: double **dnewmpar,**doldm;
6704: int i, j, nhstepm, hstepm;
6705: double *xp;
6706: double *gp, *gm;
6707: double **gradg, **trgradg;
6708: double **mgm, **mgp;
6709: double age,agelim;
6710: int theta;
6711:
6712: pstamp(ficresvbl);
6713: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6714: fprintf(ficresvbl,"# Age ");
6715: if(nresult >=1)
6716: fprintf(ficresvbl," Result# ");
6717: for(i=1; i<=nlstate;i++)
6718: fprintf(ficresvbl," %1d-%1d",i,i);
6719: fprintf(ficresvbl,"\n");
6720:
6721: xp=vector(1,npar);
6722: dnewmpar=matrix(1,nlstate,1,npar);
6723: doldm=matrix(1,nlstate,1,nlstate);
6724:
6725: hstepm=1*YEARM; /* Every year of age */
6726: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6727: agelim = AGEINF;
6728: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6729: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6730: if (stepm >= YEARM) hstepm=1;
6731: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6732: gradg=matrix(1,npar,1,nlstate);
6733: mgp=matrix(1,npar,1,nlstate);
6734: mgm=matrix(1,npar,1,nlstate);
6735: gp=vector(1,nlstate);
6736: gm=vector(1,nlstate);
6737:
6738: for(theta=1; theta <=npar; theta++){
6739: for(i=1; i<=npar; i++){ /* Computes gradient */
6740: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6741: }
6742: if(mobilavproj > 0 )
6743: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6744: else
6745: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6746: for(i=1;i<=nlstate;i++){
6747: gp[i] = bprlim[i][i];
6748: mgp[theta][i] = bprlim[i][i];
6749: }
6750: for(i=1; i<=npar; i++) /* Computes gradient */
6751: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6752: if(mobilavproj > 0 )
6753: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6754: else
6755: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6756: for(i=1;i<=nlstate;i++){
6757: gm[i] = bprlim[i][i];
6758: mgm[theta][i] = bprlim[i][i];
6759: }
6760: for(i=1;i<=nlstate;i++)
6761: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6762: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6763: } /* End theta */
6764:
6765: trgradg =matrix(1,nlstate,1,npar);
6766:
6767: for(j=1; j<=nlstate;j++)
6768: for(theta=1; theta <=npar; theta++)
6769: trgradg[j][theta]=gradg[theta][j];
6770: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6771: /* printf("\nmgm mgp %d ",(int)age); */
6772: /* for(j=1; j<=nlstate;j++){ */
6773: /* printf(" %d ",j); */
6774: /* for(theta=1; theta <=npar; theta++) */
6775: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6776: /* printf("\n "); */
6777: /* } */
6778: /* } */
6779: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6780: /* printf("\n gradg %d ",(int)age); */
6781: /* for(j=1; j<=nlstate;j++){ */
6782: /* printf("%d ",j); */
6783: /* for(theta=1; theta <=npar; theta++) */
6784: /* printf("%d %lf ",theta,gradg[theta][j]); */
6785: /* printf("\n "); */
6786: /* } */
6787: /* } */
6788:
6789: for(i=1;i<=nlstate;i++)
6790: varbpl[i][(int)age] =0.;
6791: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6792: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6793: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6794: }else{
6795: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6796: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6797: }
6798: for(i=1;i<=nlstate;i++)
6799: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6800:
6801: fprintf(ficresvbl,"%.0f ",age );
6802: if(nresult >=1)
6803: fprintf(ficresvbl,"%d ",nres );
6804: for(i=1; i<=nlstate;i++)
6805: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6806: fprintf(ficresvbl,"\n");
6807: free_vector(gp,1,nlstate);
6808: free_vector(gm,1,nlstate);
6809: free_matrix(mgm,1,npar,1,nlstate);
6810: free_matrix(mgp,1,npar,1,nlstate);
6811: free_matrix(gradg,1,npar,1,nlstate);
6812: free_matrix(trgradg,1,nlstate,1,npar);
6813: } /* End age */
6814:
6815: free_vector(xp,1,npar);
6816: free_matrix(doldm,1,nlstate,1,npar);
6817: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6818:
6819: }
6820:
6821: /************ Variance of one-step probabilities ******************/
6822: 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 6823: {
6824: int i, j=0, k1, l1, tj;
6825: int k2, l2, j1, z1;
6826: int k=0, l;
6827: int first=1, first1, first2;
1.326 brouard 6828: int nres=0; /* New */
1.222 brouard 6829: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6830: double **dnewm,**doldm;
6831: double *xp;
6832: double *gp, *gm;
6833: double **gradg, **trgradg;
6834: double **mu;
6835: double age, cov[NCOVMAX+1];
6836: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6837: int theta;
6838: char fileresprob[FILENAMELENGTH];
6839: char fileresprobcov[FILENAMELENGTH];
6840: char fileresprobcor[FILENAMELENGTH];
6841: double ***varpij;
6842:
6843: strcpy(fileresprob,"PROB_");
6844: strcat(fileresprob,fileres);
6845: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6846: printf("Problem with resultfile: %s\n", fileresprob);
6847: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6848: }
6849: strcpy(fileresprobcov,"PROBCOV_");
6850: strcat(fileresprobcov,fileresu);
6851: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6852: printf("Problem with resultfile: %s\n", fileresprobcov);
6853: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6854: }
6855: strcpy(fileresprobcor,"PROBCOR_");
6856: strcat(fileresprobcor,fileresu);
6857: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6858: printf("Problem with resultfile: %s\n", fileresprobcor);
6859: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6860: }
6861: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6862: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6863: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6864: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6865: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6866: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6867: pstamp(ficresprob);
6868: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6869: fprintf(ficresprob,"# Age");
6870: pstamp(ficresprobcov);
6871: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6872: fprintf(ficresprobcov,"# Age");
6873: pstamp(ficresprobcor);
6874: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6875: fprintf(ficresprobcor,"# Age");
1.126 brouard 6876:
6877:
1.222 brouard 6878: for(i=1; i<=nlstate;i++)
6879: for(j=1; j<=(nlstate+ndeath);j++){
6880: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6881: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6882: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6883: }
6884: /* fprintf(ficresprob,"\n");
6885: fprintf(ficresprobcov,"\n");
6886: fprintf(ficresprobcor,"\n");
6887: */
6888: xp=vector(1,npar);
6889: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6890: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6891: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6892: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6893: first=1;
6894: fprintf(ficgp,"\n# Routine varprob");
6895: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6896: fprintf(fichtm,"\n");
6897:
1.288 brouard 6898: 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 6899: 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);
6900: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6901: and drawn. It helps understanding how is the covariance between two incidences.\
6902: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6903: 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 6904: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6905: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6906: standard deviations wide on each axis. <br>\
6907: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6908: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6909: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6910:
1.222 brouard 6911: cov[1]=1;
6912: /* tj=cptcoveff; */
1.225 brouard 6913: tj = (int) pow(2,cptcoveff);
1.222 brouard 6914: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6915: j1=0;
1.224 brouard 6916: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.326 brouard 6917: for(nres=1;nres <=1; nres++){ /* For each resultline */
6918: /* for(nres=1;nres <=nresult; nres++){ /\* For each resultline *\/ */
1.222 brouard 6919: if (cptcovn>0) {
6920: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6921: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6922: fprintf(ficresprob, "**********\n#\n");
6923: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6924: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6925: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6926:
1.222 brouard 6927: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6928: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6929: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6930:
6931:
1.222 brouard 6932: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.319 brouard 6933: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]); */
6934: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtmcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6935: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6936:
1.222 brouard 6937: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6938: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6939: fprintf(ficresprobcor, "**********\n#");
6940: if(invalidvarcomb[j1]){
6941: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6942: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6943: continue;
6944: }
6945: }
6946: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6947: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6948: gp=vector(1,(nlstate)*(nlstate+ndeath));
6949: gm=vector(1,(nlstate)*(nlstate+ndeath));
6950: for (age=bage; age<=fage; age ++){
6951: cov[2]=age;
6952: if(nagesqr==1)
6953: cov[3]= age*age;
1.326 brouard 6954: /* for (k=1; k<=cptcovn;k++) { */
6955: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; */
6956: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
6957: /* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates */
6958: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,k)];
1.222 brouard 6959: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6960: * 1 1 1 1 1
6961: * 2 2 1 1 1
6962: * 3 1 2 1 1
6963: */
6964: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6965: }
1.319 brouard 6966: /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 */
6967: /* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] */
6968: /*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
1.326 brouard 6969: for (k=1; k<=cptcovage;k++){ /* For product with age */
6970: if(Dummy[Tage[k]]==2){ /* dummy with age */
6971: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,k)]*cov[2];
6972: /* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
6973: } else if(Dummy[Tage[k]]==3){ /* quantitative with age */
1.327 brouard 6974: 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]);
6975: exit(1);
6976: /* 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 6977: /* cov[++k1]=Tqresult[nres][k]; */
6978: }
6979: /* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
6980: }
6981: for (k=1; k<=cptcovprod;k++){/* For product without age */
6982: if(Dummy[Tvard[k][1]==0]){
6983: if(Dummy[Tvard[k][2]==0]){
6984: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,k)] * nbcode[Tvard[k][2]][codtabm(j1,k)];
6985: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
6986: }else{ /* Should we use the mean of the quantitative variables? */
6987: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,k)] * Tqresult[nres][k];
6988: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; */
6989: }
6990: }else{
6991: if(Dummy[Tvard[k][2]==0]){
6992: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,k)] * Tqinvresult[nres][Tvard[k][1]];
6993: /* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; */
6994: }else{
6995: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
6996: /* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
6997: }
6998: }
6999: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
7000: }
7001: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7002: for(theta=1; theta <=npar; theta++){
7003: for(i=1; i<=npar; i++)
7004: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7005:
1.222 brouard 7006: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7007:
1.222 brouard 7008: k=0;
7009: for(i=1; i<= (nlstate); i++){
7010: for(j=1; j<=(nlstate+ndeath);j++){
7011: k=k+1;
7012: gp[k]=pmmij[i][j];
7013: }
7014: }
1.220 brouard 7015:
1.222 brouard 7016: for(i=1; i<=npar; i++)
7017: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7018:
1.222 brouard 7019: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7020: k=0;
7021: for(i=1; i<=(nlstate); i++){
7022: for(j=1; j<=(nlstate+ndeath);j++){
7023: k=k+1;
7024: gm[k]=pmmij[i][j];
7025: }
7026: }
1.220 brouard 7027:
1.222 brouard 7028: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7029: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7030: }
1.126 brouard 7031:
1.222 brouard 7032: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7033: for(theta=1; theta <=npar; theta++)
7034: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7035:
1.222 brouard 7036: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7037: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7038:
1.222 brouard 7039: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7040:
1.222 brouard 7041: k=0;
7042: for(i=1; i<=(nlstate); i++){
7043: for(j=1; j<=(nlstate+ndeath);j++){
7044: k=k+1;
7045: mu[k][(int) age]=pmmij[i][j];
7046: }
7047: }
7048: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7049: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7050: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7051:
1.222 brouard 7052: /*printf("\n%d ",(int)age);
7053: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7054: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7055: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7056: }*/
1.220 brouard 7057:
1.222 brouard 7058: fprintf(ficresprob,"\n%d ",(int)age);
7059: fprintf(ficresprobcov,"\n%d ",(int)age);
7060: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7061:
1.222 brouard 7062: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7063: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7064: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7065: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7066: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7067: }
7068: i=0;
7069: for (k=1; k<=(nlstate);k++){
7070: for (l=1; l<=(nlstate+ndeath);l++){
7071: i++;
7072: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7073: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7074: for (j=1; j<=i;j++){
7075: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7076: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7077: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7078: }
7079: }
7080: }/* end of loop for state */
7081: } /* end of loop for age */
7082: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7083: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7084: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7085: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7086:
7087: /* Confidence intervalle of pij */
7088: /*
7089: fprintf(ficgp,"\nunset parametric;unset label");
7090: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7091: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7092: 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);
7093: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7094: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7095: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7096: */
7097:
7098: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7099: first1=1;first2=2;
7100: for (k2=1; k2<=(nlstate);k2++){
7101: for (l2=1; l2<=(nlstate+ndeath);l2++){
7102: if(l2==k2) continue;
7103: j=(k2-1)*(nlstate+ndeath)+l2;
7104: for (k1=1; k1<=(nlstate);k1++){
7105: for (l1=1; l1<=(nlstate+ndeath);l1++){
7106: if(l1==k1) continue;
7107: i=(k1-1)*(nlstate+ndeath)+l1;
7108: if(i<=j) continue;
7109: for (age=bage; age<=fage; age ++){
7110: if ((int)age %5==0){
7111: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7112: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7113: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7114: mu1=mu[i][(int) age]/stepm*YEARM ;
7115: mu2=mu[j][(int) age]/stepm*YEARM;
7116: c12=cv12/sqrt(v1*v2);
7117: /* Computing eigen value of matrix of covariance */
7118: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7119: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7120: if ((lc2 <0) || (lc1 <0) ){
7121: if(first2==1){
7122: first1=0;
7123: 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);
7124: }
7125: 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);
7126: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7127: /* lc2=fabs(lc2); */
7128: }
1.220 brouard 7129:
1.222 brouard 7130: /* Eigen vectors */
1.280 brouard 7131: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7132: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7133: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7134: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7135: }else
7136: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7137: /*v21=sqrt(1.-v11*v11); *//* error */
7138: v21=(lc1-v1)/cv12*v11;
7139: v12=-v21;
7140: v22=v11;
7141: tnalp=v21/v11;
7142: if(first1==1){
7143: first1=0;
7144: 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);
7145: }
7146: 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);
7147: /*printf(fignu*/
7148: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7149: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7150: if(first==1){
7151: first=0;
7152: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7153: fprintf(ficgp,"\nset parametric;unset label");
7154: 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);
7155: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7156: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7157: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7158: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7159: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7160: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7161: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7162: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7163: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7164: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7165: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7166: 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 7167: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7168: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7169: }else{
7170: first=0;
7171: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7172: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7173: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7174: 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 7175: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7176: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7177: }/* if first */
7178: } /* age mod 5 */
7179: } /* end loop age */
7180: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7181: first=1;
7182: } /*l12 */
7183: } /* k12 */
7184: } /*l1 */
7185: }/* k1 */
1.326 brouard 7186: } /* loop on nres */
1.222 brouard 7187: } /* loop on combination of covariates j1 */
7188: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7189: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7190: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7191: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7192: free_vector(xp,1,npar);
7193: fclose(ficresprob);
7194: fclose(ficresprobcov);
7195: fclose(ficresprobcor);
7196: fflush(ficgp);
7197: fflush(fichtmcov);
7198: }
1.126 brouard 7199:
7200:
7201: /******************* Printing html file ***********/
1.201 brouard 7202: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7203: int lastpass, int stepm, int weightopt, char model[],\
7204: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7205: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7206: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7207: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7208: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7209: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7210: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7211: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7212: </ul>");
1.319 brouard 7213: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7214: /* </ul>", model); */
1.214 brouard 7215: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7216: 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",
7217: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
7218: 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 7219: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7220: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7221: fprintf(fichtm,"\
7222: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7223: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7224: fprintf(fichtm,"\
1.217 brouard 7225: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7226: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7227: fprintf(fichtm,"\
1.288 brouard 7228: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7229: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7230: fprintf(fichtm,"\
1.288 brouard 7231: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7232: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7233: fprintf(fichtm,"\
1.211 brouard 7234: - (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 7235: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7236: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7237: if(prevfcast==1){
7238: fprintf(fichtm,"\
7239: - Prevalence projections by age and states: \
1.201 brouard 7240: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7241: }
1.126 brouard 7242:
7243:
1.225 brouard 7244: m=pow(2,cptcoveff);
1.222 brouard 7245: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7246:
1.317 brouard 7247: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7248:
7249: jj1=0;
7250:
7251: fprintf(fichtm," \n<ul>");
7252: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7253: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7254: if(m != 1 && TKresult[nres]!= k1)
7255: continue;
7256: jj1++;
7257: if (cptcovn > 0) {
7258: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
7259: for (cpt=1; cpt<=cptcoveff;cpt++){
7260: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7261: }
7262: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7263: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7264: }
7265: fprintf(fichtm,"\">");
7266:
7267: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7268: fprintf(fichtm,"************ Results for covariates");
7269: for (cpt=1; cpt<=cptcoveff;cpt++){
7270: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7271: }
7272: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7273: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7274: }
7275: if(invalidvarcomb[k1]){
7276: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7277: continue;
7278: }
7279: fprintf(fichtm,"</a></li>");
7280: } /* cptcovn >0 */
7281: }
1.317 brouard 7282: fprintf(fichtm," \n</ul>");
1.264 brouard 7283:
1.222 brouard 7284: jj1=0;
1.237 brouard 7285:
7286: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7287: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7288: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7289: continue;
1.220 brouard 7290:
1.222 brouard 7291: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7292: jj1++;
7293: if (cptcovn > 0) {
1.264 brouard 7294: fprintf(fichtm,"\n<p><a name=\"rescov");
7295: for (cpt=1; cpt<=cptcoveff;cpt++){
7296: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7297: }
7298: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7299: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7300: }
7301: fprintf(fichtm,"\"</a>");
7302:
1.222 brouard 7303: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7304: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7305: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7306: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7307: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7308: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7309: }
1.237 brouard 7310: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7311: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7312: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7313: }
7314:
1.230 brouard 7315: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.321 brouard 7316: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7317: if(invalidvarcomb[k1]){
7318: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7319: printf("\nCombination (%d) ignored because no cases \n",k1);
7320: continue;
7321: }
7322: }
7323: /* aij, bij */
1.259 brouard 7324: 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 7325: <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 7326: /* Pij */
1.241 brouard 7327: 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> \
7328: <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 7329: /* Quasi-incidences */
7330: 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 7331: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7332: 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 7333: 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> \
7334: <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 7335: /* Survival functions (period) in state j */
7336: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7337: fprintf(fichtm,"<br>\n- Survival functions in state %d. And probability to be observed in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 7338: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7339: }
7340: /* State specific survival functions (period) */
7341: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7342: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7343: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 7344: <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7345: }
1.288 brouard 7346: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7347: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7348: 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> \
7349: <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7350: }
1.296 brouard 7351: if(prevbcast==1){
1.288 brouard 7352: /* Backward prevalence in each health state */
1.222 brouard 7353: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7354: 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 7355: <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 7356: }
1.217 brouard 7357: }
1.222 brouard 7358: if(prevfcast==1){
1.288 brouard 7359: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7360: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7361: 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);
7362: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7363: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7364: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7365: }
7366: }
1.296 brouard 7367: if(prevbcast==1){
1.268 brouard 7368: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7369: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7370: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7371: 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 \
7372: 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 7373: 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);
7374: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7375: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7376: }
7377: }
1.220 brouard 7378:
1.222 brouard 7379: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7380: 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);
7381: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7382: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7383: }
7384: /* } /\* end i1 *\/ */
7385: }/* End k1 */
7386: fprintf(fichtm,"</ul>");
1.126 brouard 7387:
1.222 brouard 7388: fprintf(fichtm,"\
1.126 brouard 7389: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7390: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7391: - 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 7392: But because parameters are usually highly correlated (a higher incidence of disability \
7393: and a higher incidence of recovery can give very close observed transition) it might \
7394: be very useful to look not only at linear confidence intervals estimated from the \
7395: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7396: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7397: covariance matrix of the one-step probabilities. \
7398: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7399:
1.222 brouard 7400: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7401: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7402: fprintf(fichtm,"\
1.126 brouard 7403: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7404: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7405:
1.222 brouard 7406: fprintf(fichtm,"\
1.126 brouard 7407: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7408: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7409: fprintf(fichtm,"\
1.126 brouard 7410: - 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): \
7411: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7412: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7413: fprintf(fichtm,"\
1.126 brouard 7414: - (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): \
7415: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7416: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7417: fprintf(fichtm,"\
1.288 brouard 7418: - 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 7419: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7420: fprintf(fichtm,"\
1.128 brouard 7421: - 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 7422: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7423: fprintf(fichtm,"\
1.288 brouard 7424: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7425: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7426:
7427: /* if(popforecast==1) fprintf(fichtm,"\n */
7428: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7429: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7430: /* <br>",fileres,fileres,fileres,fileres); */
7431: /* else */
7432: /* 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 7433: fflush(fichtm);
1.126 brouard 7434:
1.225 brouard 7435: m=pow(2,cptcoveff);
1.222 brouard 7436: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7437:
1.317 brouard 7438: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7439:
7440: jj1=0;
7441:
7442: fprintf(fichtm," \n<ul>");
7443: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7444: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7445: if(m != 1 && TKresult[nres]!= k1)
7446: continue;
7447: jj1++;
7448: if (cptcovn > 0) {
7449: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
7450: for (cpt=1; cpt<=cptcoveff;cpt++){
7451: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7452: }
7453: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7454: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7455: }
7456: fprintf(fichtm,"\">");
7457:
7458: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7459: fprintf(fichtm,"************ Results for covariates");
7460: for (cpt=1; cpt<=cptcoveff;cpt++){
7461: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7462: }
7463: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7464: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7465: }
7466: if(invalidvarcomb[k1]){
7467: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7468: continue;
7469: }
7470: fprintf(fichtm,"</a></li>");
7471: } /* cptcovn >0 */
7472: }
7473: fprintf(fichtm," \n</ul>");
7474:
1.222 brouard 7475: jj1=0;
1.237 brouard 7476:
1.241 brouard 7477: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7478: for(k1=1; k1<=m;k1++){
1.253 brouard 7479: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7480: continue;
1.222 brouard 7481: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7482: jj1++;
1.126 brouard 7483: if (cptcovn > 0) {
1.317 brouard 7484: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
7485: for (cpt=1; cpt<=cptcoveff;cpt++){
7486: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7487: }
7488: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7489: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7490: }
7491: fprintf(fichtm,"\"</a>");
7492:
1.126 brouard 7493: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.317 brouard 7494: for (cpt=1; cpt<=cptcoveff;cpt++){ /**< cptcoveff number of variables */
1.237 brouard 7495: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
1.317 brouard 7496: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
1.237 brouard 7497: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 7498: }
1.237 brouard 7499: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7500: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7501: }
7502:
1.321 brouard 7503: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 7504:
1.222 brouard 7505: if(invalidvarcomb[k1]){
7506: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7507: continue;
7508: }
1.126 brouard 7509: }
7510: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7511: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7512: 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);
7513: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7514: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7515: }
7516: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7517: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7518: true period expectancies (those weighted with period prevalences are also\
7519: drawn in addition to the population based expectancies computed using\
1.314 brouard 7520: 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);
7521: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7522: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7523: /* } /\* end i1 *\/ */
7524: }/* End k1 */
1.241 brouard 7525: }/* End nres */
1.222 brouard 7526: fprintf(fichtm,"</ul>");
7527: fflush(fichtm);
1.126 brouard 7528: }
7529:
7530: /******************* Gnuplot file **************/
1.296 brouard 7531: 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 7532:
7533: char dirfileres[132],optfileres[132];
1.264 brouard 7534: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7535: 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 7536: int lv=0, vlv=0, kl=0;
1.130 brouard 7537: int ng=0;
1.201 brouard 7538: int vpopbased;
1.223 brouard 7539: int ioffset; /* variable offset for columns */
1.270 brouard 7540: int iyearc=1; /* variable column for year of projection */
7541: int iagec=1; /* variable column for age of projection */
1.235 brouard 7542: int nres=0; /* Index of resultline */
1.266 brouard 7543: int istart=1; /* For starting graphs in projections */
1.219 brouard 7544:
1.126 brouard 7545: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7546: /* printf("Problem with file %s",optionfilegnuplot); */
7547: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7548: /* } */
7549:
7550: /*#ifdef windows */
7551: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7552: /*#endif */
1.225 brouard 7553: m=pow(2,cptcoveff);
1.126 brouard 7554:
1.274 brouard 7555: /* diagram of the model */
7556: fprintf(ficgp,"\n#Diagram of the model \n");
7557: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7558: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7559: 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);
7560:
7561: 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);
7562: fprintf(ficgp,"\n#show arrow\nunset label\n");
7563: 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);
7564: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7565: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7566: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7567: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7568:
1.202 brouard 7569: /* Contribution to likelihood */
7570: /* Plot the probability implied in the likelihood */
1.223 brouard 7571: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7572: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7573: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7574: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7575: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7576: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7577: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7578: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7579: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7580: 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));
7581: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7582: 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));
7583: for (i=1; i<= nlstate ; i ++) {
7584: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7585: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7586: 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);
7587: for (j=2; j<= nlstate+ndeath ; j ++) {
7588: 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);
7589: }
7590: fprintf(ficgp,";\nset out; unset ylabel;\n");
7591: }
7592: /* 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 */
7593: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7594: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7595: fprintf(ficgp,"\nset out;unset log\n");
7596: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7597:
1.126 brouard 7598: strcpy(dirfileres,optionfilefiname);
7599: strcpy(optfileres,"vpl");
1.223 brouard 7600: /* 1eme*/
1.238 brouard 7601: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7602: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7603: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7604: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7605: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7606: continue;
7607: /* We are interested in selected combination by the resultline */
1.246 brouard 7608: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7609: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7610: strcpy(gplotlabel,"(");
1.238 brouard 7611: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7612: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7613: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7614: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7615: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7616: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7617: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7618: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7619: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7620: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7621: }
7622: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7623: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7624: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7625: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7626: }
7627: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7628: /* printf("\n#\n"); */
1.238 brouard 7629: fprintf(ficgp,"\n#\n");
7630: if(invalidvarcomb[k1]){
1.260 brouard 7631: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7632: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7633: continue;
7634: }
1.235 brouard 7635:
1.241 brouard 7636: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7637: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7638: /* 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 7639: fprintf(ficgp,"set title \"Alive state %d %s model=%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 7640: 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);
7641: /* 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); */
7642: /* k1-1 error should be nres-1*/
1.238 brouard 7643: for (i=1; i<= nlstate ; i ++) {
7644: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7645: else fprintf(ficgp," %%*lf (%%*lf)");
7646: }
1.288 brouard 7647: 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 7648: for (i=1; i<= nlstate ; i ++) {
7649: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7650: else fprintf(ficgp," %%*lf (%%*lf)");
7651: }
1.260 brouard 7652: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7653: for (i=1; i<= nlstate ; i ++) {
7654: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7655: else fprintf(ficgp," %%*lf (%%*lf)");
7656: }
1.265 brouard 7657: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7658:
7659: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7660: if(cptcoveff ==0){
1.271 brouard 7661: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7662: }else{
7663: kl=0;
7664: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7665: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7666: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7667: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7668: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7669: vlv= nbcode[Tvaraff[k]][lv];
7670: kl++;
7671: /* 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 *\/ */
7672: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7673: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7674: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7675: if(k==cptcoveff){
7676: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7677: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7678: }else{
7679: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7680: kl++;
7681: }
7682: } /* end covariate */
7683: } /* end if no covariate */
7684:
1.296 brouard 7685: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7686: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7687: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7688: if(cptcoveff ==0){
1.245 brouard 7689: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7690: }else{
7691: kl=0;
7692: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7693: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7694: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7695: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7696: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7697: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7698: kl++;
1.238 brouard 7699: /* 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 *\/ */
7700: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7701: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7702: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7703: if(k==cptcoveff){
1.245 brouard 7704: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7705: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7706: }else{
7707: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7708: kl++;
7709: }
7710: } /* end covariate */
7711: } /* end if no covariate */
1.296 brouard 7712: if(prevbcast == 1){
1.268 brouard 7713: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7714: /* k1-1 error should be nres-1*/
7715: for (i=1; i<= nlstate ; i ++) {
7716: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7717: else fprintf(ficgp," %%*lf (%%*lf)");
7718: }
1.271 brouard 7719: 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 7720: for (i=1; i<= nlstate ; i ++) {
7721: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7722: else fprintf(ficgp," %%*lf (%%*lf)");
7723: }
1.276 brouard 7724: 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 7725: for (i=1; i<= nlstate ; i ++) {
7726: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7727: else fprintf(ficgp," %%*lf (%%*lf)");
7728: }
1.274 brouard 7729: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7730: } /* end if backprojcast */
1.296 brouard 7731: } /* end if prevbcast */
1.276 brouard 7732: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7733: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7734: } /* nres */
1.201 brouard 7735: } /* k1 */
7736: } /* cpt */
1.235 brouard 7737:
7738:
1.126 brouard 7739: /*2 eme*/
1.238 brouard 7740: for (k1=1; k1<= m ; k1 ++){
7741: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7742: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7743: continue;
7744: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7745: strcpy(gplotlabel,"(");
1.238 brouard 7746: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7747: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7748: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7749: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7750: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7751: vlv= nbcode[Tvaraff[k]][lv];
7752: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7753: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7754: }
1.237 brouard 7755: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7756: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7757: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7758: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7759: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7760: }
1.264 brouard 7761: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7762: fprintf(ficgp,"\n#\n");
1.223 brouard 7763: if(invalidvarcomb[k1]){
7764: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7765: continue;
7766: }
1.219 brouard 7767:
1.241 brouard 7768: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7769: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7770: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7771: if(vpopbased==0){
1.238 brouard 7772: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7773: }else
1.238 brouard 7774: fprintf(ficgp,"\nreplot ");
7775: for (i=1; i<= nlstate+1 ; i ++) {
7776: k=2*i;
1.261 brouard 7777: 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 7778: for (j=1; j<= nlstate+1 ; j ++) {
7779: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7780: else fprintf(ficgp," %%*lf (%%*lf)");
7781: }
7782: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7783: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7784: 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 7785: for (j=1; j<= nlstate+1 ; j ++) {
7786: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7787: else fprintf(ficgp," %%*lf (%%*lf)");
7788: }
7789: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7790: 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 7791: for (j=1; j<= nlstate+1 ; j ++) {
7792: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7793: else fprintf(ficgp," %%*lf (%%*lf)");
7794: }
7795: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7796: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7797: } /* state */
7798: } /* vpopbased */
1.264 brouard 7799: 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 7800: } /* end nres */
7801: } /* k1 end 2 eme*/
7802:
7803:
7804: /*3eme*/
7805: for (k1=1; k1<= m ; k1 ++){
7806: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7807: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7808: continue;
7809:
7810: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7811: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7812: strcpy(gplotlabel,"(");
1.238 brouard 7813: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7814: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7815: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7816: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7817: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7818: vlv= nbcode[Tvaraff[k]][lv];
7819: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7820: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7821: }
7822: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7823: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7824: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7825: }
1.264 brouard 7826: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7827: fprintf(ficgp,"\n#\n");
7828: if(invalidvarcomb[k1]){
7829: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7830: continue;
7831: }
7832:
7833: /* k=2+nlstate*(2*cpt-2); */
7834: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7835: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7836: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7837: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7838: 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 7839: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7840: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7841: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7842: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7843: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7844: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7845:
1.238 brouard 7846: */
7847: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7848: 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 7849: /* 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 7850:
1.238 brouard 7851: }
1.261 brouard 7852: 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 7853: }
1.264 brouard 7854: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7855: } /* end nres */
7856: } /* end kl 3eme */
1.126 brouard 7857:
1.223 brouard 7858: /* 4eme */
1.201 brouard 7859: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7860: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7861: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7862: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7863: continue;
1.238 brouard 7864: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7865: strcpy(gplotlabel,"(");
1.238 brouard 7866: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7867: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7868: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7869: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7870: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7871: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7872: vlv= nbcode[Tvaraff[k]][lv];
7873: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7874: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7875: }
7876: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7877: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7878: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7879: }
1.264 brouard 7880: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7881: fprintf(ficgp,"\n#\n");
7882: if(invalidvarcomb[k1]){
7883: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7884: continue;
1.223 brouard 7885: }
1.238 brouard 7886:
1.241 brouard 7887: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7888: 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 7889: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7890: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7891: k=3;
7892: for (i=1; i<= nlstate ; i ++){
7893: if(i==1){
7894: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7895: }else{
7896: fprintf(ficgp,", '' ");
7897: }
7898: l=(nlstate+ndeath)*(i-1)+1;
7899: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7900: for (j=2; j<= nlstate+ndeath ; j ++)
7901: fprintf(ficgp,"+$%d",k+l+j-1);
7902: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7903: } /* nlstate */
1.264 brouard 7904: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7905: } /* end cpt state*/
7906: } /* end nres */
7907: } /* end covariate k1 */
7908:
1.220 brouard 7909: /* 5eme */
1.201 brouard 7910: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7911: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7912: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7913: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7914: continue;
1.238 brouard 7915: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7916: strcpy(gplotlabel,"(");
1.238 brouard 7917: 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);
7918: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7919: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7920: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7921: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7922: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7923: vlv= nbcode[Tvaraff[k]][lv];
7924: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7925: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7926: }
7927: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7928: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7929: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7930: }
1.264 brouard 7931: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7932: fprintf(ficgp,"\n#\n");
7933: if(invalidvarcomb[k1]){
7934: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7935: continue;
7936: }
1.227 brouard 7937:
1.241 brouard 7938: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7939: 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 7940: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7941: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7942: k=3;
7943: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7944: if(j==1)
7945: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7946: else
7947: fprintf(ficgp,", '' ");
7948: l=(nlstate+ndeath)*(cpt-1) +j;
7949: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7950: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7951: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7952: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7953: } /* nlstate */
7954: fprintf(ficgp,", '' ");
7955: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7956: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7957: l=(nlstate+ndeath)*(cpt-1) +j;
7958: if(j < nlstate)
7959: fprintf(ficgp,"$%d +",k+l);
7960: else
7961: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7962: }
1.264 brouard 7963: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7964: } /* end cpt state*/
7965: } /* end covariate */
7966: } /* end nres */
1.227 brouard 7967:
1.220 brouard 7968: /* 6eme */
1.202 brouard 7969: /* CV preval stable (period) for each covariate */
1.237 brouard 7970: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7971: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7972: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7973: continue;
1.255 brouard 7974: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7975: strcpy(gplotlabel,"(");
1.288 brouard 7976: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7977: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7978: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7979: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7980: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7981: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7982: vlv= nbcode[Tvaraff[k]][lv];
7983: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7984: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7985: }
1.237 brouard 7986: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7987: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7988: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7989: }
1.264 brouard 7990: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7991: fprintf(ficgp,"\n#\n");
1.223 brouard 7992: if(invalidvarcomb[k1]){
1.227 brouard 7993: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7994: continue;
1.223 brouard 7995: }
1.227 brouard 7996:
1.241 brouard 7997: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7998: 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 7999: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8000: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8001: k=3; /* Offset */
1.255 brouard 8002: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8003: if(i==1)
8004: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8005: else
8006: fprintf(ficgp,", '' ");
1.255 brouard 8007: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8008: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8009: for (j=2; j<= nlstate ; j ++)
8010: fprintf(ficgp,"+$%d",k+l+j-1);
8011: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8012: } /* nlstate */
1.264 brouard 8013: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8014: } /* end cpt state*/
8015: } /* end covariate */
1.227 brouard 8016:
8017:
1.220 brouard 8018: /* 7eme */
1.296 brouard 8019: if(prevbcast == 1){
1.288 brouard 8020: /* CV backward prevalence for each covariate */
1.237 brouard 8021: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8022: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8023: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8024: continue;
1.268 brouard 8025: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8026: strcpy(gplotlabel,"(");
1.288 brouard 8027: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8028: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
8029: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
8030: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8031: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 8032: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 8033: vlv= nbcode[Tvaraff[k]][lv];
8034: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8035: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8036: }
1.237 brouard 8037: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8038: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8039: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8040: }
1.264 brouard 8041: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8042: fprintf(ficgp,"\n#\n");
8043: if(invalidvarcomb[k1]){
8044: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8045: continue;
8046: }
8047:
1.241 brouard 8048: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8049: 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 8050: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8051: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8052: k=3; /* Offset */
1.268 brouard 8053: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8054: if(i==1)
8055: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8056: else
8057: fprintf(ficgp,", '' ");
8058: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8059: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8060: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8061: /* 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 8062: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8063: /* for (j=2; j<= nlstate ; j ++) */
8064: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8065: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8066: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8067: } /* nlstate */
1.264 brouard 8068: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8069: } /* end cpt state*/
8070: } /* end covariate */
1.296 brouard 8071: } /* End if prevbcast */
1.218 brouard 8072:
1.223 brouard 8073: /* 8eme */
1.218 brouard 8074: if(prevfcast==1){
1.288 brouard 8075: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8076:
1.237 brouard 8077: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8078: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8079: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8080: continue;
1.211 brouard 8081: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8082: strcpy(gplotlabel,"(");
1.288 brouard 8083: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8084: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8085: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8086: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8087: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8088: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8089: vlv= nbcode[Tvaraff[k]][lv];
8090: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8091: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8092: }
1.237 brouard 8093: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8094: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8095: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8096: }
1.264 brouard 8097: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8098: fprintf(ficgp,"\n#\n");
8099: if(invalidvarcomb[k1]){
8100: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8101: continue;
8102: }
8103:
8104: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8105: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8106: 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 8107: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8108: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8109:
8110: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8111: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8112: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8113: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8114: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8115: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8116: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8117: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8118: if(i==istart){
1.227 brouard 8119: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8120: }else{
8121: fprintf(ficgp,",\\\n '' ");
8122: }
8123: if(cptcoveff ==0){ /* No covariate */
8124: ioffset=2; /* Age is in 2 */
8125: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8126: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8127: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8128: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8129: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8130: if(i==nlstate+1){
1.270 brouard 8131: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8132: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8133: fprintf(ficgp,",\\\n '' ");
8134: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8135: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8136: offyear, \
1.268 brouard 8137: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8138: }else
1.227 brouard 8139: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8140: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8141: }else{ /* more than 2 covariates */
1.270 brouard 8142: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8143: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8144: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8145: iyearc=ioffset-1;
8146: iagec=ioffset;
1.227 brouard 8147: fprintf(ficgp," u %d:(",ioffset);
8148: kl=0;
8149: strcpy(gplotcondition,"(");
8150: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8151: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8152: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8153: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8154: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8155: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8156: kl++;
8157: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8158: kl++;
8159: if(k <cptcoveff && cptcoveff>1)
8160: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8161: }
8162: strcpy(gplotcondition+strlen(gplotcondition),")");
8163: /* 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 *\/ */
8164: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8165: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8166: /* '' 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*/
8167: if(i==nlstate+1){
1.270 brouard 8168: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8169: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8170: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8171: fprintf(ficgp," u %d:(",iagec);
8172: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8173: iyearc, iagec, offyear, \
8174: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8175: /* '' 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 8176: }else{
8177: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8178: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8179: }
8180: } /* end if covariate */
8181: } /* nlstate */
1.264 brouard 8182: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8183: } /* end cpt state*/
8184: } /* end covariate */
8185: } /* End if prevfcast */
1.227 brouard 8186:
1.296 brouard 8187: if(prevbcast==1){
1.268 brouard 8188: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8189:
8190: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8191: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8192: if(m != 1 && TKresult[nres]!= k1)
8193: continue;
8194: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8195: strcpy(gplotlabel,"(");
8196: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
8197: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8198: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8199: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8200: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8201: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8202: vlv= nbcode[Tvaraff[k]][lv];
8203: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8204: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8205: }
8206: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8207: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8208: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8209: }
8210: strcpy(gplotlabel+strlen(gplotlabel),")");
8211: fprintf(ficgp,"\n#\n");
8212: if(invalidvarcomb[k1]){
8213: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8214: continue;
8215: }
8216:
8217: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8218: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8219: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8220: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8221: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8222:
8223: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8224: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8225: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8226: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8227: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8228: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8229: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8230: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8231: if(i==istart){
8232: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8233: }else{
8234: fprintf(ficgp,",\\\n '' ");
8235: }
8236: if(cptcoveff ==0){ /* No covariate */
8237: ioffset=2; /* Age is in 2 */
8238: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8239: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8240: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8241: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8242: fprintf(ficgp," u %d:(", ioffset);
8243: if(i==nlstate+1){
1.270 brouard 8244: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8245: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8246: fprintf(ficgp,",\\\n '' ");
8247: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8248: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8249: offbyear, \
8250: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8251: }else
8252: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8253: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8254: }else{ /* more than 2 covariates */
1.270 brouard 8255: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8256: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8257: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8258: iyearc=ioffset-1;
8259: iagec=ioffset;
1.268 brouard 8260: fprintf(ficgp," u %d:(",ioffset);
8261: kl=0;
8262: strcpy(gplotcondition,"(");
8263: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
8264: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
8265: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8266: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8267: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8268: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
8269: kl++;
8270: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8271: kl++;
8272: if(k <cptcoveff && cptcoveff>1)
8273: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8274: }
8275: strcpy(gplotcondition+strlen(gplotcondition),")");
8276: /* 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 *\/ */
8277: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8278: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8279: /* '' 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*/
8280: if(i==nlstate+1){
1.270 brouard 8281: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8282: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8283: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8284: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8285: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8286: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8287: iyearc,iagec,offbyear, \
8288: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8289: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8290: }else{
8291: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8292: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8293: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8294: }
8295: } /* end if covariate */
8296: } /* nlstate */
8297: fprintf(ficgp,"\nset out; unset label;\n");
8298: } /* end cpt state*/
8299: } /* end covariate */
1.296 brouard 8300: } /* End if prevbcast */
1.268 brouard 8301:
1.227 brouard 8302:
1.238 brouard 8303: /* 9eme writing MLE parameters */
8304: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8305: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8306: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8307: for(k=1; k <=(nlstate+ndeath); k++){
8308: if (k != i) {
1.227 brouard 8309: fprintf(ficgp,"# current state %d\n",k);
8310: for(j=1; j <=ncovmodel; j++){
8311: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8312: jk++;
8313: }
8314: fprintf(ficgp,"\n");
1.126 brouard 8315: }
8316: }
1.223 brouard 8317: }
1.187 brouard 8318: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8319:
1.145 brouard 8320: /*goto avoid;*/
1.238 brouard 8321: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8322: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8323: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8324: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8325: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8326: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8327: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8328: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8329: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8330: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8331: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8332: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8333: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8334: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8335: fprintf(ficgp,"#\n");
1.223 brouard 8336: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8337: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8338: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8339: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8340: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8341: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8342: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8343: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8344: continue;
1.264 brouard 8345: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8346: strcpy(gplotlabel,"(");
1.276 brouard 8347: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8348: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8349: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8350: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8351: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8352: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8353: vlv= nbcode[Tvaraff[k]][lv];
8354: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8355: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8356: }
1.237 brouard 8357: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8358: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8359: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8360: }
1.264 brouard 8361: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8362: fprintf(ficgp,"\n#\n");
1.264 brouard 8363: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8364: fprintf(ficgp,"\nset key outside ");
8365: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8366: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8367: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8368: if (ng==1){
8369: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8370: fprintf(ficgp,"\nunset log y");
8371: }else if (ng==2){
8372: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8373: fprintf(ficgp,"\nset log y");
8374: }else if (ng==3){
8375: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8376: fprintf(ficgp,"\nset log y");
8377: }else
8378: fprintf(ficgp,"\nunset title ");
8379: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8380: i=1;
8381: for(k2=1; k2<=nlstate; k2++) {
8382: k3=i;
8383: for(k=1; k<=(nlstate+ndeath); k++) {
8384: if (k != k2){
8385: switch( ng) {
8386: case 1:
8387: if(nagesqr==0)
8388: fprintf(ficgp," p%d+p%d*x",i,i+1);
8389: else /* nagesqr =1 */
8390: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8391: break;
8392: case 2: /* ng=2 */
8393: if(nagesqr==0)
8394: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8395: else /* nagesqr =1 */
8396: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8397: break;
8398: case 3:
8399: if(nagesqr==0)
8400: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8401: else /* nagesqr =1 */
8402: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8403: break;
8404: }
8405: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8406: ijp=1; /* product no age */
8407: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8408: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8409: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8410: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.325 brouard 8411: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
1.268 brouard 8412: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
1.325 brouard 8413: if(DummyV[j]==0){/* Bug valgrind */
1.268 brouard 8414: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8415: }else{ /* quantitative */
8416: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8417: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8418: }
8419: ij++;
1.237 brouard 8420: }
1.268 brouard 8421: }
8422: }else if(cptcovprod >0){
8423: if(j==Tprod[ijp]) { /* */
8424: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8425: if(ijp <=cptcovprod) { /* Product */
8426: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8427: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8428: /* 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)]); */
8429: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8430: }else{ /* Vn is dummy and Vm is quanti */
8431: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8432: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8433: }
8434: }else{ /* Vn*Vm Vn is quanti */
8435: if(DummyV[Tvard[ijp][2]]==0){
8436: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8437: }else{ /* Both quanti */
8438: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8439: }
1.237 brouard 8440: }
1.268 brouard 8441: ijp++;
1.237 brouard 8442: }
1.268 brouard 8443: } /* end Tprod */
1.237 brouard 8444: } else{ /* simple covariate */
1.264 brouard 8445: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8446: if(Dummy[j]==0){
8447: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8448: }else{ /* quantitative */
8449: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8450: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8451: }
1.237 brouard 8452: } /* end simple */
8453: } /* end j */
1.223 brouard 8454: }else{
8455: i=i-ncovmodel;
8456: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8457: fprintf(ficgp," (1.");
8458: }
1.227 brouard 8459:
1.223 brouard 8460: if(ng != 1){
8461: fprintf(ficgp,")/(1");
1.227 brouard 8462:
1.264 brouard 8463: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8464: if(nagesqr==0)
1.264 brouard 8465: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8466: else /* nagesqr =1 */
1.264 brouard 8467: 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 8468:
1.223 brouard 8469: ij=1;
8470: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8471: if(cptcovage >0){
8472: if((j-2)==Tage[ij]) { /* Bug valgrind */
8473: if(ij <=cptcovage) { /* Bug valgrind */
8474: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8475: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8476: ij++;
8477: }
8478: }
8479: }else
8480: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/* Valgrind bug nbcode */
1.223 brouard 8481: }
8482: fprintf(ficgp,")");
8483: }
8484: fprintf(ficgp,")");
8485: if(ng ==2)
1.276 brouard 8486: 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 8487: else /* ng= 3 */
1.276 brouard 8488: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"i%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 8489: }else{ /* end ng <> 1 */
8490: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8491: 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 8492: }
8493: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8494: fprintf(ficgp,",");
8495: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8496: fprintf(ficgp,",");
8497: i=i+ncovmodel;
8498: } /* end k */
8499: } /* end k2 */
1.276 brouard 8500: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8501: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8502: } /* end k1 */
1.223 brouard 8503: } /* end ng */
8504: /* avoid: */
8505: fflush(ficgp);
1.126 brouard 8506: } /* end gnuplot */
8507:
8508:
8509: /*************** Moving average **************/
1.219 brouard 8510: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8511: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8512:
1.222 brouard 8513: int i, cpt, cptcod;
8514: int modcovmax =1;
8515: int mobilavrange, mob;
8516: int iage=0;
1.288 brouard 8517: int firstA1=0, firstA2=0;
1.222 brouard 8518:
1.266 brouard 8519: double sum=0., sumr=0.;
1.222 brouard 8520: double age;
1.266 brouard 8521: double *sumnewp, *sumnewm, *sumnewmr;
8522: double *agemingood, *agemaxgood;
8523: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8524:
8525:
1.278 brouard 8526: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8527: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8528:
8529: sumnewp = vector(1,ncovcombmax);
8530: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8531: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8532: agemingood = vector(1,ncovcombmax);
1.266 brouard 8533: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8534: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8535: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8536:
8537: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8538: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8539: sumnewp[cptcod]=0.;
1.266 brouard 8540: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8541: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8542: }
8543: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8544:
1.266 brouard 8545: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8546: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8547: else mobilavrange=mobilav;
8548: for (age=bage; age<=fage; age++)
8549: for (i=1; i<=nlstate;i++)
8550: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8551: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8552: /* We keep the original values on the extreme ages bage, fage and for
8553: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8554: we use a 5 terms etc. until the borders are no more concerned.
8555: */
8556: for (mob=3;mob <=mobilavrange;mob=mob+2){
8557: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8558: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8559: sumnewm[cptcod]=0.;
8560: for (i=1; i<=nlstate;i++){
1.222 brouard 8561: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8562: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8563: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8564: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8565: }
8566: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8567: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8568: } /* end i */
8569: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8570: } /* end cptcod */
1.222 brouard 8571: }/* end age */
8572: }/* end mob */
1.266 brouard 8573: }else{
8574: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8575: return -1;
1.266 brouard 8576: }
8577:
8578: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8579: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8580: if(invalidvarcomb[cptcod]){
8581: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8582: continue;
8583: }
1.219 brouard 8584:
1.266 brouard 8585: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8586: sumnewm[cptcod]=0.;
8587: sumnewmr[cptcod]=0.;
8588: for (i=1; i<=nlstate;i++){
8589: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8590: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8591: }
8592: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8593: agemingoodr[cptcod]=age;
8594: }
8595: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8596: agemingood[cptcod]=age;
8597: }
8598: } /* age */
8599: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8600: sumnewm[cptcod]=0.;
1.266 brouard 8601: sumnewmr[cptcod]=0.;
1.222 brouard 8602: for (i=1; i<=nlstate;i++){
8603: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8604: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8605: }
8606: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8607: agemaxgoodr[cptcod]=age;
1.222 brouard 8608: }
8609: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8610: agemaxgood[cptcod]=age;
8611: }
8612: } /* age */
8613: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8614: /* but they will change */
1.288 brouard 8615: firstA1=0;firstA2=0;
1.266 brouard 8616: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8617: sumnewm[cptcod]=0.;
8618: sumnewmr[cptcod]=0.;
8619: for (i=1; i<=nlstate;i++){
8620: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8621: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8622: }
8623: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8624: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8625: agemaxgoodr[cptcod]=age; /* age min */
8626: for (i=1; i<=nlstate;i++)
8627: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8628: }else{ /* bad we change the value with the values of good ages */
8629: for (i=1; i<=nlstate;i++){
8630: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8631: } /* i */
8632: } /* end bad */
8633: }else{
8634: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8635: agemaxgood[cptcod]=age;
8636: }else{ /* bad we change the value with the values of good ages */
8637: for (i=1; i<=nlstate;i++){
8638: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8639: } /* i */
8640: } /* end bad */
8641: }/* end else */
8642: sum=0.;sumr=0.;
8643: for (i=1; i<=nlstate;i++){
8644: sum+=mobaverage[(int)age][i][cptcod];
8645: sumr+=probs[(int)age][i][cptcod];
8646: }
8647: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8648: if(!firstA1){
8649: firstA1=1;
8650: 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);
8651: }
8652: 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 8653: } /* end bad */
8654: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8655: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8656: if(!firstA2){
8657: firstA2=1;
8658: 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);
8659: }
8660: 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 8661: } /* end bad */
8662: }/* age */
1.266 brouard 8663:
8664: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8665: sumnewm[cptcod]=0.;
1.266 brouard 8666: sumnewmr[cptcod]=0.;
1.222 brouard 8667: for (i=1; i<=nlstate;i++){
8668: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8669: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8670: }
8671: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8672: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8673: agemingoodr[cptcod]=age;
8674: for (i=1; i<=nlstate;i++)
8675: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8676: }else{ /* bad we change the value with the values of good ages */
8677: for (i=1; i<=nlstate;i++){
8678: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8679: } /* i */
8680: } /* end bad */
8681: }else{
8682: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8683: agemingood[cptcod]=age;
8684: }else{ /* bad */
8685: for (i=1; i<=nlstate;i++){
8686: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8687: } /* i */
8688: } /* end bad */
8689: }/* end else */
8690: sum=0.;sumr=0.;
8691: for (i=1; i<=nlstate;i++){
8692: sum+=mobaverage[(int)age][i][cptcod];
8693: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8694: }
1.266 brouard 8695: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8696: 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 8697: } /* end bad */
8698: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8699: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8700: 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 8701: } /* end bad */
8702: }/* age */
1.266 brouard 8703:
1.222 brouard 8704:
8705: for (age=bage; age<=fage; age++){
1.235 brouard 8706: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8707: sumnewp[cptcod]=0.;
8708: sumnewm[cptcod]=0.;
8709: for (i=1; i<=nlstate;i++){
8710: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8711: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8712: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8713: }
8714: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8715: }
8716: /* printf("\n"); */
8717: /* } */
1.266 brouard 8718:
1.222 brouard 8719: /* brutal averaging */
1.266 brouard 8720: /* for (i=1; i<=nlstate;i++){ */
8721: /* for (age=1; age<=bage; age++){ */
8722: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8723: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8724: /* } */
8725: /* for (age=fage; age<=AGESUP; age++){ */
8726: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8727: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8728: /* } */
8729: /* } /\* end i status *\/ */
8730: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8731: /* for (age=1; age<=AGESUP; age++){ */
8732: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8733: /* mobaverage[(int)age][i][cptcod]=0.; */
8734: /* } */
8735: /* } */
1.222 brouard 8736: }/* end cptcod */
1.266 brouard 8737: free_vector(agemaxgoodr,1, ncovcombmax);
8738: free_vector(agemaxgood,1, ncovcombmax);
8739: free_vector(agemingood,1, ncovcombmax);
8740: free_vector(agemingoodr,1, ncovcombmax);
8741: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8742: free_vector(sumnewm,1, ncovcombmax);
8743: free_vector(sumnewp,1, ncovcombmax);
8744: return 0;
8745: }/* End movingaverage */
1.218 brouard 8746:
1.126 brouard 8747:
1.296 brouard 8748:
1.126 brouard 8749: /************** Forecasting ******************/
1.296 brouard 8750: /* 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)*/
8751: 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){
8752: /* dateintemean, mean date of interviews
8753: dateprojd, year, month, day of starting projection
8754: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8755: agemin, agemax range of age
8756: dateprev1 dateprev2 range of dates during which prevalence is computed
8757: */
1.296 brouard 8758: /* double anprojd, mprojd, jprojd; */
8759: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8760: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8761: double agec; /* generic age */
1.296 brouard 8762: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8763: double *popeffectif,*popcount;
8764: double ***p3mat;
1.218 brouard 8765: /* double ***mobaverage; */
1.126 brouard 8766: char fileresf[FILENAMELENGTH];
8767:
8768: agelim=AGESUP;
1.211 brouard 8769: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8770: in each health status at the date of interview (if between dateprev1 and dateprev2).
8771: We still use firstpass and lastpass as another selection.
8772: */
1.214 brouard 8773: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8774: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8775:
1.201 brouard 8776: strcpy(fileresf,"F_");
8777: strcat(fileresf,fileresu);
1.126 brouard 8778: if((ficresf=fopen(fileresf,"w"))==NULL) {
8779: printf("Problem with forecast resultfile: %s\n", fileresf);
8780: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8781: }
1.235 brouard 8782: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8783: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8784:
1.225 brouard 8785: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8786:
8787:
8788: stepsize=(int) (stepm+YEARM-1)/YEARM;
8789: if (stepm<=12) stepsize=1;
8790: if(estepm < stepm){
8791: printf ("Problem %d lower than %d\n",estepm, stepm);
8792: }
1.270 brouard 8793: else{
8794: hstepm=estepm;
8795: }
8796: if(estepm > stepm){ /* Yes every two year */
8797: stepsize=2;
8798: }
1.296 brouard 8799: hstepm=hstepm/stepm;
1.126 brouard 8800:
1.296 brouard 8801:
8802: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8803: /* fractional in yp1 *\/ */
8804: /* aintmean=yp; */
8805: /* yp2=modf((yp1*12),&yp); */
8806: /* mintmean=yp; */
8807: /* yp1=modf((yp2*30.5),&yp); */
8808: /* jintmean=yp; */
8809: /* if(jintmean==0) jintmean=1; */
8810: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8811:
1.296 brouard 8812:
8813: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8814: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8815: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8816: i1=pow(2,cptcoveff);
1.126 brouard 8817: if (cptcovn < 1){i1=1;}
8818:
1.296 brouard 8819: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8820:
8821: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8822:
1.126 brouard 8823: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8824: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8825: for(k=1; k<=i1;k++){
1.253 brouard 8826: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8827: continue;
1.227 brouard 8828: if(invalidvarcomb[k]){
8829: printf("\nCombination (%d) projection ignored because no cases \n",k);
8830: continue;
8831: }
8832: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8833: for(j=1;j<=cptcoveff;j++) {
8834: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8835: }
1.235 brouard 8836: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8837: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8838: }
1.227 brouard 8839: fprintf(ficresf," yearproj age");
8840: for(j=1; j<=nlstate+ndeath;j++){
8841: for(i=1; i<=nlstate;i++)
8842: fprintf(ficresf," p%d%d",i,j);
8843: fprintf(ficresf," wp.%d",j);
8844: }
1.296 brouard 8845: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8846: fprintf(ficresf,"\n");
1.296 brouard 8847: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8848: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8849: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8850: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8851: nhstepm = nhstepm/hstepm;
8852: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8853: oldm=oldms;savm=savms;
1.268 brouard 8854: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8855: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8856: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8857: for (h=0; h<=nhstepm; h++){
8858: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8859: break;
8860: }
8861: }
8862: fprintf(ficresf,"\n");
8863: for(j=1;j<=cptcoveff;j++)
8864: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8865: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8866:
8867: for(j=1; j<=nlstate+ndeath;j++) {
8868: ppij=0.;
8869: for(i=1; i<=nlstate;i++) {
1.278 brouard 8870: if (mobilav>=1)
8871: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8872: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8873: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8874: }
1.268 brouard 8875: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8876: } /* end i */
8877: fprintf(ficresf," %.3f", ppij);
8878: }/* end j */
1.227 brouard 8879: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8880: } /* end agec */
1.266 brouard 8881: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8882: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8883: } /* end yearp */
8884: } /* end k */
1.219 brouard 8885:
1.126 brouard 8886: fclose(ficresf);
1.215 brouard 8887: printf("End of Computing forecasting \n");
8888: fprintf(ficlog,"End of Computing forecasting\n");
8889:
1.126 brouard 8890: }
8891:
1.269 brouard 8892: /************** Back Forecasting ******************/
1.296 brouard 8893: /* 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){ */
8894: 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){
8895: /* back1, year, month, day of starting backprojection
1.267 brouard 8896: agemin, agemax range of age
8897: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8898: anback2 year of end of backprojection (same day and month as back1).
8899: prevacurrent and prev are prevalences.
1.267 brouard 8900: */
8901: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8902: double agec; /* generic age */
1.302 brouard 8903: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 8904: double *popeffectif,*popcount;
8905: double ***p3mat;
8906: /* double ***mobaverage; */
8907: char fileresfb[FILENAMELENGTH];
8908:
1.268 brouard 8909: agelim=AGEINF;
1.267 brouard 8910: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8911: in each health status at the date of interview (if between dateprev1 and dateprev2).
8912: We still use firstpass and lastpass as another selection.
8913: */
8914: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8915: /* firstpass, lastpass, stepm, weightopt, model); */
8916:
8917: /*Do we need to compute prevalence again?*/
8918:
8919: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8920:
8921: strcpy(fileresfb,"FB_");
8922: strcat(fileresfb,fileresu);
8923: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8924: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8925: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8926: }
8927: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8928: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8929:
8930: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8931:
8932:
8933: stepsize=(int) (stepm+YEARM-1)/YEARM;
8934: if (stepm<=12) stepsize=1;
8935: if(estepm < stepm){
8936: printf ("Problem %d lower than %d\n",estepm, stepm);
8937: }
1.270 brouard 8938: else{
8939: hstepm=estepm;
8940: }
8941: if(estepm >= stepm){ /* Yes every two year */
8942: stepsize=2;
8943: }
1.267 brouard 8944:
8945: hstepm=hstepm/stepm;
1.296 brouard 8946: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8947: /* fractional in yp1 *\/ */
8948: /* aintmean=yp; */
8949: /* yp2=modf((yp1*12),&yp); */
8950: /* mintmean=yp; */
8951: /* yp1=modf((yp2*30.5),&yp); */
8952: /* jintmean=yp; */
8953: /* if(jintmean==0) jintmean=1; */
8954: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8955:
8956: i1=pow(2,cptcoveff);
8957: if (cptcovn < 1){i1=1;}
8958:
1.296 brouard 8959: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8960: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8961:
8962: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8963:
8964: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8965: for(k=1; k<=i1;k++){
8966: if(i1 != 1 && TKresult[nres]!= k)
8967: continue;
8968: if(invalidvarcomb[k]){
8969: printf("\nCombination (%d) projection ignored because no cases \n",k);
8970: continue;
8971: }
1.268 brouard 8972: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8973: for(j=1;j<=cptcoveff;j++) {
8974: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8975: }
8976: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8977: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8978: }
8979: fprintf(ficresfb," yearbproj age");
8980: for(j=1; j<=nlstate+ndeath;j++){
8981: for(i=1; i<=nlstate;i++)
1.268 brouard 8982: fprintf(ficresfb," b%d%d",i,j);
8983: fprintf(ficresfb," b.%d",j);
1.267 brouard 8984: }
1.296 brouard 8985: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8986: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8987: fprintf(ficresfb,"\n");
1.296 brouard 8988: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8989: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8990: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8991: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8992: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8993: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8994: nhstepm = nhstepm/hstepm;
8995: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8996: oldm=oldms;savm=savms;
1.268 brouard 8997: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8998: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8999: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9000: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9001: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9002: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9003: for (h=0; h<=nhstepm; h++){
1.268 brouard 9004: if (h*hstepm/YEARM*stepm ==-yearp) {
9005: break;
9006: }
9007: }
9008: fprintf(ficresfb,"\n");
9009: for(j=1;j<=cptcoveff;j++)
9010: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 9011: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9012: for(i=1; i<=nlstate+ndeath;i++) {
9013: ppij=0.;ppi=0.;
9014: for(j=1; j<=nlstate;j++) {
9015: /* if (mobilav==1) */
1.269 brouard 9016: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9017: ppi=ppi+prevacurrent[(int)agec][j][k];
9018: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9019: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9020: /* else { */
9021: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9022: /* } */
1.268 brouard 9023: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9024: } /* end j */
9025: if(ppi <0.99){
9026: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9027: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9028: }
9029: fprintf(ficresfb," %.3f", ppij);
9030: }/* end j */
1.267 brouard 9031: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9032: } /* end agec */
9033: } /* end yearp */
9034: } /* end k */
1.217 brouard 9035:
1.267 brouard 9036: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9037:
1.267 brouard 9038: fclose(ficresfb);
9039: printf("End of Computing Back forecasting \n");
9040: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9041:
1.267 brouard 9042: }
1.217 brouard 9043:
1.269 brouard 9044: /* Variance of prevalence limit: varprlim */
9045: 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 9046: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9047:
9048: char fileresvpl[FILENAMELENGTH];
9049: FILE *ficresvpl;
9050: double **oldm, **savm;
9051: double **varpl; /* Variances of prevalence limits by age */
9052: int i1, k, nres, j ;
9053:
9054: strcpy(fileresvpl,"VPL_");
9055: strcat(fileresvpl,fileresu);
9056: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9057: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9058: exit(0);
9059: }
1.288 brouard 9060: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9061: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9062:
9063: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9064: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9065:
9066: i1=pow(2,cptcoveff);
9067: if (cptcovn < 1){i1=1;}
9068:
9069: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9070: for(k=1; k<=i1;k++){
9071: if(i1 != 1 && TKresult[nres]!= k)
9072: continue;
9073: fprintf(ficresvpl,"\n#****** ");
9074: printf("\n#****** ");
9075: fprintf(ficlog,"\n#****** ");
9076: for(j=1;j<=cptcoveff;j++) {
9077: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9078: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9079: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9080: }
9081: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9082: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9083: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9084: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9085: }
9086: fprintf(ficresvpl,"******\n");
9087: printf("******\n");
9088: fprintf(ficlog,"******\n");
9089:
9090: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9091: oldm=oldms;savm=savms;
9092: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9093: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9094: /*}*/
9095: }
9096:
9097: fclose(ficresvpl);
1.288 brouard 9098: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9099: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9100:
9101: }
9102: /* Variance of back prevalence: varbprlim */
9103: 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){
9104: /*------- Variance of back (stable) prevalence------*/
9105:
9106: char fileresvbl[FILENAMELENGTH];
9107: FILE *ficresvbl;
9108:
9109: double **oldm, **savm;
9110: double **varbpl; /* Variances of back prevalence limits by age */
9111: int i1, k, nres, j ;
9112:
9113: strcpy(fileresvbl,"VBL_");
9114: strcat(fileresvbl,fileresu);
9115: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9116: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9117: exit(0);
9118: }
9119: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9120: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9121:
9122:
9123: i1=pow(2,cptcoveff);
9124: if (cptcovn < 1){i1=1;}
9125:
9126: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9127: for(k=1; k<=i1;k++){
9128: if(i1 != 1 && TKresult[nres]!= k)
9129: continue;
9130: fprintf(ficresvbl,"\n#****** ");
9131: printf("\n#****** ");
9132: fprintf(ficlog,"\n#****** ");
9133: for(j=1;j<=cptcoveff;j++) {
9134: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9135: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9136: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
9137: }
9138: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
9139: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9140: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9141: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
9142: }
9143: fprintf(ficresvbl,"******\n");
9144: printf("******\n");
9145: fprintf(ficlog,"******\n");
9146:
9147: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9148: oldm=oldms;savm=savms;
9149:
9150: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9151: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9152: /*}*/
9153: }
9154:
9155: fclose(ficresvbl);
9156: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9157: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9158:
9159: } /* End of varbprlim */
9160:
1.126 brouard 9161: /************** Forecasting *****not tested NB*************/
1.227 brouard 9162: /* 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 9163:
1.227 brouard 9164: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9165: /* int *popage; */
9166: /* double calagedatem, agelim, kk1, kk2; */
9167: /* double *popeffectif,*popcount; */
9168: /* double ***p3mat,***tabpop,***tabpopprev; */
9169: /* /\* double ***mobaverage; *\/ */
9170: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9171:
1.227 brouard 9172: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9173: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9174: /* agelim=AGESUP; */
9175: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9176:
1.227 brouard 9177: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9178:
9179:
1.227 brouard 9180: /* strcpy(filerespop,"POP_"); */
9181: /* strcat(filerespop,fileresu); */
9182: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9183: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9184: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9185: /* } */
9186: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9187: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9188:
1.227 brouard 9189: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9190:
1.227 brouard 9191: /* /\* if (mobilav!=0) { *\/ */
9192: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9193: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9194: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9195: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9196: /* /\* } *\/ */
9197: /* /\* } *\/ */
1.126 brouard 9198:
1.227 brouard 9199: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9200: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9201:
1.227 brouard 9202: /* agelim=AGESUP; */
1.126 brouard 9203:
1.227 brouard 9204: /* hstepm=1; */
9205: /* hstepm=hstepm/stepm; */
1.218 brouard 9206:
1.227 brouard 9207: /* if (popforecast==1) { */
9208: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9209: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9210: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9211: /* } */
9212: /* popage=ivector(0,AGESUP); */
9213: /* popeffectif=vector(0,AGESUP); */
9214: /* popcount=vector(0,AGESUP); */
1.126 brouard 9215:
1.227 brouard 9216: /* i=1; */
9217: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9218:
1.227 brouard 9219: /* imx=i; */
9220: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9221: /* } */
1.218 brouard 9222:
1.227 brouard 9223: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9224: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9225: /* k=k+1; */
9226: /* fprintf(ficrespop,"\n#******"); */
9227: /* for(j=1;j<=cptcoveff;j++) { */
9228: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9229: /* } */
9230: /* fprintf(ficrespop,"******\n"); */
9231: /* fprintf(ficrespop,"# Age"); */
9232: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9233: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9234:
1.227 brouard 9235: /* for (cpt=0; cpt<=0;cpt++) { */
9236: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9237:
1.227 brouard 9238: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9239: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9240: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9241:
1.227 brouard 9242: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9243: /* oldm=oldms;savm=savms; */
9244: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9245:
1.227 brouard 9246: /* for (h=0; h<=nhstepm; h++){ */
9247: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9248: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9249: /* } */
9250: /* for(j=1; j<=nlstate+ndeath;j++) { */
9251: /* kk1=0.;kk2=0; */
9252: /* for(i=1; i<=nlstate;i++) { */
9253: /* if (mobilav==1) */
9254: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9255: /* else { */
9256: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9257: /* } */
9258: /* } */
9259: /* if (h==(int)(calagedatem+12*cpt)){ */
9260: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9261: /* /\*fprintf(ficrespop," %.3f", kk1); */
9262: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9263: /* } */
9264: /* } */
9265: /* for(i=1; i<=nlstate;i++){ */
9266: /* kk1=0.; */
9267: /* for(j=1; j<=nlstate;j++){ */
9268: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9269: /* } */
9270: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9271: /* } */
1.218 brouard 9272:
1.227 brouard 9273: /* if (h==(int)(calagedatem+12*cpt)) */
9274: /* for(j=1; j<=nlstate;j++) */
9275: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9276: /* } */
9277: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9278: /* } */
9279: /* } */
1.218 brouard 9280:
1.227 brouard 9281: /* /\******\/ */
1.218 brouard 9282:
1.227 brouard 9283: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9284: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9285: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9286: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9287: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9288:
1.227 brouard 9289: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9290: /* oldm=oldms;savm=savms; */
9291: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9292: /* for (h=0; h<=nhstepm; h++){ */
9293: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9294: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9295: /* } */
9296: /* for(j=1; j<=nlstate+ndeath;j++) { */
9297: /* kk1=0.;kk2=0; */
9298: /* for(i=1; i<=nlstate;i++) { */
9299: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9300: /* } */
9301: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9302: /* } */
9303: /* } */
9304: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9305: /* } */
9306: /* } */
9307: /* } */
9308: /* } */
1.218 brouard 9309:
1.227 brouard 9310: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9311:
1.227 brouard 9312: /* if (popforecast==1) { */
9313: /* free_ivector(popage,0,AGESUP); */
9314: /* free_vector(popeffectif,0,AGESUP); */
9315: /* free_vector(popcount,0,AGESUP); */
9316: /* } */
9317: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9318: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9319: /* fclose(ficrespop); */
9320: /* } /\* End of popforecast *\/ */
1.218 brouard 9321:
1.126 brouard 9322: int fileappend(FILE *fichier, char *optionfich)
9323: {
9324: if((fichier=fopen(optionfich,"a"))==NULL) {
9325: printf("Problem with file: %s\n", optionfich);
9326: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9327: return (0);
9328: }
9329: fflush(fichier);
9330: return (1);
9331: }
9332:
9333:
9334: /**************** function prwizard **********************/
9335: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9336: {
9337:
9338: /* Wizard to print covariance matrix template */
9339:
1.164 brouard 9340: char ca[32], cb[32];
9341: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9342: int numlinepar;
9343:
9344: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9345: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9346: for(i=1; i <=nlstate; i++){
9347: jj=0;
9348: for(j=1; j <=nlstate+ndeath; j++){
9349: if(j==i) continue;
9350: jj++;
9351: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9352: printf("%1d%1d",i,j);
9353: fprintf(ficparo,"%1d%1d",i,j);
9354: for(k=1; k<=ncovmodel;k++){
9355: /* printf(" %lf",param[i][j][k]); */
9356: /* fprintf(ficparo," %lf",param[i][j][k]); */
9357: printf(" 0.");
9358: fprintf(ficparo," 0.");
9359: }
9360: printf("\n");
9361: fprintf(ficparo,"\n");
9362: }
9363: }
9364: printf("# Scales (for hessian or gradient estimation)\n");
9365: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9366: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9367: for(i=1; i <=nlstate; i++){
9368: jj=0;
9369: for(j=1; j <=nlstate+ndeath; j++){
9370: if(j==i) continue;
9371: jj++;
9372: fprintf(ficparo,"%1d%1d",i,j);
9373: printf("%1d%1d",i,j);
9374: fflush(stdout);
9375: for(k=1; k<=ncovmodel;k++){
9376: /* printf(" %le",delti3[i][j][k]); */
9377: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9378: printf(" 0.");
9379: fprintf(ficparo," 0.");
9380: }
9381: numlinepar++;
9382: printf("\n");
9383: fprintf(ficparo,"\n");
9384: }
9385: }
9386: printf("# Covariance matrix\n");
9387: /* # 121 Var(a12)\n\ */
9388: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9389: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9390: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9391: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9392: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9393: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9394: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9395: fflush(stdout);
9396: fprintf(ficparo,"# Covariance matrix\n");
9397: /* # 121 Var(a12)\n\ */
9398: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9399: /* # ...\n\ */
9400: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9401:
9402: for(itimes=1;itimes<=2;itimes++){
9403: jj=0;
9404: for(i=1; i <=nlstate; i++){
9405: for(j=1; j <=nlstate+ndeath; j++){
9406: if(j==i) continue;
9407: for(k=1; k<=ncovmodel;k++){
9408: jj++;
9409: ca[0]= k+'a'-1;ca[1]='\0';
9410: if(itimes==1){
9411: printf("#%1d%1d%d",i,j,k);
9412: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9413: }else{
9414: printf("%1d%1d%d",i,j,k);
9415: fprintf(ficparo,"%1d%1d%d",i,j,k);
9416: /* printf(" %.5le",matcov[i][j]); */
9417: }
9418: ll=0;
9419: for(li=1;li <=nlstate; li++){
9420: for(lj=1;lj <=nlstate+ndeath; lj++){
9421: if(lj==li) continue;
9422: for(lk=1;lk<=ncovmodel;lk++){
9423: ll++;
9424: if(ll<=jj){
9425: cb[0]= lk +'a'-1;cb[1]='\0';
9426: if(ll<jj){
9427: if(itimes==1){
9428: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9429: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9430: }else{
9431: printf(" 0.");
9432: fprintf(ficparo," 0.");
9433: }
9434: }else{
9435: if(itimes==1){
9436: printf(" Var(%s%1d%1d)",ca,i,j);
9437: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9438: }else{
9439: printf(" 0.");
9440: fprintf(ficparo," 0.");
9441: }
9442: }
9443: }
9444: } /* end lk */
9445: } /* end lj */
9446: } /* end li */
9447: printf("\n");
9448: fprintf(ficparo,"\n");
9449: numlinepar++;
9450: } /* end k*/
9451: } /*end j */
9452: } /* end i */
9453: } /* end itimes */
9454:
9455: } /* end of prwizard */
9456: /******************* Gompertz Likelihood ******************************/
9457: double gompertz(double x[])
9458: {
1.302 brouard 9459: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9460: int i,n=0; /* n is the size of the sample */
9461:
1.220 brouard 9462: for (i=1;i<=imx ; i++) {
1.126 brouard 9463: sump=sump+weight[i];
9464: /* sump=sump+1;*/
9465: num=num+1;
9466: }
1.302 brouard 9467: L=0.0;
9468: /* agegomp=AGEGOMP; */
1.126 brouard 9469: /* for (i=0; i<=imx; i++)
9470: 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]);*/
9471:
1.302 brouard 9472: for (i=1;i<=imx ; i++) {
9473: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9474: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9475: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9476: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9477: * +
9478: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9479: */
9480: if (wav[i] > 1 || agedc[i] < AGESUP) {
9481: if (cens[i] == 1){
9482: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9483: } else if (cens[i] == 0){
1.126 brouard 9484: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9485: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9486: } else
9487: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9488: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9489: L=L+A*weight[i];
1.126 brouard 9490: /* 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 9491: }
9492: }
1.126 brouard 9493:
1.302 brouard 9494: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9495:
9496: return -2*L*num/sump;
9497: }
9498:
1.136 brouard 9499: #ifdef GSL
9500: /******************* Gompertz_f Likelihood ******************************/
9501: double gompertz_f(const gsl_vector *v, void *params)
9502: {
1.302 brouard 9503: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9504: double *x= (double *) v->data;
9505: int i,n=0; /* n is the size of the sample */
9506:
9507: for (i=0;i<=imx-1 ; i++) {
9508: sump=sump+weight[i];
9509: /* sump=sump+1;*/
9510: num=num+1;
9511: }
9512:
9513:
9514: /* for (i=0; i<=imx; i++)
9515: 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]);*/
9516: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9517: for (i=1;i<=imx ; i++)
9518: {
9519: if (cens[i] == 1 && wav[i]>1)
9520: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9521:
9522: if (cens[i] == 0 && wav[i]>1)
9523: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9524: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9525:
9526: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9527: if (wav[i] > 1 ) { /* ??? */
9528: LL=LL+A*weight[i];
9529: /* 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]);*/
9530: }
9531: }
9532:
9533: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9534: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9535:
9536: return -2*LL*num/sump;
9537: }
9538: #endif
9539:
1.126 brouard 9540: /******************* Printing html file ***********/
1.201 brouard 9541: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9542: int lastpass, int stepm, int weightopt, char model[],\
9543: int imx, double p[],double **matcov,double agemortsup){
9544: int i,k;
9545:
9546: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9547: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9548: for (i=1;i<=2;i++)
9549: 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 9550: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9551: fprintf(fichtm,"</ul>");
9552:
9553: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9554:
9555: 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>");
9556:
9557: for (k=agegomp;k<(agemortsup-2);k++)
9558: 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]);
9559:
9560:
9561: fflush(fichtm);
9562: }
9563:
9564: /******************* Gnuplot file **************/
1.201 brouard 9565: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9566:
9567: char dirfileres[132],optfileres[132];
1.164 brouard 9568:
1.126 brouard 9569: int ng;
9570:
9571:
9572: /*#ifdef windows */
9573: fprintf(ficgp,"cd \"%s\" \n",pathc);
9574: /*#endif */
9575:
9576:
9577: strcpy(dirfileres,optionfilefiname);
9578: strcpy(optfileres,"vpl");
1.199 brouard 9579: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9580: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9581: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9582: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9583: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9584:
9585: }
9586:
1.136 brouard 9587: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9588: {
1.126 brouard 9589:
1.136 brouard 9590: /*-------- data file ----------*/
9591: FILE *fic;
9592: char dummy[]=" ";
1.240 brouard 9593: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9594: int lstra;
1.136 brouard 9595: int linei, month, year,iout;
1.302 brouard 9596: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9597: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9598: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9599: char *stratrunc;
1.223 brouard 9600:
1.240 brouard 9601: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9602: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 ! brouard 9603: for(v=1;v<NCOVMAX;v++){
! 9604: DummyV[v]=0;
! 9605: FixedV[v]=0;
! 9606: }
1.126 brouard 9607:
1.240 brouard 9608: for(v=1; v <=ncovcol;v++){
9609: DummyV[v]=0;
9610: FixedV[v]=0;
9611: }
9612: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9613: DummyV[v]=1;
9614: FixedV[v]=0;
9615: }
9616: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9617: DummyV[v]=0;
9618: FixedV[v]=1;
9619: }
9620: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9621: DummyV[v]=1;
9622: FixedV[v]=1;
9623: }
9624: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9625: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9626: 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]);
9627: }
1.126 brouard 9628:
1.136 brouard 9629: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9630: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9631: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9632: }
1.126 brouard 9633:
1.302 brouard 9634: /* Is it a BOM UTF-8 Windows file? */
9635: /* First data line */
9636: linei=0;
9637: while(fgets(line, MAXLINE, fic)) {
9638: noffset=0;
9639: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9640: {
9641: noffset=noffset+3;
9642: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9643: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9644: fflush(ficlog); return 1;
9645: }
9646: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9647: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9648: {
9649: noffset=noffset+2;
1.304 brouard 9650: 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);
9651: 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 9652: fflush(ficlog); return 1;
9653: }
9654: else if( line[0] == 0 && line[1] == 0)
9655: {
9656: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9657: noffset=noffset+4;
1.304 brouard 9658: 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);
9659: 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 9660: fflush(ficlog); return 1;
9661: }
9662: } else{
9663: ;/*printf(" Not a BOM file\n");*/
9664: }
9665: /* If line starts with a # it is a comment */
9666: if (line[noffset] == '#') {
9667: linei=linei+1;
9668: break;
9669: }else{
9670: break;
9671: }
9672: }
9673: fclose(fic);
9674: if((fic=fopen(datafile,"r"))==NULL) {
9675: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9676: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9677: }
9678: /* Not a Bom file */
9679:
1.136 brouard 9680: i=1;
9681: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9682: linei=linei+1;
9683: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9684: if(line[j] == '\t')
9685: line[j] = ' ';
9686: }
9687: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9688: ;
9689: };
9690: line[j+1]=0; /* Trims blanks at end of line */
9691: if(line[0]=='#'){
9692: fprintf(ficlog,"Comment line\n%s\n",line);
9693: printf("Comment line\n%s\n",line);
9694: continue;
9695: }
9696: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9697: strcpy(line, linetmp);
1.223 brouard 9698:
9699: /* Loops on waves */
9700: for (j=maxwav;j>=1;j--){
9701: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9702: cutv(stra, strb, line, ' ');
9703: if(strb[0]=='.') { /* Missing value */
9704: lval=-1;
9705: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9706: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9707: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9708: 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);
9709: 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);
9710: return 1;
9711: }
9712: }else{
9713: errno=0;
9714: /* what_kind_of_number(strb); */
9715: dval=strtod(strb,&endptr);
9716: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9717: /* if(strb != endptr && *endptr == '\0') */
9718: /* dval=dlval; */
9719: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9720: if( strb[0]=='\0' || (*endptr != '\0')){
9721: 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);
9722: 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);
9723: return 1;
9724: }
9725: cotqvar[j][iv][i]=dval;
9726: cotvar[j][ntv+iv][i]=dval;
9727: }
9728: strcpy(line,stra);
1.223 brouard 9729: }/* end loop ntqv */
1.225 brouard 9730:
1.223 brouard 9731: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9732: cutv(stra, strb, line, ' ');
9733: if(strb[0]=='.') { /* Missing value */
9734: lval=-1;
9735: }else{
9736: errno=0;
9737: lval=strtol(strb,&endptr,10);
9738: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9739: if( strb[0]=='\0' || (*endptr != '\0')){
9740: 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);
9741: 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);
9742: return 1;
9743: }
9744: }
9745: if(lval <-1 || lval >1){
9746: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9747: 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 9748: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9749: For example, for multinomial values like 1, 2 and 3,\n \
9750: build V1=0 V2=0 for the reference value (1),\n \
9751: V1=1 V2=0 for (2) \n \
1.223 brouard 9752: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9753: output of IMaCh is often meaningless.\n \
1.319 brouard 9754: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 9755: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 9756: 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 9757: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9758: For example, for multinomial values like 1, 2 and 3,\n \
9759: build V1=0 V2=0 for the reference value (1),\n \
9760: V1=1 V2=0 for (2) \n \
1.223 brouard 9761: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9762: output of IMaCh is often meaningless.\n \
1.319 brouard 9763: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 9764: return 1;
9765: }
9766: cotvar[j][iv][i]=(double)(lval);
9767: strcpy(line,stra);
1.223 brouard 9768: }/* end loop ntv */
1.225 brouard 9769:
1.223 brouard 9770: /* Statuses at wave */
1.137 brouard 9771: cutv(stra, strb, line, ' ');
1.223 brouard 9772: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9773: lval=-1;
1.136 brouard 9774: }else{
1.238 brouard 9775: errno=0;
9776: lval=strtol(strb,&endptr,10);
9777: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9778: if( strb[0]=='\0' || (*endptr != '\0')){
9779: 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);
9780: 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);
9781: return 1;
9782: }
1.136 brouard 9783: }
1.225 brouard 9784:
1.136 brouard 9785: s[j][i]=lval;
1.225 brouard 9786:
1.223 brouard 9787: /* Date of Interview */
1.136 brouard 9788: strcpy(line,stra);
9789: cutv(stra, strb,line,' ');
1.169 brouard 9790: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9791: }
1.169 brouard 9792: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9793: month=99;
9794: year=9999;
1.136 brouard 9795: }else{
1.225 brouard 9796: 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);
9797: 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);
9798: return 1;
1.136 brouard 9799: }
9800: anint[j][i]= (double) year;
1.302 brouard 9801: mint[j][i]= (double)month;
9802: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
9803: /* 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]); */
9804: /* 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]); */
9805: /* } */
1.136 brouard 9806: strcpy(line,stra);
1.223 brouard 9807: } /* End loop on waves */
1.225 brouard 9808:
1.223 brouard 9809: /* Date of death */
1.136 brouard 9810: cutv(stra, strb,line,' ');
1.169 brouard 9811: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9812: }
1.169 brouard 9813: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9814: month=99;
9815: year=9999;
9816: }else{
1.141 brouard 9817: 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 9818: 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);
9819: return 1;
1.136 brouard 9820: }
9821: andc[i]=(double) year;
9822: moisdc[i]=(double) month;
9823: strcpy(line,stra);
9824:
1.223 brouard 9825: /* Date of birth */
1.136 brouard 9826: cutv(stra, strb,line,' ');
1.169 brouard 9827: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9828: }
1.169 brouard 9829: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9830: month=99;
9831: year=9999;
9832: }else{
1.141 brouard 9833: 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);
9834: 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 9835: return 1;
1.136 brouard 9836: }
9837: if (year==9999) {
1.141 brouard 9838: 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);
9839: 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 9840: return 1;
9841:
1.136 brouard 9842: }
9843: annais[i]=(double)(year);
1.302 brouard 9844: moisnais[i]=(double)(month);
9845: for (j=1;j<=maxwav;j++){
9846: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
9847: 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]);
9848: 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]);
9849: }
9850: }
9851:
1.136 brouard 9852: strcpy(line,stra);
1.225 brouard 9853:
1.223 brouard 9854: /* Sample weight */
1.136 brouard 9855: cutv(stra, strb,line,' ');
9856: errno=0;
9857: dval=strtod(strb,&endptr);
9858: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9859: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9860: 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 9861: fflush(ficlog);
9862: return 1;
9863: }
9864: weight[i]=dval;
9865: strcpy(line,stra);
1.225 brouard 9866:
1.223 brouard 9867: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9868: cutv(stra, strb, line, ' ');
9869: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9870: lval=-1;
1.311 brouard 9871: coqvar[iv][i]=NAN;
9872: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 9873: }else{
1.225 brouard 9874: errno=0;
9875: /* what_kind_of_number(strb); */
9876: dval=strtod(strb,&endptr);
9877: /* if(strb != endptr && *endptr == '\0') */
9878: /* dval=dlval; */
9879: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9880: if( strb[0]=='\0' || (*endptr != '\0')){
9881: 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);
9882: 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);
9883: return 1;
9884: }
9885: coqvar[iv][i]=dval;
1.226 brouard 9886: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9887: }
9888: strcpy(line,stra);
9889: }/* end loop nqv */
1.136 brouard 9890:
1.223 brouard 9891: /* Covariate values */
1.136 brouard 9892: for (j=ncovcol;j>=1;j--){
9893: cutv(stra, strb,line,' ');
1.223 brouard 9894: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9895: lval=-1;
1.136 brouard 9896: }else{
1.225 brouard 9897: errno=0;
9898: lval=strtol(strb,&endptr,10);
9899: if( strb[0]=='\0' || (*endptr != '\0')){
9900: 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);
9901: 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);
9902: return 1;
9903: }
1.136 brouard 9904: }
9905: if(lval <-1 || lval >1){
1.225 brouard 9906: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9907: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9908: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9909: For example, for multinomial values like 1, 2 and 3,\n \
9910: build V1=0 V2=0 for the reference value (1),\n \
9911: V1=1 V2=0 for (2) \n \
1.136 brouard 9912: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9913: output of IMaCh is often meaningless.\n \
1.136 brouard 9914: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9915: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9916: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9917: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9918: For example, for multinomial values like 1, 2 and 3,\n \
9919: build V1=0 V2=0 for the reference value (1),\n \
9920: V1=1 V2=0 for (2) \n \
1.136 brouard 9921: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9922: output of IMaCh is often meaningless.\n \
1.136 brouard 9923: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9924: return 1;
1.136 brouard 9925: }
9926: covar[j][i]=(double)(lval);
9927: strcpy(line,stra);
9928: }
9929: lstra=strlen(stra);
1.225 brouard 9930:
1.136 brouard 9931: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9932: stratrunc = &(stra[lstra-9]);
9933: num[i]=atol(stratrunc);
9934: }
9935: else
9936: num[i]=atol(stra);
9937: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9938: 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;}*/
9939:
9940: i=i+1;
9941: } /* End loop reading data */
1.225 brouard 9942:
1.136 brouard 9943: *imax=i-1; /* Number of individuals */
9944: fclose(fic);
1.225 brouard 9945:
1.136 brouard 9946: return (0);
1.164 brouard 9947: /* endread: */
1.225 brouard 9948: printf("Exiting readdata: ");
9949: fclose(fic);
9950: return (1);
1.223 brouard 9951: }
1.126 brouard 9952:
1.234 brouard 9953: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9954: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9955: while (*p2 == ' ')
1.234 brouard 9956: p2++;
9957: /* while ((*p1++ = *p2++) !=0) */
9958: /* ; */
9959: /* do */
9960: /* while (*p2 == ' ') */
9961: /* p2++; */
9962: /* while (*p1++ == *p2++); */
9963: *stri=p2;
1.145 brouard 9964: }
9965:
1.235 brouard 9966: int decoderesult ( char resultline[], int nres)
1.230 brouard 9967: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9968: {
1.235 brouard 9969: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9970: char resultsav[MAXLINE];
1.234 brouard 9971: int resultmodel[MAXLINE];
9972: int modelresult[MAXLINE];
1.230 brouard 9973: char stra[80], strb[80], strc[80], strd[80],stre[80];
9974:
1.234 brouard 9975: removefirstspace(&resultline);
1.230 brouard 9976:
9977: if (strstr(resultline,"v") !=0){
9978: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9979: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9980: return 1;
9981: }
9982: trimbb(resultsav, resultline);
9983: if (strlen(resultsav) >1){
9984: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9985: }
1.253 brouard 9986: if(j == 0){ /* Resultline but no = */
9987: TKresult[nres]=0; /* Combination for the nresult and the model */
9988: return (0);
9989: }
1.234 brouard 9990: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.318 brouard 9991: 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 9992: 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 9993: }
9994: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9995: if(nbocc(resultsav,'=') >1){
1.318 brouard 9996: 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" */
9997: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.234 brouard 9998: }else
9999: cutl(strc,strd,resultsav,'=');
1.318 brouard 10000: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10001:
1.230 brouard 10002: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10003: 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 10004: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10005: /* cptcovsel++; */
10006: if (nbocc(stra,'=') >0)
10007: strcpy(resultsav,stra); /* and analyzes it */
10008: }
1.235 brouard 10009: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 10010: for(k1=1; k1<= cptcovt ;k1++){ /* Loop on model. model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10011: 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 10012: match=0;
1.318 brouard 10013: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10014: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 10015: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10016: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10017: break;
10018: }
10019: }
10020: if(match == 0){
1.310 brouard 10021: printf("Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
10022: fprintf(ficlog,"Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
10023: return 1;
1.234 brouard 10024: }
10025: }
10026: }
1.235 brouard 10027: /* Checking for missing or useless values in comparison of current model needs */
1.318 brouard 10028: 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 10029: match=0;
1.318 brouard 10030: 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 10031: if(Typevar[k1]==0){ /* Single */
1.237 brouard 10032: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.318 brouard 10033: resultmodel[k1]=k2; /* k2th variable of the model corresponds to k1 variable of the model. resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 10034: ++match;
10035: }
10036: }
10037: }
10038: if(match == 0){
10039: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
1.310 brouard 10040: fprintf(ficlog,"Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
10041: return 1;
1.234 brouard 10042: }else if(match > 1){
10043: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 10044: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
10045: return 1;
1.234 brouard 10046: }
10047: }
1.235 brouard 10048:
1.234 brouard 10049: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10050: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10051: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10052: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
10053: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10054: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10055: /* 1 0 0 0 */
10056: /* 2 1 0 0 */
10057: /* 3 0 1 0 */
10058: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
10059: /* 5 0 0 1 */
10060: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
10061: /* 7 0 1 1 */
10062: /* 8 1 1 1 */
1.237 brouard 10063: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10064: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10065: /* V5*age V5 known which value for nres? */
10066: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.318 brouard 10067: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* loop on model line */
1.235 brouard 10068: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 10069: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 10070: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
10071: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 10072: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
10073: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10074: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 10075: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
10076: k4++;;
10077: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
1.318 brouard 10078: k3q= resultmodel[k1]; /* resultmodel[1(V5)] = 25.1=k3q */
10079: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.237 brouard 10080: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10081: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
10082: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 10083: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
10084: k4q++;;
10085: }
10086: }
1.234 brouard 10087:
1.235 brouard 10088: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 10089: return (0);
10090: }
1.235 brouard 10091:
1.230 brouard 10092: int decodemodel( char model[], int lastobs)
10093: /**< This routine decodes the model and returns:
1.224 brouard 10094: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10095: * - nagesqr = 1 if age*age in the model, otherwise 0.
10096: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10097: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10098: * - cptcovage number of covariates with age*products =2
10099: * - cptcovs number of simple covariates
10100: * - 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
10101: * which is a new column after the 9 (ncovcol) variables.
1.319 brouard 10102: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10103: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10104: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10105: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10106: */
1.319 brouard 10107: /* 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 10108: {
1.238 brouard 10109: int i, j, k, ks, v;
1.227 brouard 10110: int j1, k1, k2, k3, k4;
1.136 brouard 10111: char modelsav[80];
1.145 brouard 10112: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10113: char *strpt;
1.136 brouard 10114:
1.145 brouard 10115: /*removespace(model);*/
1.136 brouard 10116: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10117: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10118: if (strstr(model,"AGE") !=0){
1.192 brouard 10119: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10120: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10121: return 1;
10122: }
1.141 brouard 10123: if (strstr(model,"v") !=0){
10124: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
10125: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
10126: return 1;
10127: }
1.187 brouard 10128: strcpy(modelsav,model);
10129: if ((strpt=strstr(model,"age*age")) !=0){
10130: printf(" strpt=%s, model=%s\n",strpt, model);
10131: if(strpt != model){
1.234 brouard 10132: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10133: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10134: corresponding column of parameters.\n",model);
1.234 brouard 10135: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10136: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10137: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10138: return 1;
1.225 brouard 10139: }
1.187 brouard 10140: nagesqr=1;
10141: if (strstr(model,"+age*age") !=0)
1.234 brouard 10142: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10143: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10144: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10145: else
1.234 brouard 10146: substrchaine(modelsav, model, "age*age");
1.187 brouard 10147: }else
10148: nagesqr=0;
10149: if (strlen(modelsav) >1){
10150: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10151: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10152: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10153: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10154: * cst, age and age*age
10155: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10156: /* including age products which are counted in cptcovage.
10157: * but the covariates which are products must be treated
10158: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10159: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10160: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10161:
10162:
1.187 brouard 10163: /* Design
10164: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10165: * < ncovcol=8 >
10166: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10167: * k= 1 2 3 4 5 6 7 8
10168: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10169: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10170: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10171: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10172: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10173: * Tage[++cptcovage]=k
10174: * if products, new covar are created after ncovcol with k1
10175: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10176: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10177: * 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
10178: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10179: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10180: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10181: * < ncovcol=8 >
10182: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10183: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10184: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10185: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10186: * p Tprod[1]@2={ 6, 5}
10187: *p Tvard[1][1]@4= {7, 8, 5, 6}
10188: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10189: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10190: *How to reorganize? Tvars(orted)
1.187 brouard 10191: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10192: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
10193: * {2, 1, 4, 8, 5, 6, 3, 7}
10194: * Struct []
10195: */
1.225 brouard 10196:
1.187 brouard 10197: /* This loop fills the array Tvar from the string 'model'.*/
10198: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
10199: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
10200: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
10201: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
10202: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
10203: /* k=1 Tvar[1]=2 (from V2) */
10204: /* k=5 Tvar[5] */
10205: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 10206: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 10207: /* } */
1.198 brouard 10208: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 10209: /*
10210: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 10211: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
10212: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
10213: }
1.187 brouard 10214: cptcovage=0;
1.319 brouard 10215: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
10216: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
10217: 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" */
10218: if (nbocc(modelsav,'+')==0)
10219: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 10220: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
10221: /*scanf("%d",i);*/
1.319 brouard 10222: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
10223: 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 10224: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
10225: /* covar is not filled and then is empty */
10226: cptcovprod--;
10227: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 10228: 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 10229: Typevar[k]=1; /* 1 for age product */
1.319 brouard 10230: cptcovage++; /* Counts the number of covariates which include age as a product */
10231: 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 10232: /*printf("stre=%s ", stre);*/
10233: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
10234: cptcovprod--;
10235: cutl(stre,strb,strc,'V');
10236: Tvar[k]=atoi(stre);
10237: Typevar[k]=1; /* 1 for age product */
10238: cptcovage++;
10239: Tage[cptcovage]=k;
10240: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
10241: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
10242: cptcovn++;
10243: cptcovprodnoage++;k1++;
10244: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10245: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10246: because this model-covariate is a construction we invent a new column
10247: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.319 brouard 10248: If already ncovcol=4 and model=V2 + V1 +V1*V4 +age*V3 +V3*V2
10249: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
10250: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=4 etc */
1.234 brouard 10251: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10252: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10253: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 10254: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 10255: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
10256: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
10257: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10258: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10259: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10260: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10261: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10262: for (i=1; i<=lastobs;i++){
10263: /* Computes the new covariate which is a product of
10264: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10265: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10266: }
10267: } /* End age is not in the model */
10268: } /* End if model includes a product */
1.319 brouard 10269: else { /* not a product */
1.234 brouard 10270: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10271: /* scanf("%d",i);*/
10272: cutl(strd,strc,strb,'V');
10273: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10274: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10275: Tvar[k]=atoi(strd);
10276: Typevar[k]=0; /* 0 for simple covariates */
10277: }
10278: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10279: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10280: scanf("%d",i);*/
1.187 brouard 10281: } /* end of loop + on total covariates */
10282: } /* end if strlen(modelsave == 0) age*age might exist */
10283: } /* end if strlen(model == 0) */
1.136 brouard 10284:
10285: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10286: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10287:
1.136 brouard 10288: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10289: printf("cptcovprod=%d ", cptcovprod);
10290: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10291: scanf("%d ",i);*/
10292:
10293:
1.230 brouard 10294: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10295: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10296: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10297: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10298: k = 1 2 3 4 5 6 7 8 9
10299: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 10300: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 10301: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10302: Dummy[k] 1 0 0 0 3 1 1 2 3
10303: Tmodelind[combination of covar]=k;
1.225 brouard 10304: */
10305: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10306: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10307: /* 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 10308: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 10309: printf("Model=1+age+%s\n\
1.227 brouard 10310: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10311: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10312: 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 10313: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 10314: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10315: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10316: 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 10317: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10318: 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 */
10319: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10320: Fixed[k]= 0;
10321: Dummy[k]= 0;
1.225 brouard 10322: ncoveff++;
1.232 brouard 10323: ncovf++;
1.234 brouard 10324: nsd++;
10325: modell[k].maintype= FTYPE;
10326: TvarsD[nsd]=Tvar[k];
10327: TvarsDind[nsd]=k;
10328: TvarF[ncovf]=Tvar[k];
10329: TvarFind[ncovf]=k;
10330: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10331: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10332: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10333: Fixed[k]= 0;
10334: Dummy[k]= 0;
10335: ncoveff++;
10336: ncovf++;
10337: modell[k].maintype= FTYPE;
10338: TvarF[ncovf]=Tvar[k];
10339: TvarFind[ncovf]=k;
1.230 brouard 10340: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10341: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10342: }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 10343: Fixed[k]= 0;
10344: Dummy[k]= 1;
1.230 brouard 10345: nqfveff++;
1.234 brouard 10346: modell[k].maintype= FTYPE;
10347: modell[k].subtype= FQ;
10348: nsq++;
10349: TvarsQ[nsq]=Tvar[k];
10350: TvarsQind[nsq]=k;
1.232 brouard 10351: ncovf++;
1.234 brouard 10352: TvarF[ncovf]=Tvar[k];
10353: TvarFind[ncovf]=k;
1.231 brouard 10354: 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 10355: 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 10356: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10357: Fixed[k]= 1;
10358: Dummy[k]= 0;
1.225 brouard 10359: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10360: modell[k].maintype= VTYPE;
10361: modell[k].subtype= VD;
10362: nsd++;
10363: TvarsD[nsd]=Tvar[k];
10364: TvarsDind[nsd]=k;
10365: ncovv++; /* Only simple time varying variables */
10366: TvarV[ncovv]=Tvar[k];
1.242 brouard 10367: 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 10368: 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 */
10369: 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 10370: 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);
10371: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10372: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10373: Fixed[k]= 1;
10374: Dummy[k]= 1;
10375: nqtveff++;
10376: modell[k].maintype= VTYPE;
10377: modell[k].subtype= VQ;
10378: ncovv++; /* Only simple time varying variables */
10379: nsq++;
1.319 brouard 10380: TvarsQ[nsq]=Tvar[k]; /* k=1 Tvar=5 nsq=1 TvarsQ[1]=5 */
1.234 brouard 10381: TvarsQind[nsq]=k;
10382: TvarV[ncovv]=Tvar[k];
1.242 brouard 10383: 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 10384: 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 */
10385: 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 10386: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10387: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10388: 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 10389: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10390: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10391: ncova++;
10392: TvarA[ncova]=Tvar[k];
10393: TvarAind[ncova]=k;
1.231 brouard 10394: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10395: Fixed[k]= 2;
10396: Dummy[k]= 2;
10397: modell[k].maintype= ATYPE;
10398: modell[k].subtype= APFD;
10399: /* ncoveff++; */
1.227 brouard 10400: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10401: Fixed[k]= 2;
10402: Dummy[k]= 3;
10403: modell[k].maintype= ATYPE;
10404: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10405: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10406: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10407: Fixed[k]= 3;
10408: Dummy[k]= 2;
10409: modell[k].maintype= ATYPE;
10410: modell[k].subtype= APVD; /* Product age * varying dummy */
10411: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10412: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10413: Fixed[k]= 3;
10414: Dummy[k]= 3;
10415: modell[k].maintype= ATYPE;
10416: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10417: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10418: }
10419: }else if (Typevar[k] == 2) { /* product without age */
10420: k1=Tposprod[k];
10421: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10422: if(Tvard[k1][2] <=ncovcol){
10423: Fixed[k]= 1;
10424: Dummy[k]= 0;
10425: modell[k].maintype= FTYPE;
10426: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10427: ncovf++; /* Fixed variables without age */
10428: TvarF[ncovf]=Tvar[k];
10429: TvarFind[ncovf]=k;
10430: }else if(Tvard[k1][2] <=ncovcol+nqv){
10431: Fixed[k]= 0; /* or 2 ?*/
10432: Dummy[k]= 1;
10433: modell[k].maintype= FTYPE;
10434: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10435: ncovf++; /* Varying variables without age */
10436: TvarF[ncovf]=Tvar[k];
10437: TvarFind[ncovf]=k;
10438: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10439: Fixed[k]= 1;
10440: Dummy[k]= 0;
10441: modell[k].maintype= VTYPE;
10442: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10443: ncovv++; /* Varying variables without age */
10444: TvarV[ncovv]=Tvar[k];
10445: TvarVind[ncovv]=k;
10446: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10447: Fixed[k]= 1;
10448: Dummy[k]= 1;
10449: modell[k].maintype= VTYPE;
10450: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10451: ncovv++; /* Varying variables without age */
10452: TvarV[ncovv]=Tvar[k];
10453: TvarVind[ncovv]=k;
10454: }
1.227 brouard 10455: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10456: if(Tvard[k1][2] <=ncovcol){
10457: Fixed[k]= 0; /* or 2 ?*/
10458: Dummy[k]= 1;
10459: modell[k].maintype= FTYPE;
10460: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10461: ncovf++; /* Fixed variables without age */
10462: TvarF[ncovf]=Tvar[k];
10463: TvarFind[ncovf]=k;
10464: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10465: Fixed[k]= 1;
10466: Dummy[k]= 1;
10467: modell[k].maintype= VTYPE;
10468: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10469: ncovv++; /* Varying variables without age */
10470: TvarV[ncovv]=Tvar[k];
10471: TvarVind[ncovv]=k;
10472: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10473: Fixed[k]= 1;
10474: Dummy[k]= 1;
10475: modell[k].maintype= VTYPE;
10476: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10477: ncovv++; /* Varying variables without age */
10478: TvarV[ncovv]=Tvar[k];
10479: TvarVind[ncovv]=k;
10480: ncovv++; /* Varying variables without age */
10481: TvarV[ncovv]=Tvar[k];
10482: TvarVind[ncovv]=k;
10483: }
1.227 brouard 10484: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10485: if(Tvard[k1][2] <=ncovcol){
10486: Fixed[k]= 1;
10487: Dummy[k]= 1;
10488: modell[k].maintype= VTYPE;
10489: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10490: ncovv++; /* Varying variables without age */
10491: TvarV[ncovv]=Tvar[k];
10492: TvarVind[ncovv]=k;
10493: }else if(Tvard[k1][2] <=ncovcol+nqv){
10494: Fixed[k]= 1;
10495: Dummy[k]= 1;
10496: modell[k].maintype= VTYPE;
10497: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10498: ncovv++; /* Varying variables without age */
10499: TvarV[ncovv]=Tvar[k];
10500: TvarVind[ncovv]=k;
10501: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10502: Fixed[k]= 1;
10503: Dummy[k]= 0;
10504: modell[k].maintype= VTYPE;
10505: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10506: ncovv++; /* Varying variables without age */
10507: TvarV[ncovv]=Tvar[k];
10508: TvarVind[ncovv]=k;
10509: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10510: Fixed[k]= 1;
10511: Dummy[k]= 1;
10512: modell[k].maintype= VTYPE;
10513: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10514: ncovv++; /* Varying variables without age */
10515: TvarV[ncovv]=Tvar[k];
10516: TvarVind[ncovv]=k;
10517: }
1.227 brouard 10518: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10519: if(Tvard[k1][2] <=ncovcol){
10520: Fixed[k]= 1;
10521: Dummy[k]= 1;
10522: modell[k].maintype= VTYPE;
10523: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10524: ncovv++; /* Varying variables without age */
10525: TvarV[ncovv]=Tvar[k];
10526: TvarVind[ncovv]=k;
10527: }else if(Tvard[k1][2] <=ncovcol+nqv){
10528: Fixed[k]= 1;
10529: Dummy[k]= 1;
10530: modell[k].maintype= VTYPE;
10531: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10532: ncovv++; /* Varying variables without age */
10533: TvarV[ncovv]=Tvar[k];
10534: TvarVind[ncovv]=k;
10535: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10536: Fixed[k]= 1;
10537: Dummy[k]= 1;
10538: modell[k].maintype= VTYPE;
10539: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10540: ncovv++; /* Varying variables without age */
10541: TvarV[ncovv]=Tvar[k];
10542: TvarVind[ncovv]=k;
10543: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10544: Fixed[k]= 1;
10545: Dummy[k]= 1;
10546: modell[k].maintype= VTYPE;
10547: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10548: ncovv++; /* Varying variables without age */
10549: TvarV[ncovv]=Tvar[k];
10550: TvarVind[ncovv]=k;
10551: }
1.227 brouard 10552: }else{
1.240 brouard 10553: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10554: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10555: } /*end k1*/
1.225 brouard 10556: }else{
1.226 brouard 10557: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10558: 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 10559: }
1.227 brouard 10560: 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 10561: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10562: 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]);
10563: }
10564: /* Searching for doublons in the model */
10565: for(k1=1; k1<= cptcovt;k1++){
10566: for(k2=1; k2 <k1;k2++){
1.285 brouard 10567: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10568: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10569: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10570: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10571: 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]);
10572: 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 10573: return(1);
10574: }
10575: }else if (Typevar[k1] ==2){
10576: k3=Tposprod[k1];
10577: k4=Tposprod[k2];
10578: 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])) ){
10579: 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]]);
10580: 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);
10581: return(1);
10582: }
10583: }
1.227 brouard 10584: }
10585: }
1.225 brouard 10586: }
10587: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10588: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10589: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10590: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10591: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10592: /*endread:*/
1.225 brouard 10593: printf("Exiting decodemodel: ");
10594: return (1);
1.136 brouard 10595: }
10596:
1.169 brouard 10597: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10598: {/* Check ages at death */
1.136 brouard 10599: int i, m;
1.218 brouard 10600: int firstone=0;
10601:
1.136 brouard 10602: for (i=1; i<=imx; i++) {
10603: for(m=2; (m<= maxwav); m++) {
10604: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10605: anint[m][i]=9999;
1.216 brouard 10606: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10607: s[m][i]=-1;
1.136 brouard 10608: }
10609: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10610: *nberr = *nberr + 1;
1.218 brouard 10611: if(firstone == 0){
10612: firstone=1;
1.260 brouard 10613: 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 10614: }
1.262 brouard 10615: 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 10616: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10617: }
10618: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10619: (*nberr)++;
1.259 brouard 10620: 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 10621: 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 10622: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10623: }
10624: }
10625: }
10626:
10627: for (i=1; i<=imx; i++) {
10628: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10629: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10630: 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 10631: if (s[m][i] >= nlstate+1) {
1.169 brouard 10632: if(agedc[i]>0){
10633: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10634: agev[m][i]=agedc[i];
1.214 brouard 10635: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10636: }else {
1.136 brouard 10637: if ((int)andc[i]!=9999){
10638: nbwarn++;
10639: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10640: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10641: agev[m][i]=-1;
10642: }
10643: }
1.169 brouard 10644: } /* agedc > 0 */
1.214 brouard 10645: } /* end if */
1.136 brouard 10646: else if(s[m][i] !=9){ /* Standard case, age in fractional
10647: years but with the precision of a month */
10648: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10649: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10650: agev[m][i]=1;
10651: else if(agev[m][i] < *agemin){
10652: *agemin=agev[m][i];
10653: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10654: }
10655: else if(agev[m][i] >*agemax){
10656: *agemax=agev[m][i];
1.156 brouard 10657: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10658: }
10659: /*agev[m][i]=anint[m][i]-annais[i];*/
10660: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10661: } /* en if 9*/
1.136 brouard 10662: else { /* =9 */
1.214 brouard 10663: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10664: agev[m][i]=1;
10665: s[m][i]=-1;
10666: }
10667: }
1.214 brouard 10668: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10669: agev[m][i]=1;
1.214 brouard 10670: else{
10671: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10672: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10673: agev[m][i]=0;
10674: }
10675: } /* End for lastpass */
10676: }
1.136 brouard 10677:
10678: for (i=1; i<=imx; i++) {
10679: for(m=firstpass; (m<=lastpass); m++){
10680: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10681: (*nberr)++;
1.136 brouard 10682: 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);
10683: 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);
10684: return 1;
10685: }
10686: }
10687: }
10688:
10689: /*for (i=1; i<=imx; i++){
10690: for (m=firstpass; (m<lastpass); m++){
10691: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10692: }
10693:
10694: }*/
10695:
10696:
1.139 brouard 10697: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10698: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10699:
10700: return (0);
1.164 brouard 10701: /* endread:*/
1.136 brouard 10702: printf("Exiting calandcheckages: ");
10703: return (1);
10704: }
10705:
1.172 brouard 10706: #if defined(_MSC_VER)
10707: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10708: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10709: //#include "stdafx.h"
10710: //#include <stdio.h>
10711: //#include <tchar.h>
10712: //#include <windows.h>
10713: //#include <iostream>
10714: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10715:
10716: LPFN_ISWOW64PROCESS fnIsWow64Process;
10717:
10718: BOOL IsWow64()
10719: {
10720: BOOL bIsWow64 = FALSE;
10721:
10722: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10723: // (HANDLE, PBOOL);
10724:
10725: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10726:
10727: HMODULE module = GetModuleHandle(_T("kernel32"));
10728: const char funcName[] = "IsWow64Process";
10729: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10730: GetProcAddress(module, funcName);
10731:
10732: if (NULL != fnIsWow64Process)
10733: {
10734: if (!fnIsWow64Process(GetCurrentProcess(),
10735: &bIsWow64))
10736: //throw std::exception("Unknown error");
10737: printf("Unknown error\n");
10738: }
10739: return bIsWow64 != FALSE;
10740: }
10741: #endif
1.177 brouard 10742:
1.191 brouard 10743: void syscompilerinfo(int logged)
1.292 brouard 10744: {
10745: #include <stdint.h>
10746:
10747: /* #include "syscompilerinfo.h"*/
1.185 brouard 10748: /* command line Intel compiler 32bit windows, XP compatible:*/
10749: /* /GS /W3 /Gy
10750: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10751: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10752: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10753: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10754: */
10755: /* 64 bits */
1.185 brouard 10756: /*
10757: /GS /W3 /Gy
10758: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10759: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10760: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10761: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10762: /* Optimization are useless and O3 is slower than O2 */
10763: /*
10764: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10765: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10766: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10767: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10768: */
1.186 brouard 10769: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10770: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10771: /PDB:"visual studio
10772: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10773: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10774: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10775: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10776: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10777: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10778: uiAccess='false'"
10779: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10780: /NOLOGO /TLBID:1
10781: */
1.292 brouard 10782:
10783:
1.177 brouard 10784: #if defined __INTEL_COMPILER
1.178 brouard 10785: #if defined(__GNUC__)
10786: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10787: #endif
1.177 brouard 10788: #elif defined(__GNUC__)
1.179 brouard 10789: #ifndef __APPLE__
1.174 brouard 10790: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10791: #endif
1.177 brouard 10792: struct utsname sysInfo;
1.178 brouard 10793: int cross = CROSS;
10794: if (cross){
10795: printf("Cross-");
1.191 brouard 10796: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10797: }
1.174 brouard 10798: #endif
10799:
1.191 brouard 10800: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10801: #if defined(__clang__)
1.191 brouard 10802: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10803: #endif
10804: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10805: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10806: #endif
10807: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10808: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10809: #endif
10810: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10811: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10812: #endif
10813: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10814: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10815: #endif
10816: #if defined(_MSC_VER)
1.191 brouard 10817: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10818: #endif
10819: #if defined(__PGI)
1.191 brouard 10820: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10821: #endif
10822: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10823: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10824: #endif
1.191 brouard 10825: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10826:
1.167 brouard 10827: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10828: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10829: // Windows (x64 and x86)
1.191 brouard 10830: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10831: #elif __unix__ // all unices, not all compilers
10832: // Unix
1.191 brouard 10833: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10834: #elif __linux__
10835: // linux
1.191 brouard 10836: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10837: #elif __APPLE__
1.174 brouard 10838: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10839: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10840: #endif
10841:
10842: /* __MINGW32__ */
10843: /* __CYGWIN__ */
10844: /* __MINGW64__ */
10845: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10846: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10847: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10848: /* _WIN64 // Defined for applications for Win64. */
10849: /* _M_X64 // Defined for compilations that target x64 processors. */
10850: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10851:
1.167 brouard 10852: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10853: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10854: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10855: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10856: #else
1.191 brouard 10857: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10858: #endif
10859:
1.169 brouard 10860: #if defined(__GNUC__)
10861: # if defined(__GNUC_PATCHLEVEL__)
10862: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10863: + __GNUC_MINOR__ * 100 \
10864: + __GNUC_PATCHLEVEL__)
10865: # else
10866: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10867: + __GNUC_MINOR__ * 100)
10868: # endif
1.174 brouard 10869: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10870: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10871:
10872: if (uname(&sysInfo) != -1) {
10873: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10874: 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 10875: }
10876: else
10877: perror("uname() error");
1.179 brouard 10878: //#ifndef __INTEL_COMPILER
10879: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10880: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10881: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10882: #endif
1.169 brouard 10883: #endif
1.172 brouard 10884:
1.286 brouard 10885: // void main ()
1.172 brouard 10886: // {
1.169 brouard 10887: #if defined(_MSC_VER)
1.174 brouard 10888: if (IsWow64()){
1.191 brouard 10889: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10890: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10891: }
10892: else{
1.191 brouard 10893: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10894: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10895: }
1.172 brouard 10896: // printf("\nPress Enter to continue...");
10897: // getchar();
10898: // }
10899:
1.169 brouard 10900: #endif
10901:
1.167 brouard 10902:
1.219 brouard 10903: }
1.136 brouard 10904:
1.219 brouard 10905: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10906: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10907: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10908: /* double ftolpl = 1.e-10; */
1.180 brouard 10909: double age, agebase, agelim;
1.203 brouard 10910: double tot;
1.180 brouard 10911:
1.202 brouard 10912: strcpy(filerespl,"PL_");
10913: strcat(filerespl,fileresu);
10914: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10915: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10916: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10917: }
1.288 brouard 10918: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10919: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10920: pstamp(ficrespl);
1.288 brouard 10921: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10922: fprintf(ficrespl,"#Age ");
10923: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10924: fprintf(ficrespl,"\n");
1.180 brouard 10925:
1.219 brouard 10926: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10927:
1.219 brouard 10928: agebase=ageminpar;
10929: agelim=agemaxpar;
1.180 brouard 10930:
1.227 brouard 10931: /* i1=pow(2,ncoveff); */
1.234 brouard 10932: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10933: if (cptcovn < 1){i1=1;}
1.180 brouard 10934:
1.238 brouard 10935: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10936: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10937: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10938: continue;
1.235 brouard 10939:
1.238 brouard 10940: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10941: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10942: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10943: /* k=k+1; */
10944: /* to clean */
10945: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10946: fprintf(ficrespl,"#******");
10947: printf("#******");
10948: fprintf(ficlog,"#******");
10949: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10950: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10951: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10952: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10953: }
10954: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10955: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10956: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10957: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10958: }
10959: fprintf(ficrespl,"******\n");
10960: printf("******\n");
10961: fprintf(ficlog,"******\n");
10962: if(invalidvarcomb[k]){
10963: printf("\nCombination (%d) ignored because no case \n",k);
10964: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10965: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10966: continue;
10967: }
1.219 brouard 10968:
1.238 brouard 10969: fprintf(ficrespl,"#Age ");
10970: for(j=1;j<=cptcoveff;j++) {
10971: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10972: }
10973: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10974: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10975:
1.238 brouard 10976: for (age=agebase; age<=agelim; age++){
10977: /* for (age=agebase; age<=agebase; age++){ */
10978: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10979: fprintf(ficrespl,"%.0f ",age );
10980: for(j=1;j<=cptcoveff;j++)
10981: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10982: tot=0.;
10983: for(i=1; i<=nlstate;i++){
10984: tot += prlim[i][i];
10985: fprintf(ficrespl," %.5f", prlim[i][i]);
10986: }
10987: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10988: } /* Age */
10989: /* was end of cptcod */
10990: } /* cptcov */
10991: } /* nres */
1.219 brouard 10992: return 0;
1.180 brouard 10993: }
10994:
1.218 brouard 10995: 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 10996: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10997:
10998: /* Computes the back prevalence limit for any combination of covariate values
10999: * at any age between ageminpar and agemaxpar
11000: */
1.235 brouard 11001: int i, j, k, i1, nres=0 ;
1.217 brouard 11002: /* double ftolpl = 1.e-10; */
11003: double age, agebase, agelim;
11004: double tot;
1.218 brouard 11005: /* double ***mobaverage; */
11006: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 11007:
11008: strcpy(fileresplb,"PLB_");
11009: strcat(fileresplb,fileresu);
11010: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 11011: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
11012: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 11013: }
1.288 brouard 11014: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
11015: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 11016: pstamp(ficresplb);
1.288 brouard 11017: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 11018: fprintf(ficresplb,"#Age ");
11019: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
11020: fprintf(ficresplb,"\n");
11021:
1.218 brouard 11022:
11023: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
11024:
11025: agebase=ageminpar;
11026: agelim=agemaxpar;
11027:
11028:
1.227 brouard 11029: i1=pow(2,cptcoveff);
1.218 brouard 11030: if (cptcovn < 1){i1=1;}
1.227 brouard 11031:
1.238 brouard 11032: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11033: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11034: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11035: continue;
11036: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
11037: fprintf(ficresplb,"#******");
11038: printf("#******");
11039: fprintf(ficlog,"#******");
11040: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
11041: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11042: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11043: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11044: }
11045: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11046: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11047: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11048: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11049: }
11050: fprintf(ficresplb,"******\n");
11051: printf("******\n");
11052: fprintf(ficlog,"******\n");
11053: if(invalidvarcomb[k]){
11054: printf("\nCombination (%d) ignored because no cases \n",k);
11055: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
11056: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
11057: continue;
11058: }
1.218 brouard 11059:
1.238 brouard 11060: fprintf(ficresplb,"#Age ");
11061: for(j=1;j<=cptcoveff;j++) {
11062: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11063: }
11064: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
11065: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 11066:
11067:
1.238 brouard 11068: for (age=agebase; age<=agelim; age++){
11069: /* for (age=agebase; age<=agebase; age++){ */
11070: if(mobilavproj > 0){
11071: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11072: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11073: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11074: }else if (mobilavproj == 0){
11075: 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);
11076: 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);
11077: exit(1);
11078: }else{
11079: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11080: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11081: /* printf("TOTOT\n"); */
11082: /* exit(1); */
1.238 brouard 11083: }
11084: fprintf(ficresplb,"%.0f ",age );
11085: for(j=1;j<=cptcoveff;j++)
11086: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11087: tot=0.;
11088: for(i=1; i<=nlstate;i++){
11089: tot += bprlim[i][i];
11090: fprintf(ficresplb," %.5f", bprlim[i][i]);
11091: }
11092: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11093: } /* Age */
11094: /* was end of cptcod */
1.255 brouard 11095: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 11096: } /* end of any combination */
11097: } /* end of nres */
1.218 brouard 11098: /* hBijx(p, bage, fage); */
11099: /* fclose(ficrespijb); */
11100:
11101: return 0;
1.217 brouard 11102: }
1.218 brouard 11103:
1.180 brouard 11104: int hPijx(double *p, int bage, int fage){
11105: /*------------- h Pij x at various ages ------------*/
11106:
11107: int stepsize;
11108: int agelim;
11109: int hstepm;
11110: int nhstepm;
1.235 brouard 11111: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11112:
11113: double agedeb;
11114: double ***p3mat;
11115:
1.201 brouard 11116: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 11117: if((ficrespij=fopen(filerespij,"w"))==NULL) {
11118: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
11119: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
11120: }
11121: printf("Computing pij: result on file '%s' \n", filerespij);
11122: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
11123:
11124: stepsize=(int) (stepm+YEARM-1)/YEARM;
11125: /*if (stepm<=24) stepsize=2;*/
11126:
11127: agelim=AGESUP;
11128: hstepm=stepsize*YEARM; /* Every year of age */
11129: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 11130:
1.180 brouard 11131: /* hstepm=1; aff par mois*/
11132: pstamp(ficrespij);
11133: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 11134: i1= pow(2,cptcoveff);
1.218 brouard 11135: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11136: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11137: /* k=k+1; */
1.235 brouard 11138: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11139: for(k=1; k<=i1;k++){
1.253 brouard 11140: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11141: continue;
1.183 brouard 11142: fprintf(ficrespij,"\n#****** ");
1.227 brouard 11143: for(j=1;j<=cptcoveff;j++)
1.198 brouard 11144: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 11145: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11146: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11147: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11148: }
1.183 brouard 11149: fprintf(ficrespij,"******\n");
11150:
11151: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
11152: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
11153: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
11154:
11155: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 11156:
1.183 brouard 11157: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11158: oldm=oldms;savm=savms;
1.235 brouard 11159: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 11160: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
11161: for(i=1; i<=nlstate;i++)
11162: for(j=1; j<=nlstate+ndeath;j++)
11163: fprintf(ficrespij," %1d-%1d",i,j);
11164: fprintf(ficrespij,"\n");
11165: for (h=0; h<=nhstepm; h++){
11166: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11167: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 11168: for(i=1; i<=nlstate;i++)
11169: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 11170: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 11171: fprintf(ficrespij,"\n");
11172: }
1.183 brouard 11173: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11174: fprintf(ficrespij,"\n");
11175: }
1.180 brouard 11176: /*}*/
11177: }
1.218 brouard 11178: return 0;
1.180 brouard 11179: }
1.218 brouard 11180:
11181: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 11182: /*------------- h Bij x at various ages ------------*/
11183:
11184: int stepsize;
1.218 brouard 11185: /* int agelim; */
11186: int ageminl;
1.217 brouard 11187: int hstepm;
11188: int nhstepm;
1.238 brouard 11189: int h, i, i1, j, k, nres;
1.218 brouard 11190:
1.217 brouard 11191: double agedeb;
11192: double ***p3mat;
1.218 brouard 11193:
11194: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
11195: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
11196: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11197: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11198: }
11199: printf("Computing pij back: result on file '%s' \n", filerespijb);
11200: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
11201:
11202: stepsize=(int) (stepm+YEARM-1)/YEARM;
11203: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 11204:
1.218 brouard 11205: /* agelim=AGESUP; */
1.289 brouard 11206: ageminl=AGEINF; /* was 30 */
1.218 brouard 11207: hstepm=stepsize*YEARM; /* Every year of age */
11208: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11209:
11210: /* hstepm=1; aff par mois*/
11211: pstamp(ficrespijb);
1.255 brouard 11212: 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 11213: i1= pow(2,cptcoveff);
1.218 brouard 11214: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11215: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11216: /* k=k+1; */
1.238 brouard 11217: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11218: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11219: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11220: continue;
11221: fprintf(ficrespijb,"\n#****** ");
11222: for(j=1;j<=cptcoveff;j++)
11223: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
11224: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
11225: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
11226: }
11227: fprintf(ficrespijb,"******\n");
1.264 brouard 11228: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 11229: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
11230: continue;
11231: }
11232:
11233: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
11234: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
11235: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 11236: 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 */
11237: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 11238:
11239: /* nhstepm=nhstepm*YEARM; aff par mois*/
11240:
1.266 brouard 11241: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
11242: /* and memory limitations if stepm is small */
11243:
1.238 brouard 11244: /* oldm=oldms;savm=savms; */
11245: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.325 brouard 11246: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
1.238 brouard 11247: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 11248: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 11249: for(i=1; i<=nlstate;i++)
11250: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 11251: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 11252: fprintf(ficrespijb,"\n");
1.238 brouard 11253: for (h=0; h<=nhstepm; h++){
11254: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11255: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
11256: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
11257: for(i=1; i<=nlstate;i++)
11258: for(j=1; j<=nlstate+ndeath;j++)
1.325 brouard 11259: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.238 brouard 11260: fprintf(ficrespijb,"\n");
11261: }
11262: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11263: fprintf(ficrespijb,"\n");
11264: } /* end age deb */
11265: } /* end combination */
11266: } /* end nres */
1.218 brouard 11267: return 0;
11268: } /* hBijx */
1.217 brouard 11269:
1.180 brouard 11270:
1.136 brouard 11271: /***********************************************/
11272: /**************** Main Program *****************/
11273: /***********************************************/
11274:
11275: int main(int argc, char *argv[])
11276: {
11277: #ifdef GSL
11278: const gsl_multimin_fminimizer_type *T;
11279: size_t iteri = 0, it;
11280: int rval = GSL_CONTINUE;
11281: int status = GSL_SUCCESS;
11282: double ssval;
11283: #endif
11284: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11285: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11286: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11287: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11288: int jj, ll, li, lj, lk;
1.136 brouard 11289: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11290: int num_filled;
1.136 brouard 11291: int itimes;
11292: int NDIM=2;
11293: int vpopbased=0;
1.235 brouard 11294: int nres=0;
1.258 brouard 11295: int endishere=0;
1.277 brouard 11296: int noffset=0;
1.274 brouard 11297: int ncurrv=0; /* Temporary variable */
11298:
1.164 brouard 11299: char ca[32], cb[32];
1.136 brouard 11300: /* FILE *fichtm; *//* Html File */
11301: /* FILE *ficgp;*/ /*Gnuplot File */
11302: struct stat info;
1.191 brouard 11303: double agedeb=0.;
1.194 brouard 11304:
11305: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11306: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11307:
1.165 brouard 11308: double fret;
1.191 brouard 11309: double dum=0.; /* Dummy variable */
1.136 brouard 11310: double ***p3mat;
1.218 brouard 11311: /* double ***mobaverage; */
1.319 brouard 11312: double wald;
1.164 brouard 11313:
11314: char line[MAXLINE];
1.197 brouard 11315: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11316:
1.234 brouard 11317: char modeltemp[MAXLINE];
1.230 brouard 11318: char resultline[MAXLINE];
11319:
1.136 brouard 11320: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11321: char *tok, *val; /* pathtot */
1.290 brouard 11322: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 11323: int c, h , cpt, c2;
1.191 brouard 11324: int jl=0;
11325: int i1, j1, jk, stepsize=0;
1.194 brouard 11326: int count=0;
11327:
1.164 brouard 11328: int *tab;
1.136 brouard 11329: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11330: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11331: /* double anprojf, mprojf, jprojf; */
11332: /* double jintmean,mintmean,aintmean; */
11333: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11334: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11335: double yrfproj= 10.0; /* Number of years of forward projections */
11336: double yrbproj= 10.0; /* Number of years of backward projections */
11337: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11338: int mobilav=0,popforecast=0;
1.191 brouard 11339: int hstepm=0, nhstepm=0;
1.136 brouard 11340: int agemortsup;
11341: float sumlpop=0.;
11342: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11343: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11344:
1.191 brouard 11345: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11346: double ftolpl=FTOL;
11347: double **prlim;
1.217 brouard 11348: double **bprlim;
1.317 brouard 11349: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
11350: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11351: double ***paramstart; /* Matrix of starting parameter values */
11352: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11353: double **matcov; /* Matrix of covariance */
1.203 brouard 11354: double **hess; /* Hessian matrix */
1.136 brouard 11355: double ***delti3; /* Scale */
11356: double *delti; /* Scale */
11357: double ***eij, ***vareij;
11358: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11359:
1.136 brouard 11360: double *epj, vepp;
1.164 brouard 11361:
1.273 brouard 11362: double dateprev1, dateprev2;
1.296 brouard 11363: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11364: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11365:
1.217 brouard 11366:
1.136 brouard 11367: double **ximort;
1.145 brouard 11368: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11369: int *dcwave;
11370:
1.164 brouard 11371: char z[1]="c";
1.136 brouard 11372:
11373: /*char *strt;*/
11374: char strtend[80];
1.126 brouard 11375:
1.164 brouard 11376:
1.126 brouard 11377: /* setlocale (LC_ALL, ""); */
11378: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11379: /* textdomain (PACKAGE); */
11380: /* setlocale (LC_CTYPE, ""); */
11381: /* setlocale (LC_MESSAGES, ""); */
11382:
11383: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11384: rstart_time = time(NULL);
11385: /* (void) gettimeofday(&start_time,&tzp);*/
11386: start_time = *localtime(&rstart_time);
1.126 brouard 11387: curr_time=start_time;
1.157 brouard 11388: /*tml = *localtime(&start_time.tm_sec);*/
11389: /* strcpy(strstart,asctime(&tml)); */
11390: strcpy(strstart,asctime(&start_time));
1.126 brouard 11391:
11392: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11393: /* tp.tm_sec = tp.tm_sec +86400; */
11394: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11395: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11396: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11397: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11398: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11399: /* strt=asctime(&tmg); */
11400: /* printf("Time(after) =%s",strstart); */
11401: /* (void) time (&time_value);
11402: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11403: * tm = *localtime(&time_value);
11404: * strstart=asctime(&tm);
11405: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11406: */
11407:
11408: nberr=0; /* Number of errors and warnings */
11409: nbwarn=0;
1.184 brouard 11410: #ifdef WIN32
11411: _getcwd(pathcd, size);
11412: #else
1.126 brouard 11413: getcwd(pathcd, size);
1.184 brouard 11414: #endif
1.191 brouard 11415: syscompilerinfo(0);
1.196 brouard 11416: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11417: if(argc <=1){
11418: printf("\nEnter the parameter file name: ");
1.205 brouard 11419: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11420: printf("ERROR Empty parameter file name\n");
11421: goto end;
11422: }
1.126 brouard 11423: i=strlen(pathr);
11424: if(pathr[i-1]=='\n')
11425: pathr[i-1]='\0';
1.156 brouard 11426: i=strlen(pathr);
1.205 brouard 11427: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11428: pathr[i-1]='\0';
1.205 brouard 11429: }
11430: i=strlen(pathr);
11431: if( i==0 ){
11432: printf("ERROR Empty parameter file name\n");
11433: goto end;
11434: }
11435: for (tok = pathr; tok != NULL; ){
1.126 brouard 11436: printf("Pathr |%s|\n",pathr);
11437: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11438: printf("val= |%s| pathr=%s\n",val,pathr);
11439: strcpy (pathtot, val);
11440: if(pathr[0] == '\0') break; /* Dirty */
11441: }
11442: }
1.281 brouard 11443: else if (argc<=2){
11444: strcpy(pathtot,argv[1]);
11445: }
1.126 brouard 11446: else{
11447: strcpy(pathtot,argv[1]);
1.281 brouard 11448: strcpy(z,argv[2]);
11449: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11450: }
11451: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11452: /*cygwin_split_path(pathtot,path,optionfile);
11453: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11454: /* cutv(path,optionfile,pathtot,'\\');*/
11455:
11456: /* Split argv[0], imach program to get pathimach */
11457: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11458: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11459: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11460: /* strcpy(pathimach,argv[0]); */
11461: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11462: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11463: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11464: #ifdef WIN32
11465: _chdir(path); /* Can be a relative path */
11466: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11467: #else
1.126 brouard 11468: chdir(path); /* Can be a relative path */
1.184 brouard 11469: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11470: #endif
11471: printf("Current directory %s!\n",pathcd);
1.126 brouard 11472: strcpy(command,"mkdir ");
11473: strcat(command,optionfilefiname);
11474: if((outcmd=system(command)) != 0){
1.169 brouard 11475: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11476: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11477: /* fclose(ficlog); */
11478: /* exit(1); */
11479: }
11480: /* if((imk=mkdir(optionfilefiname))<0){ */
11481: /* perror("mkdir"); */
11482: /* } */
11483:
11484: /*-------- arguments in the command line --------*/
11485:
1.186 brouard 11486: /* Main Log file */
1.126 brouard 11487: strcat(filelog, optionfilefiname);
11488: strcat(filelog,".log"); /* */
11489: if((ficlog=fopen(filelog,"w"))==NULL) {
11490: printf("Problem with logfile %s\n",filelog);
11491: goto end;
11492: }
11493: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11494: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11495: fprintf(ficlog,"\nEnter the parameter file name: \n");
11496: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11497: path=%s \n\
11498: optionfile=%s\n\
11499: optionfilext=%s\n\
1.156 brouard 11500: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11501:
1.197 brouard 11502: syscompilerinfo(1);
1.167 brouard 11503:
1.126 brouard 11504: printf("Local time (at start):%s",strstart);
11505: fprintf(ficlog,"Local time (at start): %s",strstart);
11506: fflush(ficlog);
11507: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11508: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11509:
11510: /* */
11511: strcpy(fileres,"r");
11512: strcat(fileres, optionfilefiname);
1.201 brouard 11513: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11514: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11515: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11516:
1.186 brouard 11517: /* Main ---------arguments file --------*/
1.126 brouard 11518:
11519: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11520: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11521: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11522: fflush(ficlog);
1.149 brouard 11523: /* goto end; */
11524: exit(70);
1.126 brouard 11525: }
11526:
11527: strcpy(filereso,"o");
1.201 brouard 11528: strcat(filereso,fileresu);
1.126 brouard 11529: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11530: printf("Problem with Output resultfile: %s\n", filereso);
11531: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11532: fflush(ficlog);
11533: goto end;
11534: }
1.278 brouard 11535: /*-------- Rewriting parameter file ----------*/
11536: strcpy(rfileres,"r"); /* "Rparameterfile */
11537: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11538: strcat(rfileres,"."); /* */
11539: strcat(rfileres,optionfilext); /* Other files have txt extension */
11540: if((ficres =fopen(rfileres,"w"))==NULL) {
11541: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11542: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11543: fflush(ficlog);
11544: goto end;
11545: }
11546: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11547:
1.278 brouard 11548:
1.126 brouard 11549: /* Reads comments: lines beginning with '#' */
11550: numlinepar=0;
1.277 brouard 11551: /* Is it a BOM UTF-8 Windows file? */
11552: /* First parameter line */
1.197 brouard 11553: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11554: noffset=0;
11555: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11556: {
11557: noffset=noffset+3;
11558: printf("# File is an UTF8 Bom.\n"); // 0xBF
11559: }
1.302 brouard 11560: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11561: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11562: {
11563: noffset=noffset+2;
11564: printf("# File is an UTF16BE BOM file\n");
11565: }
11566: else if( line[0] == 0 && line[1] == 0)
11567: {
11568: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11569: noffset=noffset+4;
11570: printf("# File is an UTF16BE BOM file\n");
11571: }
11572: } else{
11573: ;/*printf(" Not a BOM file\n");*/
11574: }
11575:
1.197 brouard 11576: /* If line starts with a # it is a comment */
1.277 brouard 11577: if (line[noffset] == '#') {
1.197 brouard 11578: numlinepar++;
11579: fputs(line,stdout);
11580: fputs(line,ficparo);
1.278 brouard 11581: fputs(line,ficres);
1.197 brouard 11582: fputs(line,ficlog);
11583: continue;
11584: }else
11585: break;
11586: }
11587: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11588: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11589: if (num_filled != 5) {
11590: printf("Should be 5 parameters\n");
1.283 brouard 11591: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11592: }
1.126 brouard 11593: numlinepar++;
1.197 brouard 11594: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11595: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11596: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11597: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11598: }
11599: /* Second parameter line */
11600: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11601: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11602: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11603: if (line[0] == '#') {
11604: numlinepar++;
1.283 brouard 11605: printf("%s",line);
11606: fprintf(ficres,"%s",line);
11607: fprintf(ficparo,"%s",line);
11608: fprintf(ficlog,"%s",line);
1.197 brouard 11609: continue;
11610: }else
11611: break;
11612: }
1.223 brouard 11613: 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", \
11614: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11615: if (num_filled != 11) {
11616: 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 11617: printf("but line=%s\n",line);
1.283 brouard 11618: 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");
11619: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11620: }
1.286 brouard 11621: if( lastpass > maxwav){
11622: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11623: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11624: fflush(ficlog);
11625: goto end;
11626: }
11627: 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 11628: 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 11629: 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 11630: 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 11631: }
1.203 brouard 11632: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11633: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11634: /* Third parameter line */
11635: while(fgets(line, MAXLINE, ficpar)) {
11636: /* If line starts with a # it is a comment */
11637: if (line[0] == '#') {
11638: numlinepar++;
1.283 brouard 11639: printf("%s",line);
11640: fprintf(ficres,"%s",line);
11641: fprintf(ficparo,"%s",line);
11642: fprintf(ficlog,"%s",line);
1.197 brouard 11643: continue;
11644: }else
11645: break;
11646: }
1.201 brouard 11647: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11648: if (num_filled != 1){
1.302 brouard 11649: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
11650: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11651: model[0]='\0';
11652: goto end;
11653: }
11654: else{
11655: if (model[0]=='+'){
11656: for(i=1; i<=strlen(model);i++)
11657: modeltemp[i-1]=model[i];
1.201 brouard 11658: strcpy(model,modeltemp);
1.197 brouard 11659: }
11660: }
1.199 brouard 11661: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11662: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11663: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11664: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11665: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11666: }
11667: /* 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); */
11668: /* numlinepar=numlinepar+3; /\* In general *\/ */
11669: /* 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 11670: /* 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); */
11671: /* 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 11672: fflush(ficlog);
1.190 brouard 11673: /* if(model[0]=='#'|| model[0]== '\0'){ */
11674: if(model[0]=='#'){
1.279 brouard 11675: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11676: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11677: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11678: if(mle != -1){
1.279 brouard 11679: 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 11680: exit(1);
11681: }
11682: }
1.126 brouard 11683: while((c=getc(ficpar))=='#' && c!= EOF){
11684: ungetc(c,ficpar);
11685: fgets(line, MAXLINE, ficpar);
11686: numlinepar++;
1.195 brouard 11687: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11688: z[0]=line[1];
11689: }
11690: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11691: fputs(line, stdout);
11692: //puts(line);
1.126 brouard 11693: fputs(line,ficparo);
11694: fputs(line,ficlog);
11695: }
11696: ungetc(c,ficpar);
11697:
11698:
1.290 brouard 11699: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11700: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11701: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11702: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11703: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11704: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11705: v1+v2*age+v2*v3 makes cptcovn = 3
11706: */
11707: if (strlen(model)>1)
1.187 brouard 11708: 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 11709: else
1.187 brouard 11710: ncovmodel=2; /* Constant and age */
1.133 brouard 11711: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11712: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11713: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11714: 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);
11715: 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);
11716: fflush(stdout);
11717: fclose (ficlog);
11718: goto end;
11719: }
1.126 brouard 11720: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11721: delti=delti3[1][1];
11722: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11723: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11724: /* We could also provide initial parameters values giving by simple logistic regression
11725: * only one way, that is without matrix product. We will have nlstate maximizations */
11726: /* for(i=1;i<nlstate;i++){ */
11727: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11728: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11729: /* } */
1.126 brouard 11730: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11731: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11732: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11733: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11734: fclose (ficparo);
11735: fclose (ficlog);
11736: goto end;
11737: exit(0);
1.220 brouard 11738: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11739: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11740: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11741: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11742: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11743: matcov=matrix(1,npar,1,npar);
1.203 brouard 11744: hess=matrix(1,npar,1,npar);
1.220 brouard 11745: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11746: /* Read guessed parameters */
1.126 brouard 11747: /* Reads comments: lines beginning with '#' */
11748: while((c=getc(ficpar))=='#' && c!= EOF){
11749: ungetc(c,ficpar);
11750: fgets(line, MAXLINE, ficpar);
11751: numlinepar++;
1.141 brouard 11752: fputs(line,stdout);
1.126 brouard 11753: fputs(line,ficparo);
11754: fputs(line,ficlog);
11755: }
11756: ungetc(c,ficpar);
11757:
11758: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11759: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11760: for(i=1; i <=nlstate; i++){
1.234 brouard 11761: j=0;
1.126 brouard 11762: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11763: if(jj==i) continue;
11764: j++;
1.292 brouard 11765: while((c=getc(ficpar))=='#' && c!= EOF){
11766: ungetc(c,ficpar);
11767: fgets(line, MAXLINE, ficpar);
11768: numlinepar++;
11769: fputs(line,stdout);
11770: fputs(line,ficparo);
11771: fputs(line,ficlog);
11772: }
11773: ungetc(c,ficpar);
1.234 brouard 11774: fscanf(ficpar,"%1d%1d",&i1,&j1);
11775: if ((i1 != i) || (j1 != jj)){
11776: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11777: It might be a problem of design; if ncovcol and the model are correct\n \
11778: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11779: exit(1);
11780: }
11781: fprintf(ficparo,"%1d%1d",i1,j1);
11782: if(mle==1)
11783: printf("%1d%1d",i,jj);
11784: fprintf(ficlog,"%1d%1d",i,jj);
11785: for(k=1; k<=ncovmodel;k++){
11786: fscanf(ficpar," %lf",¶m[i][j][k]);
11787: if(mle==1){
11788: printf(" %lf",param[i][j][k]);
11789: fprintf(ficlog," %lf",param[i][j][k]);
11790: }
11791: else
11792: fprintf(ficlog," %lf",param[i][j][k]);
11793: fprintf(ficparo," %lf",param[i][j][k]);
11794: }
11795: fscanf(ficpar,"\n");
11796: numlinepar++;
11797: if(mle==1)
11798: printf("\n");
11799: fprintf(ficlog,"\n");
11800: fprintf(ficparo,"\n");
1.126 brouard 11801: }
11802: }
11803: fflush(ficlog);
1.234 brouard 11804:
1.251 brouard 11805: /* Reads parameters values */
1.126 brouard 11806: p=param[1][1];
1.251 brouard 11807: pstart=paramstart[1][1];
1.126 brouard 11808:
11809: /* Reads comments: lines beginning with '#' */
11810: while((c=getc(ficpar))=='#' && c!= EOF){
11811: ungetc(c,ficpar);
11812: fgets(line, MAXLINE, ficpar);
11813: numlinepar++;
1.141 brouard 11814: fputs(line,stdout);
1.126 brouard 11815: fputs(line,ficparo);
11816: fputs(line,ficlog);
11817: }
11818: ungetc(c,ficpar);
11819:
11820: for(i=1; i <=nlstate; i++){
11821: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11822: fscanf(ficpar,"%1d%1d",&i1,&j1);
11823: if ( (i1-i) * (j1-j) != 0){
11824: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11825: exit(1);
11826: }
11827: printf("%1d%1d",i,j);
11828: fprintf(ficparo,"%1d%1d",i1,j1);
11829: fprintf(ficlog,"%1d%1d",i1,j1);
11830: for(k=1; k<=ncovmodel;k++){
11831: fscanf(ficpar,"%le",&delti3[i][j][k]);
11832: printf(" %le",delti3[i][j][k]);
11833: fprintf(ficparo," %le",delti3[i][j][k]);
11834: fprintf(ficlog," %le",delti3[i][j][k]);
11835: }
11836: fscanf(ficpar,"\n");
11837: numlinepar++;
11838: printf("\n");
11839: fprintf(ficparo,"\n");
11840: fprintf(ficlog,"\n");
1.126 brouard 11841: }
11842: }
11843: fflush(ficlog);
1.234 brouard 11844:
1.145 brouard 11845: /* Reads covariance matrix */
1.126 brouard 11846: delti=delti3[1][1];
1.220 brouard 11847:
11848:
1.126 brouard 11849: /* 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 11850:
1.126 brouard 11851: /* Reads comments: lines beginning with '#' */
11852: while((c=getc(ficpar))=='#' && c!= EOF){
11853: ungetc(c,ficpar);
11854: fgets(line, MAXLINE, ficpar);
11855: numlinepar++;
1.141 brouard 11856: fputs(line,stdout);
1.126 brouard 11857: fputs(line,ficparo);
11858: fputs(line,ficlog);
11859: }
11860: ungetc(c,ficpar);
1.220 brouard 11861:
1.126 brouard 11862: matcov=matrix(1,npar,1,npar);
1.203 brouard 11863: hess=matrix(1,npar,1,npar);
1.131 brouard 11864: for(i=1; i <=npar; i++)
11865: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11866:
1.194 brouard 11867: /* Scans npar lines */
1.126 brouard 11868: for(i=1; i <=npar; i++){
1.226 brouard 11869: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11870: if(count != 3){
1.226 brouard 11871: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11872: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11873: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11874: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11875: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11876: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11877: exit(1);
1.220 brouard 11878: }else{
1.226 brouard 11879: if(mle==1)
11880: printf("%1d%1d%d",i1,j1,jk);
11881: }
11882: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11883: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11884: for(j=1; j <=i; j++){
1.226 brouard 11885: fscanf(ficpar," %le",&matcov[i][j]);
11886: if(mle==1){
11887: printf(" %.5le",matcov[i][j]);
11888: }
11889: fprintf(ficlog," %.5le",matcov[i][j]);
11890: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11891: }
11892: fscanf(ficpar,"\n");
11893: numlinepar++;
11894: if(mle==1)
1.220 brouard 11895: printf("\n");
1.126 brouard 11896: fprintf(ficlog,"\n");
11897: fprintf(ficparo,"\n");
11898: }
1.194 brouard 11899: /* End of read covariance matrix npar lines */
1.126 brouard 11900: for(i=1; i <=npar; i++)
11901: for(j=i+1;j<=npar;j++)
1.226 brouard 11902: matcov[i][j]=matcov[j][i];
1.126 brouard 11903:
11904: if(mle==1)
11905: printf("\n");
11906: fprintf(ficlog,"\n");
11907:
11908: fflush(ficlog);
11909:
11910: } /* End of mle != -3 */
1.218 brouard 11911:
1.186 brouard 11912: /* Main data
11913: */
1.290 brouard 11914: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11915: /* num=lvector(1,n); */
11916: /* moisnais=vector(1,n); */
11917: /* annais=vector(1,n); */
11918: /* moisdc=vector(1,n); */
11919: /* andc=vector(1,n); */
11920: /* weight=vector(1,n); */
11921: /* agedc=vector(1,n); */
11922: /* cod=ivector(1,n); */
11923: /* for(i=1;i<=n;i++){ */
11924: num=lvector(firstobs,lastobs);
11925: moisnais=vector(firstobs,lastobs);
11926: annais=vector(firstobs,lastobs);
11927: moisdc=vector(firstobs,lastobs);
11928: andc=vector(firstobs,lastobs);
11929: weight=vector(firstobs,lastobs);
11930: agedc=vector(firstobs,lastobs);
11931: cod=ivector(firstobs,lastobs);
11932: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11933: num[i]=0;
11934: moisnais[i]=0;
11935: annais[i]=0;
11936: moisdc[i]=0;
11937: andc[i]=0;
11938: agedc[i]=0;
11939: cod[i]=0;
11940: weight[i]=1.0; /* Equal weights, 1 by default */
11941: }
1.290 brouard 11942: mint=matrix(1,maxwav,firstobs,lastobs);
11943: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 11944: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
11945: printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel));
1.126 brouard 11946: tab=ivector(1,NCOVMAX);
1.144 brouard 11947: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11948: 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 11949:
1.136 brouard 11950: /* Reads data from file datafile */
11951: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11952: goto end;
11953:
11954: /* Calculation of the number of parameters from char model */
1.234 brouard 11955: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11956: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11957: k=3 V4 Tvar[k=3]= 4 (from V4)
11958: k=2 V1 Tvar[k=2]= 1 (from V1)
11959: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11960: */
11961:
11962: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11963: TvarsDind=ivector(1,NCOVMAX); /* */
11964: TvarsD=ivector(1,NCOVMAX); /* */
11965: TvarsQind=ivector(1,NCOVMAX); /* */
11966: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11967: TvarF=ivector(1,NCOVMAX); /* */
11968: TvarFind=ivector(1,NCOVMAX); /* */
11969: TvarV=ivector(1,NCOVMAX); /* */
11970: TvarVind=ivector(1,NCOVMAX); /* */
11971: TvarA=ivector(1,NCOVMAX); /* */
11972: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11973: TvarFD=ivector(1,NCOVMAX); /* */
11974: TvarFDind=ivector(1,NCOVMAX); /* */
11975: TvarFQ=ivector(1,NCOVMAX); /* */
11976: TvarFQind=ivector(1,NCOVMAX); /* */
11977: TvarVD=ivector(1,NCOVMAX); /* */
11978: TvarVDind=ivector(1,NCOVMAX); /* */
11979: TvarVQ=ivector(1,NCOVMAX); /* */
11980: TvarVQind=ivector(1,NCOVMAX); /* */
11981:
1.230 brouard 11982: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11983: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11984: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11985: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11986: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11987: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11988: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11989: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11990: */
11991: /* For model-covariate k tells which data-covariate to use but
11992: because this model-covariate is a construction we invent a new column
11993: ncovcol + k1
11994: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11995: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11996: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11997: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11998: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11999: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 12000: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 12001: */
1.145 brouard 12002: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
12003: 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 12004: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
12005: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 12006: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 12007: 4 covariates (3 plus signs)
12008: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 ! brouard 12009: */
! 12010: for(i=1;i<NCOVMAX;i++)
! 12011: Tage[i]=0;
1.230 brouard 12012: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 12013: * individual dummy, fixed or varying:
12014: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
12015: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 12016: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
12017: * V1 df, V2 qf, V3 & V4 dv, V5 qv
12018: * Tmodelind[1]@9={9,0,3,2,}*/
12019: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
12020: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 12021: * individual quantitative, fixed or varying:
12022: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
12023: * 3, 1, 0, 0, 0, 0, 0, 0},
12024: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 12025: /* Main decodemodel */
12026:
1.187 brouard 12027:
1.223 brouard 12028: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 12029: goto end;
12030:
1.137 brouard 12031: if((double)(lastobs-imx)/(double)imx > 1.10){
12032: nbwarn++;
12033: 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);
12034: 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);
12035: }
1.136 brouard 12036: /* if(mle==1){*/
1.137 brouard 12037: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
12038: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 12039: }
12040:
12041: /*-calculation of age at interview from date of interview and age at death -*/
12042: agev=matrix(1,maxwav,1,imx);
12043:
12044: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
12045: goto end;
12046:
1.126 brouard 12047:
1.136 brouard 12048: agegomp=(int)agemin;
1.290 brouard 12049: free_vector(moisnais,firstobs,lastobs);
12050: free_vector(annais,firstobs,lastobs);
1.126 brouard 12051: /* free_matrix(mint,1,maxwav,1,n);
12052: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 12053: /* free_vector(moisdc,1,n); */
12054: /* free_vector(andc,1,n); */
1.145 brouard 12055: /* */
12056:
1.126 brouard 12057: wav=ivector(1,imx);
1.214 brouard 12058: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
12059: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
12060: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
12061: 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.*/
12062: bh=imatrix(1,lastpass-firstpass+2,1,imx);
12063: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 12064:
12065: /* Concatenates waves */
1.214 brouard 12066: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
12067: Death is a valid wave (if date is known).
12068: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
12069: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
12070: and mw[mi+1][i]. dh depends on stepm.
12071: */
12072:
1.126 brouard 12073: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 12074: /* Concatenates waves */
1.145 brouard 12075:
1.290 brouard 12076: free_vector(moisdc,firstobs,lastobs);
12077: free_vector(andc,firstobs,lastobs);
1.215 brouard 12078:
1.126 brouard 12079: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12080: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12081: ncodemax[1]=1;
1.145 brouard 12082: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12083: cptcoveff=0;
1.220 brouard 12084: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
12085: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 12086: }
12087:
12088: ncovcombmax=pow(2,cptcoveff);
12089: invalidvarcomb=ivector(1, ncovcombmax);
12090: for(i=1;i<ncovcombmax;i++)
12091: invalidvarcomb[i]=0;
12092:
1.211 brouard 12093: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12094: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12095: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12096:
1.200 brouard 12097: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12098: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12099: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12100: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12101: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12102: * (currently 0 or 1) in the data.
12103: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12104: * corresponding modality (h,j).
12105: */
12106:
1.145 brouard 12107: h=0;
12108: /*if (cptcovn > 0) */
1.126 brouard 12109: m=pow(2,cptcoveff);
12110:
1.144 brouard 12111: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12112: * For k=4 covariates, h goes from 1 to m=2**k
12113: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12114: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 12115: * h\k 1 2 3 4
1.143 brouard 12116: *______________________________
12117: * 1 i=1 1 i=1 1 i=1 1 i=1 1
12118: * 2 2 1 1 1
12119: * 3 i=2 1 2 1 1
12120: * 4 2 2 1 1
12121: * 5 i=3 1 i=2 1 2 1
12122: * 6 2 1 2 1
12123: * 7 i=4 1 2 2 1
12124: * 8 2 2 2 1
1.197 brouard 12125: * 9 i=5 1 i=3 1 i=2 1 2
12126: * 10 2 1 1 2
12127: * 11 i=6 1 2 1 2
12128: * 12 2 2 1 2
12129: * 13 i=7 1 i=4 1 2 2
12130: * 14 2 1 2 2
12131: * 15 i=8 1 2 2 2
12132: * 16 2 2 2 2
1.143 brouard 12133: */
1.212 brouard 12134: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 12135: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
12136: * and the value of each covariate?
12137: * V1=1, V2=1, V3=2, V4=1 ?
12138: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
12139: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
12140: * In order to get the real value in the data, we use nbcode
12141: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
12142: * We are keeping this crazy system in order to be able (in the future?)
12143: * to have more than 2 values (0 or 1) for a covariate.
12144: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
12145: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
12146: * bbbbbbbb
12147: * 76543210
12148: * h-1 00000101 (6-1=5)
1.219 brouard 12149: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 12150: * &
12151: * 1 00000001 (1)
1.219 brouard 12152: * 00000000 = 1 & ((h-1) >> (k-1))
12153: * +1= 00000001 =1
1.211 brouard 12154: *
12155: * h=14, k=3 => h'=h-1=13, k'=k-1=2
12156: * h' 1101 =2^3+2^2+0x2^1+2^0
12157: * >>k' 11
12158: * & 00000001
12159: * = 00000001
12160: * +1 = 00000010=2 = codtabm(14,3)
12161: * Reverse h=6 and m=16?
12162: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
12163: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
12164: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
12165: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
12166: * V3=decodtabm(14,3,2**4)=2
12167: * h'=13 1101 =2^3+2^2+0x2^1+2^0
12168: *(h-1) >> (j-1) 0011 =13 >> 2
12169: * &1 000000001
12170: * = 000000001
12171: * +1= 000000010 =2
12172: * 2211
12173: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
12174: * V3=2
1.220 brouard 12175: * codtabm and decodtabm are identical
1.211 brouard 12176: */
12177:
1.145 brouard 12178:
12179: free_ivector(Ndum,-1,NCOVMAX);
12180:
12181:
1.126 brouard 12182:
1.186 brouard 12183: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 12184: strcpy(optionfilegnuplot,optionfilefiname);
12185: if(mle==-3)
1.201 brouard 12186: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 12187: strcat(optionfilegnuplot,".gp");
12188:
12189: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
12190: printf("Problem with file %s",optionfilegnuplot);
12191: }
12192: else{
1.204 brouard 12193: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 12194: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 12195: //fprintf(ficgp,"set missing 'NaNq'\n");
12196: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 12197: }
12198: /* fclose(ficgp);*/
1.186 brouard 12199:
12200:
12201: /* Initialisation of --------- index.htm --------*/
1.126 brouard 12202:
12203: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
12204: if(mle==-3)
1.201 brouard 12205: strcat(optionfilehtm,"-MORT_");
1.126 brouard 12206: strcat(optionfilehtm,".htm");
12207: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 12208: printf("Problem with %s \n",optionfilehtm);
12209: exit(0);
1.126 brouard 12210: }
12211:
12212: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
12213: strcat(optionfilehtmcov,"-cov.htm");
12214: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
12215: printf("Problem with %s \n",optionfilehtmcov), exit(0);
12216: }
12217: else{
12218: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
12219: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12220: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 12221: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
12222: }
12223:
1.324 brouard 12224: 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 12225: <hr size=\"2\" color=\"#EC5E5E\"> \n\
12226: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 12227: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12228: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 12229: \n\
12230: <hr size=\"2\" color=\"#EC5E5E\">\
12231: <ul><li><h4>Parameter files</h4>\n\
12232: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
12233: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
12234: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
12235: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
12236: - Date and time at start: %s</ul>\n",\
12237: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
12238: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
12239: fileres,fileres,\
12240: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
12241: fflush(fichtm);
12242:
12243: strcpy(pathr,path);
12244: strcat(pathr,optionfilefiname);
1.184 brouard 12245: #ifdef WIN32
12246: _chdir(optionfilefiname); /* Move to directory named optionfile */
12247: #else
1.126 brouard 12248: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12249: #endif
12250:
1.126 brouard 12251:
1.220 brouard 12252: /* Calculates basic frequencies. Computes observed prevalence at single age
12253: and for any valid combination of covariates
1.126 brouard 12254: and prints on file fileres'p'. */
1.251 brouard 12255: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12256: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12257:
12258: fprintf(fichtm,"\n");
1.286 brouard 12259: 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 12260: ftol, stepm);
12261: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12262: ncurrv=1;
12263: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12264: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12265: ncurrv=i;
12266: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12267: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12268: ncurrv=i;
12269: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12270: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12271: ncurrv=i;
12272: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12273: 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", \
12274: nlstate, ndeath, maxwav, mle, weightopt);
12275:
12276: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12277: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12278:
12279:
1.317 brouard 12280: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12281: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12282: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12283: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12284: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12285: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12286: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12287: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12288: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12289:
1.126 brouard 12290: /* For Powell, parameters are in a vector p[] starting at p[1]
12291: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12292: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12293:
12294: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12295: /* For mortality only */
1.126 brouard 12296: if (mle==-3){
1.136 brouard 12297: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12298: for(i=1;i<=NDIM;i++)
12299: for(j=1;j<=NDIM;j++)
12300: ximort[i][j]=0.;
1.186 brouard 12301: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12302: cens=ivector(firstobs,lastobs);
12303: ageexmed=vector(firstobs,lastobs);
12304: agecens=vector(firstobs,lastobs);
12305: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12306:
1.126 brouard 12307: for (i=1; i<=imx; i++){
12308: dcwave[i]=-1;
12309: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12310: if (s[m][i]>nlstate) {
12311: dcwave[i]=m;
12312: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12313: break;
12314: }
1.126 brouard 12315: }
1.226 brouard 12316:
1.126 brouard 12317: for (i=1; i<=imx; i++) {
12318: if (wav[i]>0){
1.226 brouard 12319: ageexmed[i]=agev[mw[1][i]][i];
12320: j=wav[i];
12321: agecens[i]=1.;
12322:
12323: if (ageexmed[i]> 1 && wav[i] > 0){
12324: agecens[i]=agev[mw[j][i]][i];
12325: cens[i]= 1;
12326: }else if (ageexmed[i]< 1)
12327: cens[i]= -1;
12328: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12329: cens[i]=0 ;
1.126 brouard 12330: }
12331: else cens[i]=-1;
12332: }
12333:
12334: for (i=1;i<=NDIM;i++) {
12335: for (j=1;j<=NDIM;j++)
1.226 brouard 12336: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12337: }
12338:
1.302 brouard 12339: p[1]=0.0268; p[NDIM]=0.083;
12340: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12341:
12342:
1.136 brouard 12343: #ifdef GSL
12344: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12345: #else
1.126 brouard 12346: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12347: #endif
1.201 brouard 12348: strcpy(filerespow,"POW-MORT_");
12349: strcat(filerespow,fileresu);
1.126 brouard 12350: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12351: printf("Problem with resultfile: %s\n", filerespow);
12352: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12353: }
1.136 brouard 12354: #ifdef GSL
12355: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12356: #else
1.126 brouard 12357: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12358: #endif
1.126 brouard 12359: /* for (i=1;i<=nlstate;i++)
12360: for(j=1;j<=nlstate+ndeath;j++)
12361: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12362: */
12363: fprintf(ficrespow,"\n");
1.136 brouard 12364: #ifdef GSL
12365: /* gsl starts here */
12366: T = gsl_multimin_fminimizer_nmsimplex;
12367: gsl_multimin_fminimizer *sfm = NULL;
12368: gsl_vector *ss, *x;
12369: gsl_multimin_function minex_func;
12370:
12371: /* Initial vertex size vector */
12372: ss = gsl_vector_alloc (NDIM);
12373:
12374: if (ss == NULL){
12375: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12376: }
12377: /* Set all step sizes to 1 */
12378: gsl_vector_set_all (ss, 0.001);
12379:
12380: /* Starting point */
1.126 brouard 12381:
1.136 brouard 12382: x = gsl_vector_alloc (NDIM);
12383:
12384: if (x == NULL){
12385: gsl_vector_free(ss);
12386: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12387: }
12388:
12389: /* Initialize method and iterate */
12390: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12391: /* gsl_vector_set(x, 0, 0.0268); */
12392: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12393: gsl_vector_set(x, 0, p[1]);
12394: gsl_vector_set(x, 1, p[2]);
12395:
12396: minex_func.f = &gompertz_f;
12397: minex_func.n = NDIM;
12398: minex_func.params = (void *)&p; /* ??? */
12399:
12400: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12401: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12402:
12403: printf("Iterations beginning .....\n\n");
12404: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12405:
12406: iteri=0;
12407: while (rval == GSL_CONTINUE){
12408: iteri++;
12409: status = gsl_multimin_fminimizer_iterate(sfm);
12410:
12411: if (status) printf("error: %s\n", gsl_strerror (status));
12412: fflush(0);
12413:
12414: if (status)
12415: break;
12416:
12417: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12418: ssval = gsl_multimin_fminimizer_size (sfm);
12419:
12420: if (rval == GSL_SUCCESS)
12421: printf ("converged to a local maximum at\n");
12422:
12423: printf("%5d ", iteri);
12424: for (it = 0; it < NDIM; it++){
12425: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12426: }
12427: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12428: }
12429:
12430: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12431:
12432: gsl_vector_free(x); /* initial values */
12433: gsl_vector_free(ss); /* inital step size */
12434: for (it=0; it<NDIM; it++){
12435: p[it+1]=gsl_vector_get(sfm->x,it);
12436: fprintf(ficrespow," %.12lf", p[it]);
12437: }
12438: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12439: #endif
12440: #ifdef POWELL
12441: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12442: #endif
1.126 brouard 12443: fclose(ficrespow);
12444:
1.203 brouard 12445: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12446:
12447: for(i=1; i <=NDIM; i++)
12448: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12449: matcov[i][j]=matcov[j][i];
1.126 brouard 12450:
12451: printf("\nCovariance matrix\n ");
1.203 brouard 12452: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12453: for(i=1; i <=NDIM; i++) {
12454: for(j=1;j<=NDIM;j++){
1.220 brouard 12455: printf("%f ",matcov[i][j]);
12456: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12457: }
1.203 brouard 12458: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12459: }
12460:
12461: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12462: for (i=1;i<=NDIM;i++) {
1.126 brouard 12463: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12464: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12465: }
1.302 brouard 12466: lsurv=vector(agegomp,AGESUP);
12467: lpop=vector(agegomp,AGESUP);
12468: tpop=vector(agegomp,AGESUP);
1.126 brouard 12469: lsurv[agegomp]=100000;
12470:
12471: for (k=agegomp;k<=AGESUP;k++) {
12472: agemortsup=k;
12473: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12474: }
12475:
12476: for (k=agegomp;k<agemortsup;k++)
12477: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12478:
12479: for (k=agegomp;k<agemortsup;k++){
12480: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12481: sumlpop=sumlpop+lpop[k];
12482: }
12483:
12484: tpop[agegomp]=sumlpop;
12485: for (k=agegomp;k<(agemortsup-3);k++){
12486: /* tpop[k+1]=2;*/
12487: tpop[k+1]=tpop[k]-lpop[k];
12488: }
12489:
12490:
12491: printf("\nAge lx qx dx Lx Tx e(x)\n");
12492: for (k=agegomp;k<(agemortsup-2);k++)
12493: 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]);
12494:
12495:
12496: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12497: ageminpar=50;
12498: agemaxpar=100;
1.194 brouard 12499: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12500: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12501: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12502: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12503: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12504: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12505: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12506: }else{
12507: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12508: 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 12509: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12510: }
1.201 brouard 12511: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12512: stepm, weightopt,\
12513: model,imx,p,matcov,agemortsup);
12514:
1.302 brouard 12515: free_vector(lsurv,agegomp,AGESUP);
12516: free_vector(lpop,agegomp,AGESUP);
12517: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12518: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12519: free_ivector(dcwave,firstobs,lastobs);
12520: free_vector(agecens,firstobs,lastobs);
12521: free_vector(ageexmed,firstobs,lastobs);
12522: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12523: #ifdef GSL
1.136 brouard 12524: #endif
1.186 brouard 12525: } /* Endof if mle==-3 mortality only */
1.205 brouard 12526: /* Standard */
12527: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12528: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12529: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12530: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12531: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12532: for (k=1; k<=npar;k++)
12533: printf(" %d %8.5f",k,p[k]);
12534: printf("\n");
1.205 brouard 12535: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12536: /* mlikeli uses func not funcone */
1.247 brouard 12537: /* for(i=1;i<nlstate;i++){ */
12538: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12539: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12540: /* } */
1.205 brouard 12541: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12542: }
12543: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12544: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12545: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12546: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12547: }
12548: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12549: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12550: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12551: for (k=1; k<=npar;k++)
12552: printf(" %d %8.5f",k,p[k]);
12553: printf("\n");
12554:
12555: /*--------- results files --------------*/
1.283 brouard 12556: /* 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 12557:
12558:
12559: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12560: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 12561: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12562:
12563: printf("#model= 1 + age ");
12564: fprintf(ficres,"#model= 1 + age ");
12565: fprintf(ficlog,"#model= 1 + age ");
12566: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
12567: </ul>", model);
12568:
12569: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
12570: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
12571: if(nagesqr==1){
12572: printf(" + age*age ");
12573: fprintf(ficres," + age*age ");
12574: fprintf(ficlog," + age*age ");
12575: fprintf(fichtm, "<th>+ age*age</th>");
12576: }
12577: for(j=1;j <=ncovmodel-2;j++){
12578: if(Typevar[j]==0) {
12579: printf(" + V%d ",Tvar[j]);
12580: fprintf(ficres," + V%d ",Tvar[j]);
12581: fprintf(ficlog," + V%d ",Tvar[j]);
12582: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12583: }else if(Typevar[j]==1) {
12584: printf(" + V%d*age ",Tvar[j]);
12585: fprintf(ficres," + V%d*age ",Tvar[j]);
12586: fprintf(ficlog," + V%d*age ",Tvar[j]);
12587: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12588: }else if(Typevar[j]==2) {
12589: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12590: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12591: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12592: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12593: }
12594: }
12595: printf("\n");
12596: fprintf(ficres,"\n");
12597: fprintf(ficlog,"\n");
12598: fprintf(fichtm, "</tr>");
12599: fprintf(fichtm, "\n");
12600:
12601:
1.126 brouard 12602: for(i=1,jk=1; i <=nlstate; i++){
12603: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12604: if (k != i) {
1.319 brouard 12605: fprintf(fichtm, "<tr>");
1.225 brouard 12606: printf("%d%d ",i,k);
12607: fprintf(ficlog,"%d%d ",i,k);
12608: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 12609: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12610: for(j=1; j <=ncovmodel; j++){
12611: printf("%12.7f ",p[jk]);
12612: fprintf(ficlog,"%12.7f ",p[jk]);
12613: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 12614: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 12615: jk++;
12616: }
12617: printf("\n");
12618: fprintf(ficlog,"\n");
12619: fprintf(ficres,"\n");
1.319 brouard 12620: fprintf(fichtm, "</tr>\n");
1.225 brouard 12621: }
1.126 brouard 12622: }
12623: }
1.319 brouard 12624: /* fprintf(fichtm,"</tr>\n"); */
12625: fprintf(fichtm,"</table>\n");
12626: fprintf(fichtm, "\n");
12627:
1.203 brouard 12628: if(mle != 0){
12629: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12630: ftolhess=ftol; /* Usually correct */
1.203 brouard 12631: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12632: 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");
12633: 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 12634: 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 12635: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
12636: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
12637: if(nagesqr==1){
12638: printf(" + age*age ");
12639: fprintf(ficres," + age*age ");
12640: fprintf(ficlog," + age*age ");
12641: fprintf(fichtm, "<th>+ age*age</th>");
12642: }
12643: for(j=1;j <=ncovmodel-2;j++){
12644: if(Typevar[j]==0) {
12645: printf(" + V%d ",Tvar[j]);
12646: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12647: }else if(Typevar[j]==1) {
12648: printf(" + V%d*age ",Tvar[j]);
12649: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12650: }else if(Typevar[j]==2) {
12651: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12652: }
12653: }
12654: fprintf(fichtm, "</tr>\n");
12655:
1.203 brouard 12656: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12657: for(k=1; k <=(nlstate+ndeath); k++){
12658: if (k != i) {
1.319 brouard 12659: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 12660: printf("%d%d ",i,k);
12661: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 12662: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12663: for(j=1; j <=ncovmodel; j++){
1.319 brouard 12664: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 12665: 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]));
12666: 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 12667: if(fabs(wald) > 1.96){
1.321 brouard 12668: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 12669: }else{
12670: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
12671: }
1.324 brouard 12672: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 12673: 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 12674: jk++;
12675: }
12676: printf("\n");
12677: fprintf(ficlog,"\n");
1.319 brouard 12678: fprintf(fichtm, "</tr>\n");
1.225 brouard 12679: }
12680: }
1.193 brouard 12681: }
1.203 brouard 12682: } /* end of hesscov and Wald tests */
1.319 brouard 12683: fprintf(fichtm,"</table>\n");
1.225 brouard 12684:
1.203 brouard 12685: /* */
1.126 brouard 12686: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12687: printf("# Scales (for hessian or gradient estimation)\n");
12688: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12689: for(i=1,jk=1; i <=nlstate; i++){
12690: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12691: if (j!=i) {
12692: fprintf(ficres,"%1d%1d",i,j);
12693: printf("%1d%1d",i,j);
12694: fprintf(ficlog,"%1d%1d",i,j);
12695: for(k=1; k<=ncovmodel;k++){
12696: printf(" %.5e",delti[jk]);
12697: fprintf(ficlog," %.5e",delti[jk]);
12698: fprintf(ficres," %.5e",delti[jk]);
12699: jk++;
12700: }
12701: printf("\n");
12702: fprintf(ficlog,"\n");
12703: fprintf(ficres,"\n");
12704: }
1.126 brouard 12705: }
12706: }
12707:
12708: 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 12709: if(mle >= 1) /* To big for the screen */
1.126 brouard 12710: 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");
12711: 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");
12712: /* # 121 Var(a12)\n\ */
12713: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12714: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12715: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12716: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12717: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12718: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12719: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12720:
12721:
12722: /* Just to have a covariance matrix which will be more understandable
12723: even is we still don't want to manage dictionary of variables
12724: */
12725: for(itimes=1;itimes<=2;itimes++){
12726: jj=0;
12727: for(i=1; i <=nlstate; i++){
1.225 brouard 12728: for(j=1; j <=nlstate+ndeath; j++){
12729: if(j==i) continue;
12730: for(k=1; k<=ncovmodel;k++){
12731: jj++;
12732: ca[0]= k+'a'-1;ca[1]='\0';
12733: if(itimes==1){
12734: if(mle>=1)
12735: printf("#%1d%1d%d",i,j,k);
12736: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12737: fprintf(ficres,"#%1d%1d%d",i,j,k);
12738: }else{
12739: if(mle>=1)
12740: printf("%1d%1d%d",i,j,k);
12741: fprintf(ficlog,"%1d%1d%d",i,j,k);
12742: fprintf(ficres,"%1d%1d%d",i,j,k);
12743: }
12744: ll=0;
12745: for(li=1;li <=nlstate; li++){
12746: for(lj=1;lj <=nlstate+ndeath; lj++){
12747: if(lj==li) continue;
12748: for(lk=1;lk<=ncovmodel;lk++){
12749: ll++;
12750: if(ll<=jj){
12751: cb[0]= lk +'a'-1;cb[1]='\0';
12752: if(ll<jj){
12753: if(itimes==1){
12754: if(mle>=1)
12755: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12756: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12757: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12758: }else{
12759: if(mle>=1)
12760: printf(" %.5e",matcov[jj][ll]);
12761: fprintf(ficlog," %.5e",matcov[jj][ll]);
12762: fprintf(ficres," %.5e",matcov[jj][ll]);
12763: }
12764: }else{
12765: if(itimes==1){
12766: if(mle>=1)
12767: printf(" Var(%s%1d%1d)",ca,i,j);
12768: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12769: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12770: }else{
12771: if(mle>=1)
12772: printf(" %.7e",matcov[jj][ll]);
12773: fprintf(ficlog," %.7e",matcov[jj][ll]);
12774: fprintf(ficres," %.7e",matcov[jj][ll]);
12775: }
12776: }
12777: }
12778: } /* end lk */
12779: } /* end lj */
12780: } /* end li */
12781: if(mle>=1)
12782: printf("\n");
12783: fprintf(ficlog,"\n");
12784: fprintf(ficres,"\n");
12785: numlinepar++;
12786: } /* end k*/
12787: } /*end j */
1.126 brouard 12788: } /* end i */
12789: } /* end itimes */
12790:
12791: fflush(ficlog);
12792: fflush(ficres);
1.225 brouard 12793: while(fgets(line, MAXLINE, ficpar)) {
12794: /* If line starts with a # it is a comment */
12795: if (line[0] == '#') {
12796: numlinepar++;
12797: fputs(line,stdout);
12798: fputs(line,ficparo);
12799: fputs(line,ficlog);
1.299 brouard 12800: fputs(line,ficres);
1.225 brouard 12801: continue;
12802: }else
12803: break;
12804: }
12805:
1.209 brouard 12806: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12807: /* ungetc(c,ficpar); */
12808: /* fgets(line, MAXLINE, ficpar); */
12809: /* fputs(line,stdout); */
12810: /* fputs(line,ficparo); */
12811: /* } */
12812: /* ungetc(c,ficpar); */
1.126 brouard 12813:
12814: estepm=0;
1.209 brouard 12815: 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 12816:
12817: if (num_filled != 6) {
12818: 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);
12819: 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);
12820: goto end;
12821: }
12822: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12823: }
12824: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12825: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12826:
1.209 brouard 12827: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12828: if (estepm==0 || estepm < stepm) estepm=stepm;
12829: if (fage <= 2) {
12830: bage = ageminpar;
12831: fage = agemaxpar;
12832: }
12833:
12834: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12835: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12836: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12837:
1.186 brouard 12838: /* Other stuffs, more or less useful */
1.254 brouard 12839: while(fgets(line, MAXLINE, ficpar)) {
12840: /* If line starts with a # it is a comment */
12841: if (line[0] == '#') {
12842: numlinepar++;
12843: fputs(line,stdout);
12844: fputs(line,ficparo);
12845: fputs(line,ficlog);
1.299 brouard 12846: fputs(line,ficres);
1.254 brouard 12847: continue;
12848: }else
12849: break;
12850: }
12851:
12852: 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){
12853:
12854: if (num_filled != 7) {
12855: 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);
12856: 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);
12857: goto end;
12858: }
12859: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12860: 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);
12861: 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);
12862: 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 12863: }
1.254 brouard 12864:
12865: while(fgets(line, MAXLINE, ficpar)) {
12866: /* If line starts with a # it is a comment */
12867: if (line[0] == '#') {
12868: numlinepar++;
12869: fputs(line,stdout);
12870: fputs(line,ficparo);
12871: fputs(line,ficlog);
1.299 brouard 12872: fputs(line,ficres);
1.254 brouard 12873: continue;
12874: }else
12875: break;
1.126 brouard 12876: }
12877:
12878:
12879: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12880: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12881:
1.254 brouard 12882: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12883: if (num_filled != 1) {
12884: 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);
12885: 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);
12886: goto end;
12887: }
12888: printf("pop_based=%d\n",popbased);
12889: fprintf(ficlog,"pop_based=%d\n",popbased);
12890: fprintf(ficparo,"pop_based=%d\n",popbased);
12891: fprintf(ficres,"pop_based=%d\n",popbased);
12892: }
12893:
1.258 brouard 12894: /* Results */
1.307 brouard 12895: endishere=0;
1.258 brouard 12896: nresult=0;
1.308 brouard 12897: parameterline=0;
1.258 brouard 12898: do{
12899: if(!fgets(line, MAXLINE, ficpar)){
12900: endishere=1;
1.308 brouard 12901: parameterline=15;
1.258 brouard 12902: }else if (line[0] == '#') {
12903: /* If line starts with a # it is a comment */
1.254 brouard 12904: numlinepar++;
12905: fputs(line,stdout);
12906: fputs(line,ficparo);
12907: fputs(line,ficlog);
1.299 brouard 12908: fputs(line,ficres);
1.254 brouard 12909: continue;
1.258 brouard 12910: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12911: parameterline=11;
1.296 brouard 12912: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12913: parameterline=12;
1.307 brouard 12914: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 12915: parameterline=13;
1.307 brouard 12916: }
1.258 brouard 12917: else{
12918: parameterline=14;
1.254 brouard 12919: }
1.308 brouard 12920: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 12921: case 11:
1.296 brouard 12922: 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)){
12923: 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 12924: 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);
12925: 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);
12926: 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);
12927: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12928: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12929: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12930: prvforecast = 1;
12931: }
12932: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 12933: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12934: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12935: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12936: prvforecast = 2;
12937: }
12938: else {
12939: 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);
12940: 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);
12941: goto end;
1.258 brouard 12942: }
1.254 brouard 12943: break;
1.258 brouard 12944: case 12:
1.296 brouard 12945: 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)){
12946: 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);
12947: 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);
12948: 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);
12949: 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);
12950: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12951: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12952: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12953: prvbackcast = 1;
12954: }
12955: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 12956: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12957: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12958: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12959: prvbackcast = 2;
12960: }
12961: else {
12962: 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);
12963: 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);
12964: goto end;
1.258 brouard 12965: }
1.230 brouard 12966: break;
1.258 brouard 12967: case 13:
1.307 brouard 12968: num_filled=sscanf(line,"result:%[^\n]\n",resultline);
12969: nresult++; /* Sum of resultlines */
12970: printf("Result %d: result:%s\n",nresult, resultline);
1.318 brouard 12971: if(nresult > MAXRESULTLINESPONE-1){
12972: 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);
12973: 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 12974: goto end;
12975: }
1.310 brouard 12976: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 12977: fprintf(ficparo,"result: %s\n",resultline);
12978: fprintf(ficres,"result: %s\n",resultline);
12979: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 12980: } else
12981: goto end;
1.307 brouard 12982: break;
12983: case 14:
12984: printf("Error: Unknown command '%s'\n",line);
12985: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 12986: if(line[0] == ' ' || line[0] == '\n'){
12987: printf("It should not be an empty line '%s'\n",line);
12988: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
12989: }
1.307 brouard 12990: if(ncovmodel >=2 && nresult==0 ){
12991: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
12992: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12993: }
1.307 brouard 12994: /* goto end; */
12995: break;
1.308 brouard 12996: case 15:
12997: printf("End of resultlines.\n");
12998: fprintf(ficlog,"End of resultlines.\n");
12999: break;
13000: default: /* parameterline =0 */
1.307 brouard 13001: nresult=1;
13002: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 13003: } /* End switch parameterline */
13004: }while(endishere==0); /* End do */
1.126 brouard 13005:
1.230 brouard 13006: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 13007: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 13008:
13009: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 13010: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 13011: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13012: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13013: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 13014: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13015: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13016: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13017: }else{
1.270 brouard 13018: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 13019: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
13020: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
13021: if(prvforecast==1){
13022: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
13023: jprojd=jproj1;
13024: mprojd=mproj1;
13025: anprojd=anproj1;
13026: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
13027: jprojf=jproj2;
13028: mprojf=mproj2;
13029: anprojf=anproj2;
13030: } else if(prvforecast == 2){
13031: dateprojd=dateintmean;
13032: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
13033: dateprojf=dateintmean+yrfproj;
13034: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
13035: }
13036: if(prvbackcast==1){
13037: datebackd=(jback1+12*mback1+365*anback1)/365;
13038: jbackd=jback1;
13039: mbackd=mback1;
13040: anbackd=anback1;
13041: datebackf=(jback2+12*mback2+365*anback2)/365;
13042: jbackf=jback2;
13043: mbackf=mback2;
13044: anbackf=anback2;
13045: } else if(prvbackcast == 2){
13046: datebackd=dateintmean;
13047: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
13048: datebackf=dateintmean-yrbproj;
13049: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
13050: }
13051:
13052: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 13053: }
13054: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 13055: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
13056: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 13057:
1.225 brouard 13058: /*------------ free_vector -------------*/
13059: /* chdir(path); */
1.220 brouard 13060:
1.215 brouard 13061: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
13062: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
13063: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
13064: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 13065: free_lvector(num,firstobs,lastobs);
13066: free_vector(agedc,firstobs,lastobs);
1.126 brouard 13067: /*free_matrix(covar,0,NCOVMAX,1,n);*/
13068: /*free_matrix(covar,1,NCOVMAX,1,n);*/
13069: fclose(ficparo);
13070: fclose(ficres);
1.220 brouard 13071:
13072:
1.186 brouard 13073: /* Other results (useful)*/
1.220 brouard 13074:
13075:
1.126 brouard 13076: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13077: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13078: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 13079: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13080: fclose(ficrespl);
13081:
13082: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13083: /*#include "hpijx.h"*/
13084: hPijx(p, bage, fage);
1.145 brouard 13085: fclose(ficrespij);
1.227 brouard 13086:
1.220 brouard 13087: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 13088: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 13089: k=1;
1.126 brouard 13090: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13091:
1.269 brouard 13092: /* Prevalence for each covariate combination in probs[age][status][cov] */
13093: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13094: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 13095: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 13096: for(k=1;k<=ncovcombmax;k++)
13097: probs[i][j][k]=0.;
1.269 brouard 13098: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
13099: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 13100: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 13101: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13102: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 13103: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 13104: for(k=1;k<=ncovcombmax;k++)
13105: mobaverages[i][j][k]=0.;
1.219 brouard 13106: mobaverage=mobaverages;
13107: if (mobilav!=0) {
1.235 brouard 13108: printf("Movingaveraging observed prevalence\n");
1.258 brouard 13109: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 13110: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
13111: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
13112: printf(" Error in movingaverage mobilav=%d\n",mobilav);
13113: }
1.269 brouard 13114: } else if (mobilavproj !=0) {
1.235 brouard 13115: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 13116: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 13117: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
13118: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
13119: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
13120: }
1.269 brouard 13121: }else{
13122: printf("Internal error moving average\n");
13123: fflush(stdout);
13124: exit(1);
1.219 brouard 13125: }
13126: }/* end if moving average */
1.227 brouard 13127:
1.126 brouard 13128: /*---------- Forecasting ------------------*/
1.296 brouard 13129: if(prevfcast==1){
13130: /* /\* if(stepm ==1){*\/ */
13131: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13132: /*This done previously after freqsummary.*/
13133: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
13134: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
13135:
13136: /* } else if (prvforecast==2){ */
13137: /* /\* if(stepm ==1){*\/ */
13138: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13139: /* } */
13140: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
13141: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 13142: }
1.269 brouard 13143:
1.296 brouard 13144: /* Prevbcasting */
13145: if(prevbcast==1){
1.219 brouard 13146: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13147: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13148: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13149:
13150: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
13151:
13152: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 13153:
1.219 brouard 13154: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
13155: fclose(ficresplb);
13156:
1.222 brouard 13157: hBijx(p, bage, fage, mobaverage);
13158: fclose(ficrespijb);
1.219 brouard 13159:
1.296 brouard 13160: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
13161: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
13162: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
13163: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
13164: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
13165: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
13166:
13167:
1.269 brouard 13168: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13169:
13170:
1.269 brouard 13171: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 13172: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13173: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13174: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 13175: } /* end Prevbcasting */
1.268 brouard 13176:
1.186 brouard 13177:
13178: /* ------ Other prevalence ratios------------ */
1.126 brouard 13179:
1.215 brouard 13180: free_ivector(wav,1,imx);
13181: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
13182: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
13183: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 13184:
13185:
1.127 brouard 13186: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 13187:
1.201 brouard 13188: strcpy(filerese,"E_");
13189: strcat(filerese,fileresu);
1.126 brouard 13190: if((ficreseij=fopen(filerese,"w"))==NULL) {
13191: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13192: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13193: }
1.208 brouard 13194: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
13195: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 13196:
13197: pstamp(ficreseij);
1.219 brouard 13198:
1.235 brouard 13199: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13200: if (cptcovn < 1){i1=1;}
13201:
13202: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13203: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13204: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13205: continue;
1.219 brouard 13206: fprintf(ficreseij,"\n#****** ");
1.235 brouard 13207: printf("\n#****** ");
1.225 brouard 13208: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 13209: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 13210: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13211: }
13212: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13213: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13214: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 13215: }
13216: fprintf(ficreseij,"******\n");
1.235 brouard 13217: printf("******\n");
1.219 brouard 13218:
13219: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13220: oldm=oldms;savm=savms;
1.235 brouard 13221: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 13222:
1.219 brouard 13223: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 13224: }
13225: fclose(ficreseij);
1.208 brouard 13226: printf("done evsij\n");fflush(stdout);
13227: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 13228:
1.218 brouard 13229:
1.227 brouard 13230: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 13231:
1.201 brouard 13232: strcpy(filerest,"T_");
13233: strcat(filerest,fileresu);
1.127 brouard 13234: if((ficrest=fopen(filerest,"w"))==NULL) {
13235: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
13236: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
13237: }
1.208 brouard 13238: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
13239: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 13240: strcpy(fileresstde,"STDE_");
13241: strcat(fileresstde,fileresu);
1.126 brouard 13242: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 13243: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
13244: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 13245: }
1.227 brouard 13246: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
13247: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 13248:
1.201 brouard 13249: strcpy(filerescve,"CVE_");
13250: strcat(filerescve,fileresu);
1.126 brouard 13251: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 13252: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
13253: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 13254: }
1.227 brouard 13255: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
13256: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 13257:
1.201 brouard 13258: strcpy(fileresv,"V_");
13259: strcat(fileresv,fileresu);
1.126 brouard 13260: if((ficresvij=fopen(fileresv,"w"))==NULL) {
13261: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
13262: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
13263: }
1.227 brouard 13264: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
13265: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 13266:
1.235 brouard 13267: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13268: if (cptcovn < 1){i1=1;}
13269:
13270: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13271: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13272: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13273: continue;
1.321 brouard 13274: printf("\n# model %s \n#****** Result for:", model);
13275: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
13276: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.227 brouard 13277: for(j=1;j<=cptcoveff;j++){
13278: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13279: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13280: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13281: }
1.235 brouard 13282: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13283: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13284: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13285: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13286: }
1.208 brouard 13287: fprintf(ficrest,"******\n");
1.227 brouard 13288: fprintf(ficlog,"******\n");
13289: printf("******\n");
1.208 brouard 13290:
13291: fprintf(ficresstdeij,"\n#****** ");
13292: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 13293: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 13294: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
13295: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 13296: }
1.235 brouard 13297: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13298: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13299: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13300: }
1.208 brouard 13301: fprintf(ficresstdeij,"******\n");
13302: fprintf(ficrescveij,"******\n");
13303:
13304: fprintf(ficresvij,"\n#****** ");
1.238 brouard 13305: /* pstamp(ficresvij); */
1.225 brouard 13306: for(j=1;j<=cptcoveff;j++)
1.227 brouard 13307: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 13308: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
13309: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
13310: }
1.208 brouard 13311: fprintf(ficresvij,"******\n");
13312:
13313: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13314: oldm=oldms;savm=savms;
1.235 brouard 13315: printf(" cvevsij ");
13316: fprintf(ficlog, " cvevsij ");
13317: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 13318: printf(" end cvevsij \n ");
13319: fprintf(ficlog, " end cvevsij \n ");
13320:
13321: /*
13322: */
13323: /* goto endfree; */
13324:
13325: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13326: pstamp(ficrest);
13327:
1.269 brouard 13328: epj=vector(1,nlstate+1);
1.208 brouard 13329: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 13330: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
13331: cptcod= 0; /* To be deleted */
13332: printf("varevsij vpopbased=%d \n",vpopbased);
13333: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 13334: 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 13335: 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 ");
13336: if(vpopbased==1)
13337: 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);
13338: else
1.288 brouard 13339: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13340: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
13341: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
13342: fprintf(ficrest,"\n");
13343: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 13344: printf("Computing age specific forward period (stable) prevalences in each health state \n");
13345: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13346: for(age=bage; age <=fage ;age++){
1.235 brouard 13347: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 13348: if (vpopbased==1) {
13349: if(mobilav ==0){
13350: for(i=1; i<=nlstate;i++)
13351: prlim[i][i]=probs[(int)age][i][k];
13352: }else{ /* mobilav */
13353: for(i=1; i<=nlstate;i++)
13354: prlim[i][i]=mobaverage[(int)age][i][k];
13355: }
13356: }
1.219 brouard 13357:
1.227 brouard 13358: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
13359: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
13360: /* printf(" age %4.0f ",age); */
13361: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
13362: for(i=1, epj[j]=0.;i <=nlstate;i++) {
13363: epj[j] += prlim[i][i]*eij[i][j][(int)age];
13364: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
13365: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
13366: }
13367: epj[nlstate+1] +=epj[j];
13368: }
13369: /* printf(" age %4.0f \n",age); */
1.219 brouard 13370:
1.227 brouard 13371: for(i=1, vepp=0.;i <=nlstate;i++)
13372: for(j=1;j <=nlstate;j++)
13373: vepp += vareij[i][j][(int)age];
13374: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
13375: for(j=1;j <=nlstate;j++){
13376: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
13377: }
13378: fprintf(ficrest,"\n");
13379: }
1.208 brouard 13380: } /* End vpopbased */
1.269 brouard 13381: free_vector(epj,1,nlstate+1);
1.208 brouard 13382: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
13383: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 13384: printf("done selection\n");fflush(stdout);
13385: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 13386:
1.235 brouard 13387: } /* End k selection */
1.227 brouard 13388:
13389: printf("done State-specific expectancies\n");fflush(stdout);
13390: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
13391:
1.288 brouard 13392: /* variance-covariance of forward period prevalence*/
1.269 brouard 13393: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13394:
1.227 brouard 13395:
1.290 brouard 13396: free_vector(weight,firstobs,lastobs);
1.227 brouard 13397: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 13398: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
13399: free_matrix(anint,1,maxwav,firstobs,lastobs);
13400: free_matrix(mint,1,maxwav,firstobs,lastobs);
13401: free_ivector(cod,firstobs,lastobs);
1.227 brouard 13402: free_ivector(tab,1,NCOVMAX);
13403: fclose(ficresstdeij);
13404: fclose(ficrescveij);
13405: fclose(ficresvij);
13406: fclose(ficrest);
13407: fclose(ficpar);
13408:
13409:
1.126 brouard 13410: /*---------- End : free ----------------*/
1.219 brouard 13411: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 13412: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
13413: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 13414: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
13415: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 13416: } /* mle==-3 arrives here for freeing */
1.227 brouard 13417: /* endfree:*/
13418: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
13419: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
13420: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 13421: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
13422: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
13423: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
13424: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 13425: free_matrix(matcov,1,npar,1,npar);
13426: free_matrix(hess,1,npar,1,npar);
13427: /*free_vector(delti,1,npar);*/
13428: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13429: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 13430: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 13431: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13432:
13433: free_ivector(ncodemax,1,NCOVMAX);
13434: free_ivector(ncodemaxwundef,1,NCOVMAX);
13435: free_ivector(Dummy,-1,NCOVMAX);
13436: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 13437: free_ivector(DummyV,1,NCOVMAX);
13438: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 13439: free_ivector(Typevar,-1,NCOVMAX);
13440: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 13441: free_ivector(TvarsQ,1,NCOVMAX);
13442: free_ivector(TvarsQind,1,NCOVMAX);
13443: free_ivector(TvarsD,1,NCOVMAX);
13444: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 13445: free_ivector(TvarFD,1,NCOVMAX);
13446: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 13447: free_ivector(TvarF,1,NCOVMAX);
13448: free_ivector(TvarFind,1,NCOVMAX);
13449: free_ivector(TvarV,1,NCOVMAX);
13450: free_ivector(TvarVind,1,NCOVMAX);
13451: free_ivector(TvarA,1,NCOVMAX);
13452: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 13453: free_ivector(TvarFQ,1,NCOVMAX);
13454: free_ivector(TvarFQind,1,NCOVMAX);
13455: free_ivector(TvarVD,1,NCOVMAX);
13456: free_ivector(TvarVDind,1,NCOVMAX);
13457: free_ivector(TvarVQ,1,NCOVMAX);
13458: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 13459: free_ivector(Tvarsel,1,NCOVMAX);
13460: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 13461: free_ivector(Tposprod,1,NCOVMAX);
13462: free_ivector(Tprod,1,NCOVMAX);
13463: free_ivector(Tvaraff,1,NCOVMAX);
13464: free_ivector(invalidvarcomb,1,ncovcombmax);
13465: free_ivector(Tage,1,NCOVMAX);
13466: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 13467: free_ivector(TmodelInvind,1,NCOVMAX);
13468: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 13469:
13470: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
13471: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 13472: fflush(fichtm);
13473: fflush(ficgp);
13474:
1.227 brouard 13475:
1.126 brouard 13476: if((nberr >0) || (nbwarn>0)){
1.216 brouard 13477: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
13478: 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 13479: }else{
13480: printf("End of Imach\n");
13481: fprintf(ficlog,"End of Imach\n");
13482: }
13483: printf("See log file on %s\n",filelog);
13484: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 13485: /*(void) gettimeofday(&end_time,&tzp);*/
13486: rend_time = time(NULL);
13487: end_time = *localtime(&rend_time);
13488: /* tml = *localtime(&end_time.tm_sec); */
13489: strcpy(strtend,asctime(&end_time));
1.126 brouard 13490: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
13491: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 13492: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13493:
1.157 brouard 13494: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13495: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13496: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13497: /* printf("Total time was %d uSec.\n", total_usecs);*/
13498: /* if(fileappend(fichtm,optionfilehtm)){ */
13499: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13500: fclose(fichtm);
13501: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13502: fclose(fichtmcov);
13503: fclose(ficgp);
13504: fclose(ficlog);
13505: /*------ End -----------*/
1.227 brouard 13506:
1.281 brouard 13507:
13508: /* Executes gnuplot */
1.227 brouard 13509:
13510: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13511: #ifdef WIN32
1.227 brouard 13512: if (_chdir(pathcd) != 0)
13513: printf("Can't move to directory %s!\n",path);
13514: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13515: #else
1.227 brouard 13516: if(chdir(pathcd) != 0)
13517: printf("Can't move to directory %s!\n", path);
13518: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13519: #endif
1.126 brouard 13520: printf("Current directory %s!\n",pathcd);
13521: /*strcat(plotcmd,CHARSEPARATOR);*/
13522: sprintf(plotcmd,"gnuplot");
1.157 brouard 13523: #ifdef _WIN32
1.126 brouard 13524: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13525: #endif
13526: if(!stat(plotcmd,&info)){
1.158 brouard 13527: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13528: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13529: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13530: }else
13531: strcpy(pplotcmd,plotcmd);
1.157 brouard 13532: #ifdef __unix
1.126 brouard 13533: strcpy(plotcmd,GNUPLOTPROGRAM);
13534: if(!stat(plotcmd,&info)){
1.158 brouard 13535: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13536: }else
13537: strcpy(pplotcmd,plotcmd);
13538: #endif
13539: }else
13540: strcpy(pplotcmd,plotcmd);
13541:
13542: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13543: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13544: strcpy(pplotcmd,plotcmd);
1.227 brouard 13545:
1.126 brouard 13546: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13547: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13548: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13549: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13550: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13551: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13552: strcpy(plotcmd,pplotcmd);
13553: }
1.126 brouard 13554: }
1.158 brouard 13555: printf(" Successful, please wait...");
1.126 brouard 13556: while (z[0] != 'q') {
13557: /* chdir(path); */
1.154 brouard 13558: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13559: scanf("%s",z);
13560: /* if (z[0] == 'c') system("./imach"); */
13561: if (z[0] == 'e') {
1.158 brouard 13562: #ifdef __APPLE__
1.152 brouard 13563: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13564: #elif __linux
13565: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13566: #else
1.152 brouard 13567: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13568: #endif
13569: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13570: system(pplotcmd);
1.126 brouard 13571: }
13572: else if (z[0] == 'g') system(plotcmd);
13573: else if (z[0] == 'q') exit(0);
13574: }
1.227 brouard 13575: end:
1.126 brouard 13576: while (z[0] != 'q') {
1.195 brouard 13577: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13578: scanf("%s",z);
13579: }
1.283 brouard 13580: printf("End\n");
1.282 brouard 13581: exit(0);
1.126 brouard 13582: }
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