Annotation of imach/src/imach.c, revision 1.333
1.333 ! brouard 1: /* $Id: imach.c,v 1.332 2022/08/21 09:06:25 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.333 ! brouard 4: Revision 1.332 2022/08/21 09:06:25 brouard
! 5: Summary: Version 0.99r33
! 6:
! 7: * src/imach.c (Module): Version 0.99r33 A lot of changes in
! 8: reassigning covariates: my first idea was that people will always
! 9: use the first covariate V1 into the model but in fact they are
! 10: producing data with many covariates and can use an equation model
! 11: with some of the covariate; it means that in a model V2+V3 instead
! 12: of codtabm(k,Tvaraff[j]) which calculates for combination k, for
! 13: three covariates (V1, V2, V3) the value of Tvaraff[j], but in fact
! 14: the equation model is restricted to two variables only (V2, V3)
! 15: and the combination for V2 should be codtabm(k,1) instead of
! 16: (codtabm(k,2), and the code should be
! 17: codtabm(k,TnsdVar[Tvaraff[j]]. Many many changes have been
! 18: made. All of these should be simplified once a day like we did in
! 19: hpxij() for example by using precov[nres] which is computed in
! 20: decoderesult for each nres of each resultline. Loop should be done
! 21: on the equation model globally by distinguishing only product with
! 22: age (which are changing with age) and no more on type of
! 23: covariates, single dummies, single covariates.
! 24:
1.332 brouard 25: Revision 1.331 2022/08/07 05:40:09 brouard
26: *** empty log message ***
27:
1.331 brouard 28: Revision 1.330 2022/08/06 07:18:25 brouard
29: Summary: last 0.99r31
30:
31: * imach.c (Module): Version of imach using partly decoderesult to rebuild xpxij function
32:
1.330 brouard 33: Revision 1.329 2022/08/03 17:29:54 brouard
34: * imach.c (Module): Many errors in graphs fixed with Vn*age covariates.
35:
1.329 brouard 36: Revision 1.328 2022/07/27 17:40:48 brouard
37: Summary: valgrind bug fixed by initializing to zero DummyV as well as Tage
38:
1.328 brouard 39: Revision 1.327 2022/07/27 14:47:35 brouard
40: Summary: Still a problem for one-step probabilities in case of quantitative variables
41:
1.327 brouard 42: Revision 1.326 2022/07/26 17:33:55 brouard
43: Summary: some test with nres=1
44:
1.326 brouard 45: Revision 1.325 2022/07/25 14:27:23 brouard
46: Summary: r30
47:
48: * imach.c (Module): Error cptcovn instead of nsd in bmij (was
49: coredumped, revealed by Feiuno, thank you.
50:
1.325 brouard 51: Revision 1.324 2022/07/23 17:44:26 brouard
52: *** empty log message ***
53:
1.324 brouard 54: Revision 1.323 2022/07/22 12:30:08 brouard
55: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
56:
1.323 brouard 57: Revision 1.322 2022/07/22 12:27:48 brouard
58: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
59:
1.322 brouard 60: Revision 1.321 2022/07/22 12:04:24 brouard
61: Summary: r28
62:
63: * imach.c (Module): Output of Wald test in the htm file and not only in the log.
64:
1.321 brouard 65: Revision 1.320 2022/06/02 05:10:11 brouard
66: *** empty log message ***
67:
1.320 brouard 68: Revision 1.319 2022/06/02 04:45:11 brouard
69: * imach.c (Module): Adding the Wald tests from the log to the main
70: htm for better display of the maximum likelihood estimators.
71:
1.319 brouard 72: Revision 1.318 2022/05/24 08:10:59 brouard
73: * imach.c (Module): Some attempts to find a bug of wrong estimates
74: of confidencce intervals with product in the equation modelC
75:
1.318 brouard 76: Revision 1.317 2022/05/15 15:06:23 brouard
77: * imach.c (Module): Some minor improvements
78:
1.317 brouard 79: Revision 1.316 2022/05/11 15:11:31 brouard
80: Summary: r27
81:
1.316 brouard 82: Revision 1.315 2022/05/11 15:06:32 brouard
83: *** empty log message ***
84:
1.315 brouard 85: Revision 1.314 2022/04/13 17:43:09 brouard
86: * imach.c (Module): Adding link to text data files
87:
1.314 brouard 88: Revision 1.313 2022/04/11 15:57:42 brouard
89: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
90:
1.313 brouard 91: Revision 1.312 2022/04/05 21:24:39 brouard
92: *** empty log message ***
93:
1.312 brouard 94: Revision 1.311 2022/04/05 21:03:51 brouard
95: Summary: Fixed quantitative covariates
96:
97: Fixed covariates (dummy or quantitative)
98: with missing values have never been allowed but are ERRORS and
99: program quits. Standard deviations of fixed covariates were
100: wrongly computed. Mean and standard deviations of time varying
101: covariates are still not computed.
102:
1.311 brouard 103: Revision 1.310 2022/03/17 08:45:53 brouard
104: Summary: 99r25
105:
106: Improving detection of errors: result lines should be compatible with
107: the model.
108:
1.310 brouard 109: Revision 1.309 2021/05/20 12:39:14 brouard
110: Summary: Version 0.99r24
111:
1.309 brouard 112: Revision 1.308 2021/03/31 13:11:57 brouard
113: Summary: Version 0.99r23
114:
115:
116: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
117:
1.308 brouard 118: Revision 1.307 2021/03/08 18:11:32 brouard
119: Summary: 0.99r22 fixed bug on result:
120:
1.307 brouard 121: Revision 1.306 2021/02/20 15:44:02 brouard
122: Summary: Version 0.99r21
123:
124: * imach.c (Module): Fix bug on quitting after result lines!
125: (Module): Version 0.99r21
126:
1.306 brouard 127: Revision 1.305 2021/02/20 15:28:30 brouard
128: * imach.c (Module): Fix bug on quitting after result lines!
129:
1.305 brouard 130: Revision 1.304 2021/02/12 11:34:20 brouard
131: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
132:
1.304 brouard 133: Revision 1.303 2021/02/11 19:50:15 brouard
134: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
135:
1.303 brouard 136: Revision 1.302 2020/02/22 21:00:05 brouard
137: * (Module): imach.c Update mle=-3 (for computing Life expectancy
138: and life table from the data without any state)
139:
1.302 brouard 140: Revision 1.301 2019/06/04 13:51:20 brouard
141: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
142:
1.301 brouard 143: Revision 1.300 2019/05/22 19:09:45 brouard
144: Summary: version 0.99r19 of May 2019
145:
1.300 brouard 146: Revision 1.299 2019/05/22 18:37:08 brouard
147: Summary: Cleaned 0.99r19
148:
1.299 brouard 149: Revision 1.298 2019/05/22 18:19:56 brouard
150: *** empty log message ***
151:
1.298 brouard 152: Revision 1.297 2019/05/22 17:56:10 brouard
153: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
154:
1.297 brouard 155: Revision 1.296 2019/05/20 13:03:18 brouard
156: Summary: Projection syntax simplified
157:
158:
159: We can now start projections, forward or backward, from the mean date
160: of inteviews up to or down to a number of years of projection:
161: prevforecast=1 yearsfproj=15.3 mobil_average=0
162: or
163: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
164: or
165: prevbackcast=1 yearsbproj=12.3 mobil_average=1
166: or
167: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
168:
1.296 brouard 169: Revision 1.295 2019/05/18 09:52:50 brouard
170: Summary: doxygen tex bug
171:
1.295 brouard 172: Revision 1.294 2019/05/16 14:54:33 brouard
173: Summary: There was some wrong lines added
174:
1.294 brouard 175: Revision 1.293 2019/05/09 15:17:34 brouard
176: *** empty log message ***
177:
1.293 brouard 178: Revision 1.292 2019/05/09 14:17:20 brouard
179: Summary: Some updates
180:
1.292 brouard 181: Revision 1.291 2019/05/09 13:44:18 brouard
182: Summary: Before ncovmax
183:
1.291 brouard 184: Revision 1.290 2019/05/09 13:39:37 brouard
185: Summary: 0.99r18 unlimited number of individuals
186:
187: 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.
188:
1.290 brouard 189: Revision 1.289 2018/12/13 09:16:26 brouard
190: Summary: Bug for young ages (<-30) will be in r17
191:
1.289 brouard 192: Revision 1.288 2018/05/02 20:58:27 brouard
193: Summary: Some bugs fixed
194:
1.288 brouard 195: Revision 1.287 2018/05/01 17:57:25 brouard
196: Summary: Bug fixed by providing frequencies only for non missing covariates
197:
1.287 brouard 198: Revision 1.286 2018/04/27 14:27:04 brouard
199: Summary: some minor bugs
200:
1.286 brouard 201: Revision 1.285 2018/04/21 21:02:16 brouard
202: Summary: Some bugs fixed, valgrind tested
203:
1.285 brouard 204: Revision 1.284 2018/04/20 05:22:13 brouard
205: Summary: Computing mean and stdeviation of fixed quantitative variables
206:
1.284 brouard 207: Revision 1.283 2018/04/19 14:49:16 brouard
208: Summary: Some minor bugs fixed
209:
1.283 brouard 210: Revision 1.282 2018/02/27 22:50:02 brouard
211: *** empty log message ***
212:
1.282 brouard 213: Revision 1.281 2018/02/27 19:25:23 brouard
214: Summary: Adding second argument for quitting
215:
1.281 brouard 216: Revision 1.280 2018/02/21 07:58:13 brouard
217: Summary: 0.99r15
218:
219: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
220:
1.280 brouard 221: Revision 1.279 2017/07/20 13:35:01 brouard
222: Summary: temporary working
223:
1.279 brouard 224: Revision 1.278 2017/07/19 14:09:02 brouard
225: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
226:
1.278 brouard 227: Revision 1.277 2017/07/17 08:53:49 brouard
228: Summary: BOM files can be read now
229:
1.277 brouard 230: Revision 1.276 2017/06/30 15:48:31 brouard
231: Summary: Graphs improvements
232:
1.276 brouard 233: Revision 1.275 2017/06/30 13:39:33 brouard
234: Summary: Saito's color
235:
1.275 brouard 236: Revision 1.274 2017/06/29 09:47:08 brouard
237: Summary: Version 0.99r14
238:
1.274 brouard 239: Revision 1.273 2017/06/27 11:06:02 brouard
240: Summary: More documentation on projections
241:
1.273 brouard 242: Revision 1.272 2017/06/27 10:22:40 brouard
243: Summary: Color of backprojection changed from 6 to 5(yellow)
244:
1.272 brouard 245: Revision 1.271 2017/06/27 10:17:50 brouard
246: Summary: Some bug with rint
247:
1.271 brouard 248: Revision 1.270 2017/05/24 05:45:29 brouard
249: *** empty log message ***
250:
1.270 brouard 251: Revision 1.269 2017/05/23 08:39:25 brouard
252: Summary: Code into subroutine, cleanings
253:
1.269 brouard 254: Revision 1.268 2017/05/18 20:09:32 brouard
255: Summary: backprojection and confidence intervals of backprevalence
256:
1.268 brouard 257: Revision 1.267 2017/05/13 10:25:05 brouard
258: Summary: temporary save for backprojection
259:
1.267 brouard 260: Revision 1.266 2017/05/13 07:26:12 brouard
261: Summary: Version 0.99r13 (improvements and bugs fixed)
262:
1.266 brouard 263: Revision 1.265 2017/04/26 16:22:11 brouard
264: Summary: imach 0.99r13 Some bugs fixed
265:
1.265 brouard 266: Revision 1.264 2017/04/26 06:01:29 brouard
267: Summary: Labels in graphs
268:
1.264 brouard 269: Revision 1.263 2017/04/24 15:23:15 brouard
270: Summary: to save
271:
1.263 brouard 272: Revision 1.262 2017/04/18 16:48:12 brouard
273: *** empty log message ***
274:
1.262 brouard 275: Revision 1.261 2017/04/05 10:14:09 brouard
276: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
277:
1.261 brouard 278: Revision 1.260 2017/04/04 17:46:59 brouard
279: Summary: Gnuplot indexations fixed (humm)
280:
1.260 brouard 281: Revision 1.259 2017/04/04 13:01:16 brouard
282: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
283:
1.259 brouard 284: Revision 1.258 2017/04/03 10:17:47 brouard
285: Summary: Version 0.99r12
286:
287: Some cleanings, conformed with updated documentation.
288:
1.258 brouard 289: Revision 1.257 2017/03/29 16:53:30 brouard
290: Summary: Temp
291:
1.257 brouard 292: Revision 1.256 2017/03/27 05:50:23 brouard
293: Summary: Temporary
294:
1.256 brouard 295: Revision 1.255 2017/03/08 16:02:28 brouard
296: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
297:
1.255 brouard 298: Revision 1.254 2017/03/08 07:13:00 brouard
299: Summary: Fixing data parameter line
300:
1.254 brouard 301: Revision 1.253 2016/12/15 11:59:41 brouard
302: Summary: 0.99 in progress
303:
1.253 brouard 304: Revision 1.252 2016/09/15 21:15:37 brouard
305: *** empty log message ***
306:
1.252 brouard 307: Revision 1.251 2016/09/15 15:01:13 brouard
308: Summary: not working
309:
1.251 brouard 310: Revision 1.250 2016/09/08 16:07:27 brouard
311: Summary: continue
312:
1.250 brouard 313: Revision 1.249 2016/09/07 17:14:18 brouard
314: Summary: Starting values from frequencies
315:
1.249 brouard 316: Revision 1.248 2016/09/07 14:10:18 brouard
317: *** empty log message ***
318:
1.248 brouard 319: Revision 1.247 2016/09/02 11:11:21 brouard
320: *** empty log message ***
321:
1.247 brouard 322: Revision 1.246 2016/09/02 08:49:22 brouard
323: *** empty log message ***
324:
1.246 brouard 325: Revision 1.245 2016/09/02 07:25:01 brouard
326: *** empty log message ***
327:
1.245 brouard 328: Revision 1.244 2016/09/02 07:17:34 brouard
329: *** empty log message ***
330:
1.244 brouard 331: Revision 1.243 2016/09/02 06:45:35 brouard
332: *** empty log message ***
333:
1.243 brouard 334: Revision 1.242 2016/08/30 15:01:20 brouard
335: Summary: Fixing a lots
336:
1.242 brouard 337: Revision 1.241 2016/08/29 17:17:25 brouard
338: Summary: gnuplot problem in Back projection to fix
339:
1.241 brouard 340: Revision 1.240 2016/08/29 07:53:18 brouard
341: Summary: Better
342:
1.240 brouard 343: Revision 1.239 2016/08/26 15:51:03 brouard
344: Summary: Improvement in Powell output in order to copy and paste
345:
346: Author:
347:
1.239 brouard 348: Revision 1.238 2016/08/26 14:23:35 brouard
349: Summary: Starting tests of 0.99
350:
1.238 brouard 351: Revision 1.237 2016/08/26 09:20:19 brouard
352: Summary: to valgrind
353:
1.237 brouard 354: Revision 1.236 2016/08/25 10:50:18 brouard
355: *** empty log message ***
356:
1.236 brouard 357: Revision 1.235 2016/08/25 06:59:23 brouard
358: *** empty log message ***
359:
1.235 brouard 360: Revision 1.234 2016/08/23 16:51:20 brouard
361: *** empty log message ***
362:
1.234 brouard 363: Revision 1.233 2016/08/23 07:40:50 brouard
364: Summary: not working
365:
1.233 brouard 366: Revision 1.232 2016/08/22 14:20:21 brouard
367: Summary: not working
368:
1.232 brouard 369: Revision 1.231 2016/08/22 07:17:15 brouard
370: Summary: not working
371:
1.231 brouard 372: Revision 1.230 2016/08/22 06:55:53 brouard
373: Summary: Not working
374:
1.230 brouard 375: Revision 1.229 2016/07/23 09:45:53 brouard
376: Summary: Completing for func too
377:
1.229 brouard 378: Revision 1.228 2016/07/22 17:45:30 brouard
379: Summary: Fixing some arrays, still debugging
380:
1.227 brouard 381: Revision 1.226 2016/07/12 18:42:34 brouard
382: Summary: temp
383:
1.226 brouard 384: Revision 1.225 2016/07/12 08:40:03 brouard
385: Summary: saving but not running
386:
1.225 brouard 387: Revision 1.224 2016/07/01 13:16:01 brouard
388: Summary: Fixes
389:
1.224 brouard 390: Revision 1.223 2016/02/19 09:23:35 brouard
391: Summary: temporary
392:
1.223 brouard 393: Revision 1.222 2016/02/17 08:14:50 brouard
394: Summary: Probably last 0.98 stable version 0.98r6
395:
1.222 brouard 396: Revision 1.221 2016/02/15 23:35:36 brouard
397: Summary: minor bug
398:
1.220 brouard 399: Revision 1.219 2016/02/15 00:48:12 brouard
400: *** empty log message ***
401:
1.219 brouard 402: Revision 1.218 2016/02/12 11:29:23 brouard
403: Summary: 0.99 Back projections
404:
1.218 brouard 405: Revision 1.217 2015/12/23 17:18:31 brouard
406: Summary: Experimental backcast
407:
1.217 brouard 408: Revision 1.216 2015/12/18 17:32:11 brouard
409: Summary: 0.98r4 Warning and status=-2
410:
411: Version 0.98r4 is now:
412: - displaying an error when status is -1, date of interview unknown and date of death known;
413: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
414: Older changes concerning s=-2, dating from 2005 have been supersed.
415:
1.216 brouard 416: Revision 1.215 2015/12/16 08:52:24 brouard
417: Summary: 0.98r4 working
418:
1.215 brouard 419: Revision 1.214 2015/12/16 06:57:54 brouard
420: Summary: temporary not working
421:
1.214 brouard 422: Revision 1.213 2015/12/11 18:22:17 brouard
423: Summary: 0.98r4
424:
1.213 brouard 425: Revision 1.212 2015/11/21 12:47:24 brouard
426: Summary: minor typo
427:
1.212 brouard 428: Revision 1.211 2015/11/21 12:41:11 brouard
429: Summary: 0.98r3 with some graph of projected cross-sectional
430:
431: Author: Nicolas Brouard
432:
1.211 brouard 433: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 434: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 435: Summary: Adding ftolpl parameter
436: Author: N Brouard
437:
438: We had difficulties to get smoothed confidence intervals. It was due
439: to the period prevalence which wasn't computed accurately. The inner
440: parameter ftolpl is now an outer parameter of the .imach parameter
441: file after estepm. If ftolpl is small 1.e-4 and estepm too,
442: computation are long.
443:
1.209 brouard 444: Revision 1.208 2015/11/17 14:31:57 brouard
445: Summary: temporary
446:
1.208 brouard 447: Revision 1.207 2015/10/27 17:36:57 brouard
448: *** empty log message ***
449:
1.207 brouard 450: Revision 1.206 2015/10/24 07:14:11 brouard
451: *** empty log message ***
452:
1.206 brouard 453: Revision 1.205 2015/10/23 15:50:53 brouard
454: Summary: 0.98r3 some clarification for graphs on likelihood contributions
455:
1.205 brouard 456: Revision 1.204 2015/10/01 16:20:26 brouard
457: Summary: Some new graphs of contribution to likelihood
458:
1.204 brouard 459: Revision 1.203 2015/09/30 17:45:14 brouard
460: Summary: looking at better estimation of the hessian
461:
462: Also a better criteria for convergence to the period prevalence And
463: therefore adding the number of years needed to converge. (The
464: prevalence in any alive state shold sum to one
465:
1.203 brouard 466: Revision 1.202 2015/09/22 19:45:16 brouard
467: Summary: Adding some overall graph on contribution to likelihood. Might change
468:
1.202 brouard 469: Revision 1.201 2015/09/15 17:34:58 brouard
470: Summary: 0.98r0
471:
472: - Some new graphs like suvival functions
473: - Some bugs fixed like model=1+age+V2.
474:
1.201 brouard 475: Revision 1.200 2015/09/09 16:53:55 brouard
476: Summary: Big bug thanks to Flavia
477:
478: Even model=1+age+V2. did not work anymore
479:
1.200 brouard 480: Revision 1.199 2015/09/07 14:09:23 brouard
481: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
482:
1.199 brouard 483: Revision 1.198 2015/09/03 07:14:39 brouard
484: Summary: 0.98q5 Flavia
485:
1.198 brouard 486: Revision 1.197 2015/09/01 18:24:39 brouard
487: *** empty log message ***
488:
1.197 brouard 489: Revision 1.196 2015/08/18 23:17:52 brouard
490: Summary: 0.98q5
491:
1.196 brouard 492: Revision 1.195 2015/08/18 16:28:39 brouard
493: Summary: Adding a hack for testing purpose
494:
495: After reading the title, ftol and model lines, if the comment line has
496: a q, starting with #q, the answer at the end of the run is quit. It
497: permits to run test files in batch with ctest. The former workaround was
498: $ echo q | imach foo.imach
499:
1.195 brouard 500: Revision 1.194 2015/08/18 13:32:00 brouard
501: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
502:
1.194 brouard 503: Revision 1.193 2015/08/04 07:17:42 brouard
504: Summary: 0.98q4
505:
1.193 brouard 506: Revision 1.192 2015/07/16 16:49:02 brouard
507: Summary: Fixing some outputs
508:
1.192 brouard 509: Revision 1.191 2015/07/14 10:00:33 brouard
510: Summary: Some fixes
511:
1.191 brouard 512: Revision 1.190 2015/05/05 08:51:13 brouard
513: Summary: Adding digits in output parameters (7 digits instead of 6)
514:
515: Fix 1+age+.
516:
1.190 brouard 517: Revision 1.189 2015/04/30 14:45:16 brouard
518: Summary: 0.98q2
519:
1.189 brouard 520: Revision 1.188 2015/04/30 08:27:53 brouard
521: *** empty log message ***
522:
1.188 brouard 523: Revision 1.187 2015/04/29 09:11:15 brouard
524: *** empty log message ***
525:
1.187 brouard 526: Revision 1.186 2015/04/23 12:01:52 brouard
527: Summary: V1*age is working now, version 0.98q1
528:
529: Some codes had been disabled in order to simplify and Vn*age was
530: working in the optimization phase, ie, giving correct MLE parameters,
531: but, as usual, outputs were not correct and program core dumped.
532:
1.186 brouard 533: Revision 1.185 2015/03/11 13:26:42 brouard
534: Summary: Inclusion of compile and links command line for Intel Compiler
535:
1.185 brouard 536: Revision 1.184 2015/03/11 11:52:39 brouard
537: Summary: Back from Windows 8. Intel Compiler
538:
1.184 brouard 539: Revision 1.183 2015/03/10 20:34:32 brouard
540: Summary: 0.98q0, trying with directest, mnbrak fixed
541:
542: We use directest instead of original Powell test; probably no
543: incidence on the results, but better justifications;
544: We fixed Numerical Recipes mnbrak routine which was wrong and gave
545: wrong results.
546:
1.183 brouard 547: Revision 1.182 2015/02/12 08:19:57 brouard
548: Summary: Trying to keep directest which seems simpler and more general
549: Author: Nicolas Brouard
550:
1.182 brouard 551: Revision 1.181 2015/02/11 23:22:24 brouard
552: Summary: Comments on Powell added
553:
554: Author:
555:
1.181 brouard 556: Revision 1.180 2015/02/11 17:33:45 brouard
557: Summary: Finishing move from main to function (hpijx and prevalence_limit)
558:
1.180 brouard 559: Revision 1.179 2015/01/04 09:57:06 brouard
560: Summary: back to OS/X
561:
1.179 brouard 562: Revision 1.178 2015/01/04 09:35:48 brouard
563: *** empty log message ***
564:
1.178 brouard 565: Revision 1.177 2015/01/03 18:40:56 brouard
566: Summary: Still testing ilc32 on OSX
567:
1.177 brouard 568: Revision 1.176 2015/01/03 16:45:04 brouard
569: *** empty log message ***
570:
1.176 brouard 571: Revision 1.175 2015/01/03 16:33:42 brouard
572: *** empty log message ***
573:
1.175 brouard 574: Revision 1.174 2015/01/03 16:15:49 brouard
575: Summary: Still in cross-compilation
576:
1.174 brouard 577: Revision 1.173 2015/01/03 12:06:26 brouard
578: Summary: trying to detect cross-compilation
579:
1.173 brouard 580: Revision 1.172 2014/12/27 12:07:47 brouard
581: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
582:
1.172 brouard 583: Revision 1.171 2014/12/23 13:26:59 brouard
584: Summary: Back from Visual C
585:
586: Still problem with utsname.h on Windows
587:
1.171 brouard 588: Revision 1.170 2014/12/23 11:17:12 brouard
589: Summary: Cleaning some \%% back to %%
590:
591: The escape was mandatory for a specific compiler (which one?), but too many warnings.
592:
1.170 brouard 593: Revision 1.169 2014/12/22 23:08:31 brouard
594: Summary: 0.98p
595:
596: Outputs some informations on compiler used, OS etc. Testing on different platforms.
597:
1.169 brouard 598: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 599: Summary: update
1.169 brouard 600:
1.168 brouard 601: Revision 1.167 2014/12/22 13:50:56 brouard
602: Summary: Testing uname and compiler version and if compiled 32 or 64
603:
604: Testing on Linux 64
605:
1.167 brouard 606: Revision 1.166 2014/12/22 11:40:47 brouard
607: *** empty log message ***
608:
1.166 brouard 609: Revision 1.165 2014/12/16 11:20:36 brouard
610: Summary: After compiling on Visual C
611:
612: * imach.c (Module): Merging 1.61 to 1.162
613:
1.165 brouard 614: Revision 1.164 2014/12/16 10:52:11 brouard
615: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
616:
617: * imach.c (Module): Merging 1.61 to 1.162
618:
1.164 brouard 619: Revision 1.163 2014/12/16 10:30:11 brouard
620: * imach.c (Module): Merging 1.61 to 1.162
621:
1.163 brouard 622: Revision 1.162 2014/09/25 11:43:39 brouard
623: Summary: temporary backup 0.99!
624:
1.162 brouard 625: Revision 1.1 2014/09/16 11:06:58 brouard
626: Summary: With some code (wrong) for nlopt
627:
628: Author:
629:
630: Revision 1.161 2014/09/15 20:41:41 brouard
631: Summary: Problem with macro SQR on Intel compiler
632:
1.161 brouard 633: Revision 1.160 2014/09/02 09:24:05 brouard
634: *** empty log message ***
635:
1.160 brouard 636: Revision 1.159 2014/09/01 10:34:10 brouard
637: Summary: WIN32
638: Author: Brouard
639:
1.159 brouard 640: Revision 1.158 2014/08/27 17:11:51 brouard
641: *** empty log message ***
642:
1.158 brouard 643: Revision 1.157 2014/08/27 16:26:55 brouard
644: Summary: Preparing windows Visual studio version
645: Author: Brouard
646:
647: In order to compile on Visual studio, time.h is now correct and time_t
648: and tm struct should be used. difftime should be used but sometimes I
649: just make the differences in raw time format (time(&now).
650: Trying to suppress #ifdef LINUX
651: Add xdg-open for __linux in order to open default browser.
652:
1.157 brouard 653: Revision 1.156 2014/08/25 20:10:10 brouard
654: *** empty log message ***
655:
1.156 brouard 656: Revision 1.155 2014/08/25 18:32:34 brouard
657: Summary: New compile, minor changes
658: Author: Brouard
659:
1.155 brouard 660: Revision 1.154 2014/06/20 17:32:08 brouard
661: Summary: Outputs now all graphs of convergence to period prevalence
662:
1.154 brouard 663: Revision 1.153 2014/06/20 16:45:46 brouard
664: Summary: If 3 live state, convergence to period prevalence on same graph
665: Author: Brouard
666:
1.153 brouard 667: Revision 1.152 2014/06/18 17:54:09 brouard
668: Summary: open browser, use gnuplot on same dir than imach if not found in the path
669:
1.152 brouard 670: Revision 1.151 2014/06/18 16:43:30 brouard
671: *** empty log message ***
672:
1.151 brouard 673: Revision 1.150 2014/06/18 16:42:35 brouard
674: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
675: Author: brouard
676:
1.150 brouard 677: Revision 1.149 2014/06/18 15:51:14 brouard
678: Summary: Some fixes in parameter files errors
679: Author: Nicolas Brouard
680:
1.149 brouard 681: Revision 1.148 2014/06/17 17:38:48 brouard
682: Summary: Nothing new
683: Author: Brouard
684:
685: Just a new packaging for OS/X version 0.98nS
686:
1.148 brouard 687: Revision 1.147 2014/06/16 10:33:11 brouard
688: *** empty log message ***
689:
1.147 brouard 690: Revision 1.146 2014/06/16 10:20:28 brouard
691: Summary: Merge
692: Author: Brouard
693:
694: Merge, before building revised version.
695:
1.146 brouard 696: Revision 1.145 2014/06/10 21:23:15 brouard
697: Summary: Debugging with valgrind
698: Author: Nicolas Brouard
699:
700: Lot of changes in order to output the results with some covariates
701: After the Edimburgh REVES conference 2014, it seems mandatory to
702: improve the code.
703: No more memory valgrind error but a lot has to be done in order to
704: continue the work of splitting the code into subroutines.
705: Also, decodemodel has been improved. Tricode is still not
706: optimal. nbcode should be improved. Documentation has been added in
707: the source code.
708:
1.144 brouard 709: Revision 1.143 2014/01/26 09:45:38 brouard
710: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
711:
712: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
713: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
714:
1.143 brouard 715: Revision 1.142 2014/01/26 03:57:36 brouard
716: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
717:
718: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
719:
1.142 brouard 720: Revision 1.141 2014/01/26 02:42:01 brouard
721: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
722:
1.141 brouard 723: Revision 1.140 2011/09/02 10:37:54 brouard
724: Summary: times.h is ok with mingw32 now.
725:
1.140 brouard 726: Revision 1.139 2010/06/14 07:50:17 brouard
727: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
728: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
729:
1.139 brouard 730: Revision 1.138 2010/04/30 18:19:40 brouard
731: *** empty log message ***
732:
1.138 brouard 733: Revision 1.137 2010/04/29 18:11:38 brouard
734: (Module): Checking covariates for more complex models
735: than V1+V2. A lot of change to be done. Unstable.
736:
1.137 brouard 737: Revision 1.136 2010/04/26 20:30:53 brouard
738: (Module): merging some libgsl code. Fixing computation
739: of likelione (using inter/intrapolation if mle = 0) in order to
740: get same likelihood as if mle=1.
741: Some cleaning of code and comments added.
742:
1.136 brouard 743: Revision 1.135 2009/10/29 15:33:14 brouard
744: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
745:
1.135 brouard 746: Revision 1.134 2009/10/29 13:18:53 brouard
747: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
748:
1.134 brouard 749: Revision 1.133 2009/07/06 10:21:25 brouard
750: just nforces
751:
1.133 brouard 752: Revision 1.132 2009/07/06 08:22:05 brouard
753: Many tings
754:
1.132 brouard 755: Revision 1.131 2009/06/20 16:22:47 brouard
756: Some dimensions resccaled
757:
1.131 brouard 758: Revision 1.130 2009/05/26 06:44:34 brouard
759: (Module): Max Covariate is now set to 20 instead of 8. A
760: lot of cleaning with variables initialized to 0. Trying to make
761: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
762:
1.130 brouard 763: Revision 1.129 2007/08/31 13:49:27 lievre
764: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
765:
1.129 lievre 766: Revision 1.128 2006/06/30 13:02:05 brouard
767: (Module): Clarifications on computing e.j
768:
1.128 brouard 769: Revision 1.127 2006/04/28 18:11:50 brouard
770: (Module): Yes the sum of survivors was wrong since
771: imach-114 because nhstepm was no more computed in the age
772: loop. Now we define nhstepma in the age loop.
773: (Module): In order to speed up (in case of numerous covariates) we
774: compute health expectancies (without variances) in a first step
775: and then all the health expectancies with variances or standard
776: deviation (needs data from the Hessian matrices) which slows the
777: computation.
778: In the future we should be able to stop the program is only health
779: expectancies and graph are needed without standard deviations.
780:
1.127 brouard 781: Revision 1.126 2006/04/28 17:23:28 brouard
782: (Module): Yes the sum of survivors was wrong since
783: imach-114 because nhstepm was no more computed in the age
784: loop. Now we define nhstepma in the age loop.
785: Version 0.98h
786:
1.126 brouard 787: Revision 1.125 2006/04/04 15:20:31 lievre
788: Errors in calculation of health expectancies. Age was not initialized.
789: Forecasting file added.
790:
791: Revision 1.124 2006/03/22 17:13:53 lievre
792: Parameters are printed with %lf instead of %f (more numbers after the comma).
793: The log-likelihood is printed in the log file
794:
795: Revision 1.123 2006/03/20 10:52:43 brouard
796: * imach.c (Module): <title> changed, corresponds to .htm file
797: name. <head> headers where missing.
798:
799: * imach.c (Module): Weights can have a decimal point as for
800: English (a comma might work with a correct LC_NUMERIC environment,
801: otherwise the weight is truncated).
802: Modification of warning when the covariates values are not 0 or
803: 1.
804: Version 0.98g
805:
806: Revision 1.122 2006/03/20 09:45:41 brouard
807: (Module): Weights can have a decimal point as for
808: English (a comma might work with a correct LC_NUMERIC environment,
809: otherwise the weight is truncated).
810: Modification of warning when the covariates values are not 0 or
811: 1.
812: Version 0.98g
813:
814: Revision 1.121 2006/03/16 17:45:01 lievre
815: * imach.c (Module): Comments concerning covariates added
816:
817: * imach.c (Module): refinements in the computation of lli if
818: status=-2 in order to have more reliable computation if stepm is
819: not 1 month. Version 0.98f
820:
821: Revision 1.120 2006/03/16 15:10:38 lievre
822: (Module): refinements in the computation of lli if
823: status=-2 in order to have more reliable computation if stepm is
824: not 1 month. Version 0.98f
825:
826: Revision 1.119 2006/03/15 17:42:26 brouard
827: (Module): Bug if status = -2, the loglikelihood was
828: computed as likelihood omitting the logarithm. Version O.98e
829:
830: Revision 1.118 2006/03/14 18:20:07 brouard
831: (Module): varevsij Comments added explaining the second
832: table of variances if popbased=1 .
833: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
834: (Module): Function pstamp added
835: (Module): Version 0.98d
836:
837: Revision 1.117 2006/03/14 17:16:22 brouard
838: (Module): varevsij Comments added explaining the second
839: table of variances if popbased=1 .
840: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
841: (Module): Function pstamp added
842: (Module): Version 0.98d
843:
844: Revision 1.116 2006/03/06 10:29:27 brouard
845: (Module): Variance-covariance wrong links and
846: varian-covariance of ej. is needed (Saito).
847:
848: Revision 1.115 2006/02/27 12:17:45 brouard
849: (Module): One freematrix added in mlikeli! 0.98c
850:
851: Revision 1.114 2006/02/26 12:57:58 brouard
852: (Module): Some improvements in processing parameter
853: filename with strsep.
854:
855: Revision 1.113 2006/02/24 14:20:24 brouard
856: (Module): Memory leaks checks with valgrind and:
857: datafile was not closed, some imatrix were not freed and on matrix
858: allocation too.
859:
860: Revision 1.112 2006/01/30 09:55:26 brouard
861: (Module): Back to gnuplot.exe instead of wgnuplot.exe
862:
863: Revision 1.111 2006/01/25 20:38:18 brouard
864: (Module): Lots of cleaning and bugs added (Gompertz)
865: (Module): Comments can be added in data file. Missing date values
866: can be a simple dot '.'.
867:
868: Revision 1.110 2006/01/25 00:51:50 brouard
869: (Module): Lots of cleaning and bugs added (Gompertz)
870:
871: Revision 1.109 2006/01/24 19:37:15 brouard
872: (Module): Comments (lines starting with a #) are allowed in data.
873:
874: Revision 1.108 2006/01/19 18:05:42 lievre
875: Gnuplot problem appeared...
876: To be fixed
877:
878: Revision 1.107 2006/01/19 16:20:37 brouard
879: Test existence of gnuplot in imach path
880:
881: Revision 1.106 2006/01/19 13:24:36 brouard
882: Some cleaning and links added in html output
883:
884: Revision 1.105 2006/01/05 20:23:19 lievre
885: *** empty log message ***
886:
887: Revision 1.104 2005/09/30 16:11:43 lievre
888: (Module): sump fixed, loop imx fixed, and simplifications.
889: (Module): If the status is missing at the last wave but we know
890: that the person is alive, then we can code his/her status as -2
891: (instead of missing=-1 in earlier versions) and his/her
892: contributions to the likelihood is 1 - Prob of dying from last
893: health status (= 1-p13= p11+p12 in the easiest case of somebody in
894: the healthy state at last known wave). Version is 0.98
895:
896: Revision 1.103 2005/09/30 15:54:49 lievre
897: (Module): sump fixed, loop imx fixed, and simplifications.
898:
899: Revision 1.102 2004/09/15 17:31:30 brouard
900: Add the possibility to read data file including tab characters.
901:
902: Revision 1.101 2004/09/15 10:38:38 brouard
903: Fix on curr_time
904:
905: Revision 1.100 2004/07/12 18:29:06 brouard
906: Add version for Mac OS X. Just define UNIX in Makefile
907:
908: Revision 1.99 2004/06/05 08:57:40 brouard
909: *** empty log message ***
910:
911: Revision 1.98 2004/05/16 15:05:56 brouard
912: New version 0.97 . First attempt to estimate force of mortality
913: directly from the data i.e. without the need of knowing the health
914: state at each age, but using a Gompertz model: log u =a + b*age .
915: This is the basic analysis of mortality and should be done before any
916: other analysis, in order to test if the mortality estimated from the
917: cross-longitudinal survey is different from the mortality estimated
918: from other sources like vital statistic data.
919:
920: The same imach parameter file can be used but the option for mle should be -3.
921:
1.324 brouard 922: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 923: former routines in order to include the new code within the former code.
924:
925: The output is very simple: only an estimate of the intercept and of
926: the slope with 95% confident intervals.
927:
928: Current limitations:
929: A) Even if you enter covariates, i.e. with the
930: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
931: B) There is no computation of Life Expectancy nor Life Table.
932:
933: Revision 1.97 2004/02/20 13:25:42 lievre
934: Version 0.96d. Population forecasting command line is (temporarily)
935: suppressed.
936:
937: Revision 1.96 2003/07/15 15:38:55 brouard
938: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
939: rewritten within the same printf. Workaround: many printfs.
940:
941: Revision 1.95 2003/07/08 07:54:34 brouard
942: * imach.c (Repository):
943: (Repository): Using imachwizard code to output a more meaningful covariance
944: matrix (cov(a12,c31) instead of numbers.
945:
946: Revision 1.94 2003/06/27 13:00:02 brouard
947: Just cleaning
948:
949: Revision 1.93 2003/06/25 16:33:55 brouard
950: (Module): On windows (cygwin) function asctime_r doesn't
951: exist so I changed back to asctime which exists.
952: (Module): Version 0.96b
953:
954: Revision 1.92 2003/06/25 16:30:45 brouard
955: (Module): On windows (cygwin) function asctime_r doesn't
956: exist so I changed back to asctime which exists.
957:
958: Revision 1.91 2003/06/25 15:30:29 brouard
959: * imach.c (Repository): Duplicated warning errors corrected.
960: (Repository): Elapsed time after each iteration is now output. It
961: helps to forecast when convergence will be reached. Elapsed time
962: is stamped in powell. We created a new html file for the graphs
963: concerning matrix of covariance. It has extension -cov.htm.
964:
965: Revision 1.90 2003/06/24 12:34:15 brouard
966: (Module): Some bugs corrected for windows. Also, when
967: mle=-1 a template is output in file "or"mypar.txt with the design
968: of the covariance matrix to be input.
969:
970: Revision 1.89 2003/06/24 12:30:52 brouard
971: (Module): Some bugs corrected for windows. Also, when
972: mle=-1 a template is output in file "or"mypar.txt with the design
973: of the covariance matrix to be input.
974:
975: Revision 1.88 2003/06/23 17:54:56 brouard
976: * 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.
977:
978: Revision 1.87 2003/06/18 12:26:01 brouard
979: Version 0.96
980:
981: Revision 1.86 2003/06/17 20:04:08 brouard
982: (Module): Change position of html and gnuplot routines and added
983: routine fileappend.
984:
985: Revision 1.85 2003/06/17 13:12:43 brouard
986: * imach.c (Repository): Check when date of death was earlier that
987: current date of interview. It may happen when the death was just
988: prior to the death. In this case, dh was negative and likelihood
989: was wrong (infinity). We still send an "Error" but patch by
990: assuming that the date of death was just one stepm after the
991: interview.
992: (Repository): Because some people have very long ID (first column)
993: we changed int to long in num[] and we added a new lvector for
994: memory allocation. But we also truncated to 8 characters (left
995: truncation)
996: (Repository): No more line truncation errors.
997:
998: Revision 1.84 2003/06/13 21:44:43 brouard
999: * imach.c (Repository): Replace "freqsummary" at a correct
1000: place. It differs from routine "prevalence" which may be called
1001: many times. Probs is memory consuming and must be used with
1002: parcimony.
1003: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
1004:
1005: Revision 1.83 2003/06/10 13:39:11 lievre
1006: *** empty log message ***
1007:
1008: Revision 1.82 2003/06/05 15:57:20 brouard
1009: Add log in imach.c and fullversion number is now printed.
1010:
1011: */
1012: /*
1013: Interpolated Markov Chain
1014:
1015: Short summary of the programme:
1016:
1.227 brouard 1017: This program computes Healthy Life Expectancies or State-specific
1018: (if states aren't health statuses) Expectancies from
1019: cross-longitudinal data. Cross-longitudinal data consist in:
1020:
1021: -1- a first survey ("cross") where individuals from different ages
1022: are interviewed on their health status or degree of disability (in
1023: the case of a health survey which is our main interest)
1024:
1025: -2- at least a second wave of interviews ("longitudinal") which
1026: measure each change (if any) in individual health status. Health
1027: expectancies are computed from the time spent in each health state
1028: according to a model. More health states you consider, more time is
1029: necessary to reach the Maximum Likelihood of the parameters involved
1030: in the model. The simplest model is the multinomial logistic model
1031: where pij is the probability to be observed in state j at the second
1032: wave conditional to be observed in state i at the first
1033: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
1034: etc , where 'age' is age and 'sex' is a covariate. If you want to
1035: have a more complex model than "constant and age", you should modify
1036: the program where the markup *Covariates have to be included here
1037: again* invites you to do it. More covariates you add, slower the
1.126 brouard 1038: convergence.
1039:
1040: The advantage of this computer programme, compared to a simple
1041: multinomial logistic model, is clear when the delay between waves is not
1042: identical for each individual. Also, if a individual missed an
1043: intermediate interview, the information is lost, but taken into
1044: account using an interpolation or extrapolation.
1045:
1046: hPijx is the probability to be observed in state i at age x+h
1047: conditional to the observed state i at age x. The delay 'h' can be
1048: split into an exact number (nh*stepm) of unobserved intermediate
1049: states. This elementary transition (by month, quarter,
1050: semester or year) is modelled as a multinomial logistic. The hPx
1051: matrix is simply the matrix product of nh*stepm elementary matrices
1052: and the contribution of each individual to the likelihood is simply
1053: hPijx.
1054:
1055: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 1056: of the life expectancies. It also computes the period (stable) prevalence.
1057:
1058: Back prevalence and projections:
1.227 brouard 1059:
1060: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
1061: double agemaxpar, double ftolpl, int *ncvyearp, double
1062: dateprev1,double dateprev2, int firstpass, int lastpass, int
1063: mobilavproj)
1064:
1065: Computes the back prevalence limit for any combination of
1066: covariate values k at any age between ageminpar and agemaxpar and
1067: returns it in **bprlim. In the loops,
1068:
1069: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
1070: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
1071:
1072: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 1073: Computes for any combination of covariates k and any age between bage and fage
1074: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
1075: oldm=oldms;savm=savms;
1.227 brouard 1076:
1.267 brouard 1077: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 1078: Computes the transition matrix starting at age 'age' over
1079: 'nhstepm*hstepm*stepm' months (i.e. until
1080: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 1081: nhstepm*hstepm matrices.
1082:
1083: Returns p3mat[i][j][h] after calling
1084: p3mat[i][j][h]=matprod2(newm,
1085: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1086: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1087: oldm);
1.226 brouard 1088:
1089: Important routines
1090:
1091: - func (or funcone), computes logit (pij) distinguishing
1092: o fixed variables (single or product dummies or quantitative);
1093: o varying variables by:
1094: (1) wave (single, product dummies, quantitative),
1095: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1096: % fixed dummy (treated) or quantitative (not done because time-consuming);
1097: % varying dummy (not done) or quantitative (not done);
1098: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1099: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1100: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1.325 brouard 1101: o There are 2**cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1.226 brouard 1102: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1103:
1.226 brouard 1104:
1105:
1.324 brouard 1106: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1107: Institut national d'études démographiques, Paris.
1.126 brouard 1108: This software have been partly granted by Euro-REVES, a concerted action
1109: from the European Union.
1110: It is copyrighted identically to a GNU software product, ie programme and
1111: software can be distributed freely for non commercial use. Latest version
1112: can be accessed at http://euroreves.ined.fr/imach .
1113:
1114: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1115: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1116:
1117: **********************************************************************/
1118: /*
1119: main
1120: read parameterfile
1121: read datafile
1122: concatwav
1123: freqsummary
1124: if (mle >= 1)
1125: mlikeli
1126: print results files
1127: if mle==1
1128: computes hessian
1129: read end of parameter file: agemin, agemax, bage, fage, estepm
1130: begin-prev-date,...
1131: open gnuplot file
1132: open html file
1.145 brouard 1133: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1134: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1135: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1136: freexexit2 possible for memory heap.
1137:
1138: h Pij x | pij_nom ficrestpij
1139: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1140: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1141: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1142:
1143: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1144: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1145: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1146: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1147: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1148:
1.126 brouard 1149: forecasting if prevfcast==1 prevforecast call prevalence()
1150: health expectancies
1151: Variance-covariance of DFLE
1152: prevalence()
1153: movingaverage()
1154: varevsij()
1155: if popbased==1 varevsij(,popbased)
1156: total life expectancies
1157: Variance of period (stable) prevalence
1158: end
1159: */
1160:
1.187 brouard 1161: /* #define DEBUG */
1162: /* #define DEBUGBRENT */
1.203 brouard 1163: /* #define DEBUGLINMIN */
1164: /* #define DEBUGHESS */
1165: #define DEBUGHESSIJ
1.224 brouard 1166: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1167: #define POWELL /* Instead of NLOPT */
1.224 brouard 1168: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1169: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1170: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.319 brouard 1171: /* #define FLATSUP *//* Suppresses directions where likelihood is flat */
1.126 brouard 1172:
1173: #include <math.h>
1174: #include <stdio.h>
1175: #include <stdlib.h>
1176: #include <string.h>
1.226 brouard 1177: #include <ctype.h>
1.159 brouard 1178:
1179: #ifdef _WIN32
1180: #include <io.h>
1.172 brouard 1181: #include <windows.h>
1182: #include <tchar.h>
1.159 brouard 1183: #else
1.126 brouard 1184: #include <unistd.h>
1.159 brouard 1185: #endif
1.126 brouard 1186:
1187: #include <limits.h>
1188: #include <sys/types.h>
1.171 brouard 1189:
1190: #if defined(__GNUC__)
1191: #include <sys/utsname.h> /* Doesn't work on Windows */
1192: #endif
1193:
1.126 brouard 1194: #include <sys/stat.h>
1195: #include <errno.h>
1.159 brouard 1196: /* extern int errno; */
1.126 brouard 1197:
1.157 brouard 1198: /* #ifdef LINUX */
1199: /* #include <time.h> */
1200: /* #include "timeval.h" */
1201: /* #else */
1202: /* #include <sys/time.h> */
1203: /* #endif */
1204:
1.126 brouard 1205: #include <time.h>
1206:
1.136 brouard 1207: #ifdef GSL
1208: #include <gsl/gsl_errno.h>
1209: #include <gsl/gsl_multimin.h>
1210: #endif
1211:
1.167 brouard 1212:
1.162 brouard 1213: #ifdef NLOPT
1214: #include <nlopt.h>
1215: typedef struct {
1216: double (* function)(double [] );
1217: } myfunc_data ;
1218: #endif
1219:
1.126 brouard 1220: /* #include <libintl.h> */
1221: /* #define _(String) gettext (String) */
1222:
1.251 brouard 1223: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1224:
1225: #define GNUPLOTPROGRAM "gnuplot"
1226: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1.329 brouard 1227: #define FILENAMELENGTH 256
1.126 brouard 1228:
1229: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1230: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1231:
1.144 brouard 1232: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1233: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1234:
1235: #define NINTERVMAX 8
1.144 brouard 1236: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1237: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.325 brouard 1238: #define NCOVMAX 30 /**< Maximum number of covariates used in the model, including generated covariates V1*V2 or V1*age */
1.197 brouard 1239: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1240: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1241: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1242: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1243: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1244: /* #define AGESUP 130 */
1.288 brouard 1245: /* #define AGESUP 150 */
1246: #define AGESUP 200
1.268 brouard 1247: #define AGEINF 0
1.218 brouard 1248: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1249: #define AGEBASE 40
1.194 brouard 1250: #define AGEOVERFLOW 1.e20
1.164 brouard 1251: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1252: #ifdef _WIN32
1253: #define DIRSEPARATOR '\\'
1254: #define CHARSEPARATOR "\\"
1255: #define ODIRSEPARATOR '/'
1256: #else
1.126 brouard 1257: #define DIRSEPARATOR '/'
1258: #define CHARSEPARATOR "/"
1259: #define ODIRSEPARATOR '\\'
1260: #endif
1261:
1.333 ! brouard 1262: /* $Id: imach.c,v 1.332 2022/08/21 09:06:25 brouard Exp $ */
1.126 brouard 1263: /* $State: Exp $ */
1.196 brouard 1264: #include "version.h"
1265: char version[]=__IMACH_VERSION__;
1.332 brouard 1266: char copyright[]="August 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";
1.333 ! brouard 1267: char fullversion[]="$Revision: 1.332 $ $Date: 2022/08/21 09:06:25 $";
1.126 brouard 1268: char strstart[80];
1269: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1270: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1271: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.330 brouard 1272: /* Number of covariates model (1)=V2+V1+ V3*age+V2*V4 */
1273: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
1274: int cptcovn=0; /**< cptcovn decodemodel: number of covariates k of the models excluding age*products =6 and age*age */
1275: int cptcovt=0; /**< cptcovt: total number of covariates of the model (2) nbocc(+)+1 = 8 excepting constant and age and age*age */
1276: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 (dummy or quantit or time varying) */
1.225 brouard 1277: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1278: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1279: int cptcovprodnoage=0; /**< Number of covariate products without age */
1.330 brouard 1280: int cptcoveff=0; /* Total number of covariates to vary for printing results (2**cptcoveff combinations of dummies)(computed in tricode as cptcov) */
1.233 brouard 1281: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1282: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1283: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1284: int nsd=0; /**< Total number of single dummy variables (output) */
1285: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1286: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1287: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1288: int ntveff=0; /**< ntveff number of effective time varying variables */
1289: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1290: int cptcov=0; /* Working variable */
1.290 brouard 1291: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1292: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1293: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1294: int nlstate=2; /* Number of live states */
1295: int ndeath=1; /* Number of dead states */
1.130 brouard 1296: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1297: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1298: int popbased=0;
1299:
1300: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1301: int maxwav=0; /* Maxim number of waves */
1302: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1303: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1304: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1305: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1306: int mle=1, weightopt=0;
1.126 brouard 1307: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1308: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1309: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1310: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1311: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1312: int selected(int kvar); /* Is covariate kvar selected for printing results */
1313:
1.130 brouard 1314: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1315: double **matprod2(); /* test */
1.126 brouard 1316: double **oldm, **newm, **savm; /* Working pointers to matrices */
1317: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1318: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1319:
1.136 brouard 1320: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1321: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1322: FILE *ficlog, *ficrespow;
1.130 brouard 1323: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1324: double fretone; /* Only one call to likelihood */
1.130 brouard 1325: long ipmx=0; /* Number of contributions */
1.126 brouard 1326: double sw; /* Sum of weights */
1327: char filerespow[FILENAMELENGTH];
1328: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1329: FILE *ficresilk;
1330: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1331: FILE *ficresprobmorprev;
1332: FILE *fichtm, *fichtmcov; /* Html File */
1333: FILE *ficreseij;
1334: char filerese[FILENAMELENGTH];
1335: FILE *ficresstdeij;
1336: char fileresstde[FILENAMELENGTH];
1337: FILE *ficrescveij;
1338: char filerescve[FILENAMELENGTH];
1339: FILE *ficresvij;
1340: char fileresv[FILENAMELENGTH];
1.269 brouard 1341:
1.126 brouard 1342: char title[MAXLINE];
1.234 brouard 1343: char model[MAXLINE]; /**< The model line */
1.217 brouard 1344: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1345: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1346: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1347: char command[FILENAMELENGTH];
1348: int outcmd=0;
1349:
1.217 brouard 1350: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1351: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1352: char filelog[FILENAMELENGTH]; /* Log file */
1353: char filerest[FILENAMELENGTH];
1354: char fileregp[FILENAMELENGTH];
1355: char popfile[FILENAMELENGTH];
1356:
1357: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1358:
1.157 brouard 1359: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1360: /* struct timezone tzp; */
1361: /* extern int gettimeofday(); */
1362: struct tm tml, *gmtime(), *localtime();
1363:
1364: extern time_t time();
1365:
1366: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1367: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1368: struct tm tm;
1369:
1.126 brouard 1370: char strcurr[80], strfor[80];
1371:
1372: char *endptr;
1373: long lval;
1374: double dval;
1375:
1376: #define NR_END 1
1377: #define FREE_ARG char*
1378: #define FTOL 1.0e-10
1379:
1380: #define NRANSI
1.240 brouard 1381: #define ITMAX 200
1382: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1383:
1384: #define TOL 2.0e-4
1385:
1386: #define CGOLD 0.3819660
1387: #define ZEPS 1.0e-10
1388: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1389:
1390: #define GOLD 1.618034
1391: #define GLIMIT 100.0
1392: #define TINY 1.0e-20
1393:
1394: static double maxarg1,maxarg2;
1395: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1396: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1397:
1398: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1399: #define rint(a) floor(a+0.5)
1.166 brouard 1400: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1401: #define mytinydouble 1.0e-16
1.166 brouard 1402: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1403: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1404: /* static double dsqrarg; */
1405: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1406: static double sqrarg;
1407: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1408: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1409: int agegomp= AGEGOMP;
1410:
1411: int imx;
1412: int stepm=1;
1413: /* Stepm, step in month: minimum step interpolation*/
1414:
1415: int estepm;
1416: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1417:
1418: int m,nb;
1419: long *num;
1.197 brouard 1420: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1421: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1422: covariate for which somebody answered excluding
1423: undefined. Usually 2: 0 and 1. */
1424: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1425: covariate for which somebody answered including
1426: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1427: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1428: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1429: double ***mobaverage, ***mobaverages; /* New global variable */
1.332 brouard 1430: double **precov; /* New global variable to store for each resultline, values of model covariates given by the resultlines (in order to speed up) */
1.126 brouard 1431: double *ageexmed,*agecens;
1432: double dateintmean=0;
1.296 brouard 1433: double anprojd, mprojd, jprojd; /* For eventual projections */
1434: double anprojf, mprojf, jprojf;
1.126 brouard 1435:
1.296 brouard 1436: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1437: double anbackf, mbackf, jbackf;
1438: double jintmean,mintmean,aintmean;
1.126 brouard 1439: double *weight;
1440: int **s; /* Status */
1.141 brouard 1441: double *agedc;
1.145 brouard 1442: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1443: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1444: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1445: double **coqvar; /* Fixed quantitative covariate nqv */
1446: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1447: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1448: double idx;
1449: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.319 brouard 1450: /* Some documentation */
1451: /* Design original data
1452: * V1 V2 V3 V4 V5 V6 V7 V8 Weight ddb ddth d1st s1 V9 V10 V11 V12 s2 V9 V10 V11 V12
1453: * < ncovcol=6 > nqv=2 (V7 V8) dv dv dv qtv dv dv dvv qtv
1454: * ntv=3 nqtv=1
1.330 brouard 1455: * cptcovn number of covariates (not including constant and age or age*age) = number of plus sign + 1 = 10+1=11
1.319 brouard 1456: * For time varying covariate, quanti or dummies
1457: * cotqvar[wav][iv(1 to nqtv)][i]= [1][12][i]=(V12) quanti
1458: * cotvar[wav][ntv+iv][i]= [3+(1 to nqtv)][i]=(V12) quanti
1459: * cotvar[wav][iv(1 to ntv)][i]= [1][1][i]=(V9) dummies at wav 1
1460: * cotvar[wav][iv(1 to ntv)][i]= [1][2][i]=(V10) dummies at wav 1
1.332 brouard 1461: * covar[Vk,i], value of the Vkth fixed covariate dummy or quanti for individual i:
1.319 brouard 1462: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
1463: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 + V9 + V9*age + V10
1464: * k= 1 2 3 4 5 6 7 8 9 10 11
1465: */
1466: /* According to the model, more columns can be added to covar by the product of covariates */
1.318 brouard 1467: /* 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
1468: # States 1=Coresidence, 2 Living alone, 3 Institution
1469: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1470: */
1.319 brouard 1471: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1472: /* k 1 2 3 4 5 6 7 8 9 */
1473: /*Typevar[k]= 0 0 0 2 1 0 2 1 0 *//*0 for simple covariate (dummy, quantitative,*/
1474: /* fixed or varying), 1 for age product, 2 for*/
1475: /* product */
1476: /*Dummy[k]= 1 0 0 1 3 1 1 2 0 *//*Dummy[k] 0=dummy (0 1), 1 quantitative */
1477: /*(single or product without age), 2 dummy*/
1478: /* with age product, 3 quant with age product*/
1479: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1480: /* nsd 1 2 3 */ /* Counting single dummies covar fixed or tv */
1.330 brouard 1481: /*TnsdVar[Tvar] 1 2 3 */
1.319 brouard 1482: /*TvarsD[nsd] 4 3 1 */ /* ID of single dummy cova fixed or timevary*/
1483: /*TvarsDind[k] 2 3 9 */ /* position K of single dummy cova */
1484: /* nsq 1 2 */ /* Counting single quantit tv */
1485: /* TvarsQ[k] 5 2 */ /* Number of single quantitative cova */
1486: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1487: /* Tprod[i]=k 1 2 */ /* Position in model of the ith prod without age */
1488: /* cptcovage 1 2 */ /* Counting cov*age in the model equation */
1489: /* Tage[cptcovage]=k 5 8 */ /* Position in the model of ith cov*age */
1490: /* Tvard[1][1]@4={4,3,1,2} V4*V3 V1*V2 */ /* Position in model of the ith prod without age */
1.330 brouard 1491: /* Tvardk[4][1]=4;Tvardk[4][2]=3;Tvardk[7][1]=1;Tvardk[7][2]=2 */ /* Variables of a prod at position in the model equation*/
1.319 brouard 1492: /* 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 1493: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.234 brouard 1494: /* Type */
1495: /* V 1 2 3 4 5 */
1496: /* F F V V V */
1497: /* D Q D D Q */
1498: /* */
1499: int *TvarsD;
1.330 brouard 1500: int *TnsdVar;
1.234 brouard 1501: int *TvarsDind;
1502: int *TvarsQ;
1503: int *TvarsQind;
1504:
1.318 brouard 1505: #define MAXRESULTLINESPONE 10+1
1.235 brouard 1506: int nresult=0;
1.258 brouard 1507: int parameterline=0; /* # of the parameter (type) line */
1.318 brouard 1508: int TKresult[MAXRESULTLINESPONE];
1.330 brouard 1509: int resultmodel[MAXRESULTLINESPONE][NCOVMAX];/* resultmodel[k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
1.318 brouard 1510: int Tresult[MAXRESULTLINESPONE][NCOVMAX];/* For dummy variable , value (output) */
1.332 brouard 1511: int Tinvresult[MAXRESULTLINESPONE][NCOVMAX];/* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1512: double TinvDoQresult[MAXRESULTLINESPONE][NCOVMAX];/* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.318 brouard 1513: int Tvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For dummy variable , variable # (output) */
1.332 brouard 1514: double Tqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.318 brouard 1515: double Tqinvresult[MAXRESULTLINESPONE][NCOVMAX]; /* For quantitative variable , value (output) */
1.332 brouard 1516: int Tvqresult[MAXRESULTLINESPONE][NCOVMAX]; /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.318 brouard 1517:
1518: /* 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
1519: # States 1=Coresidence, 2 Living alone, 3 Institution
1520: # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi
1521: */
1.234 brouard 1522: /* 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 1523: 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 */
1524: 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 */
1525: 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 */
1526: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1527: 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 */
1528: 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 1529: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1530: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1531: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1532: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1533: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1534: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1535: 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 */
1536: 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 */
1537:
1.230 brouard 1538: int *Tvarsel; /**< Selected covariates for output */
1539: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1540: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1541: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1542: 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 1543: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1544: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1545: int *Tage;
1.227 brouard 1546: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1547: 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 1548: 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*/
1549: 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 1550: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1551: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1552: int **Tvard;
1.330 brouard 1553: int **Tvardk;
1.227 brouard 1554: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1555: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1556: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1557: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1558: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1559: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1560: double *lsurv, *lpop, *tpop;
1561:
1.231 brouard 1562: #define FD 1; /* Fixed dummy covariate */
1563: #define FQ 2; /* Fixed quantitative covariate */
1564: #define FP 3; /* Fixed product covariate */
1565: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1566: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1567: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1568: #define VD 10; /* Varying dummy covariate */
1569: #define VQ 11; /* Varying quantitative covariate */
1570: #define VP 12; /* Varying product covariate */
1571: #define VPDD 13; /* Varying product dummy*dummy covariate */
1572: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1573: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1574: #define APFD 16; /* Age product * fixed dummy covariate */
1575: #define APFQ 17; /* Age product * fixed quantitative covariate */
1576: #define APVD 18; /* Age product * varying dummy covariate */
1577: #define APVQ 19; /* Age product * varying quantitative covariate */
1578:
1579: #define FTYPE 1; /* Fixed covariate */
1580: #define VTYPE 2; /* Varying covariate (loop in wave) */
1581: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1582:
1583: struct kmodel{
1584: int maintype; /* main type */
1585: int subtype; /* subtype */
1586: };
1587: struct kmodel modell[NCOVMAX];
1588:
1.143 brouard 1589: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1590: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1591:
1592: /**************** split *************************/
1593: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1594: {
1595: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1596: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1597: */
1598: char *ss; /* pointer */
1.186 brouard 1599: int l1=0, l2=0; /* length counters */
1.126 brouard 1600:
1601: l1 = strlen(path ); /* length of path */
1602: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1603: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1604: if ( ss == NULL ) { /* no directory, so determine current directory */
1605: strcpy( name, path ); /* we got the fullname name because no directory */
1606: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1607: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1608: /* get current working directory */
1609: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1610: #ifdef WIN32
1611: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1612: #else
1613: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1614: #endif
1.126 brouard 1615: return( GLOCK_ERROR_GETCWD );
1616: }
1617: /* got dirc from getcwd*/
1618: printf(" DIRC = %s \n",dirc);
1.205 brouard 1619: } else { /* strip directory from path */
1.126 brouard 1620: ss++; /* after this, the filename */
1621: l2 = strlen( ss ); /* length of filename */
1622: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1623: strcpy( name, ss ); /* save file name */
1624: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1625: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1626: printf(" DIRC2 = %s \n",dirc);
1627: }
1628: /* We add a separator at the end of dirc if not exists */
1629: l1 = strlen( dirc ); /* length of directory */
1630: if( dirc[l1-1] != DIRSEPARATOR ){
1631: dirc[l1] = DIRSEPARATOR;
1632: dirc[l1+1] = 0;
1633: printf(" DIRC3 = %s \n",dirc);
1634: }
1635: ss = strrchr( name, '.' ); /* find last / */
1636: if (ss >0){
1637: ss++;
1638: strcpy(ext,ss); /* save extension */
1639: l1= strlen( name);
1640: l2= strlen(ss)+1;
1641: strncpy( finame, name, l1-l2);
1642: finame[l1-l2]= 0;
1643: }
1644:
1645: return( 0 ); /* we're done */
1646: }
1647:
1648:
1649: /******************************************/
1650:
1651: void replace_back_to_slash(char *s, char*t)
1652: {
1653: int i;
1654: int lg=0;
1655: i=0;
1656: lg=strlen(t);
1657: for(i=0; i<= lg; i++) {
1658: (s[i] = t[i]);
1659: if (t[i]== '\\') s[i]='/';
1660: }
1661: }
1662:
1.132 brouard 1663: char *trimbb(char *out, char *in)
1.137 brouard 1664: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1665: char *s;
1666: s=out;
1667: while (*in != '\0'){
1.137 brouard 1668: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1669: in++;
1670: }
1671: *out++ = *in++;
1672: }
1673: *out='\0';
1674: return s;
1675: }
1676:
1.187 brouard 1677: /* char *substrchaine(char *out, char *in, char *chain) */
1678: /* { */
1679: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1680: /* char *s, *t; */
1681: /* t=in;s=out; */
1682: /* while ((*in != *chain) && (*in != '\0')){ */
1683: /* *out++ = *in++; */
1684: /* } */
1685:
1686: /* /\* *in matches *chain *\/ */
1687: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1688: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1689: /* } */
1690: /* in--; chain--; */
1691: /* while ( (*in != '\0')){ */
1692: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1693: /* *out++ = *in++; */
1694: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1695: /* } */
1696: /* *out='\0'; */
1697: /* out=s; */
1698: /* return out; */
1699: /* } */
1700: char *substrchaine(char *out, char *in, char *chain)
1701: {
1702: /* Substract chain 'chain' from 'in', return and output 'out' */
1703: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1704:
1705: char *strloc;
1706:
1707: strcpy (out, in);
1708: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1709: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1710: if(strloc != NULL){
1711: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1712: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1713: /* strcpy (strloc, strloc +strlen(chain));*/
1714: }
1715: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1716: return out;
1717: }
1718:
1719:
1.145 brouard 1720: char *cutl(char *blocc, char *alocc, char *in, char occ)
1721: {
1.187 brouard 1722: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1723: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1724: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1725: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1726: */
1.160 brouard 1727: char *s, *t;
1.145 brouard 1728: t=in;s=in;
1729: while ((*in != occ) && (*in != '\0')){
1730: *alocc++ = *in++;
1731: }
1732: if( *in == occ){
1733: *(alocc)='\0';
1734: s=++in;
1735: }
1736:
1737: if (s == t) {/* occ not found */
1738: *(alocc-(in-s))='\0';
1739: in=s;
1740: }
1741: while ( *in != '\0'){
1742: *blocc++ = *in++;
1743: }
1744:
1745: *blocc='\0';
1746: return t;
1747: }
1.137 brouard 1748: char *cutv(char *blocc, char *alocc, char *in, char occ)
1749: {
1.187 brouard 1750: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1751: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1752: gives blocc="abcdef2ghi" and alocc="j".
1753: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1754: */
1755: char *s, *t;
1756: t=in;s=in;
1757: while (*in != '\0'){
1758: while( *in == occ){
1759: *blocc++ = *in++;
1760: s=in;
1761: }
1762: *blocc++ = *in++;
1763: }
1764: if (s == t) /* occ not found */
1765: *(blocc-(in-s))='\0';
1766: else
1767: *(blocc-(in-s)-1)='\0';
1768: in=s;
1769: while ( *in != '\0'){
1770: *alocc++ = *in++;
1771: }
1772:
1773: *alocc='\0';
1774: return s;
1775: }
1776:
1.126 brouard 1777: int nbocc(char *s, char occ)
1778: {
1779: int i,j=0;
1780: int lg=20;
1781: i=0;
1782: lg=strlen(s);
1783: for(i=0; i<= lg; i++) {
1.234 brouard 1784: if (s[i] == occ ) j++;
1.126 brouard 1785: }
1786: return j;
1787: }
1788:
1.137 brouard 1789: /* void cutv(char *u,char *v, char*t, char occ) */
1790: /* { */
1791: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1792: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1793: /* gives u="abcdef2ghi" and v="j" *\/ */
1794: /* int i,lg,j,p=0; */
1795: /* i=0; */
1796: /* lg=strlen(t); */
1797: /* for(j=0; j<=lg-1; j++) { */
1798: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1799: /* } */
1.126 brouard 1800:
1.137 brouard 1801: /* for(j=0; j<p; j++) { */
1802: /* (u[j] = t[j]); */
1803: /* } */
1804: /* u[p]='\0'; */
1.126 brouard 1805:
1.137 brouard 1806: /* for(j=0; j<= lg; j++) { */
1807: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1808: /* } */
1809: /* } */
1.126 brouard 1810:
1.160 brouard 1811: #ifdef _WIN32
1812: char * strsep(char **pp, const char *delim)
1813: {
1814: char *p, *q;
1815:
1816: if ((p = *pp) == NULL)
1817: return 0;
1818: if ((q = strpbrk (p, delim)) != NULL)
1819: {
1820: *pp = q + 1;
1821: *q = '\0';
1822: }
1823: else
1824: *pp = 0;
1825: return p;
1826: }
1827: #endif
1828:
1.126 brouard 1829: /********************** nrerror ********************/
1830:
1831: void nrerror(char error_text[])
1832: {
1833: fprintf(stderr,"ERREUR ...\n");
1834: fprintf(stderr,"%s\n",error_text);
1835: exit(EXIT_FAILURE);
1836: }
1837: /*********************** vector *******************/
1838: double *vector(int nl, int nh)
1839: {
1840: double *v;
1841: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1842: if (!v) nrerror("allocation failure in vector");
1843: return v-nl+NR_END;
1844: }
1845:
1846: /************************ free vector ******************/
1847: void free_vector(double*v, int nl, int nh)
1848: {
1849: free((FREE_ARG)(v+nl-NR_END));
1850: }
1851:
1852: /************************ivector *******************************/
1853: int *ivector(long nl,long nh)
1854: {
1855: int *v;
1856: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1857: if (!v) nrerror("allocation failure in ivector");
1858: return v-nl+NR_END;
1859: }
1860:
1861: /******************free ivector **************************/
1862: void free_ivector(int *v, long nl, long nh)
1863: {
1864: free((FREE_ARG)(v+nl-NR_END));
1865: }
1866:
1867: /************************lvector *******************************/
1868: long *lvector(long nl,long nh)
1869: {
1870: long *v;
1871: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1872: if (!v) nrerror("allocation failure in ivector");
1873: return v-nl+NR_END;
1874: }
1875:
1876: /******************free lvector **************************/
1877: void free_lvector(long *v, long nl, long nh)
1878: {
1879: free((FREE_ARG)(v+nl-NR_END));
1880: }
1881:
1882: /******************* imatrix *******************************/
1883: int **imatrix(long nrl, long nrh, long ncl, long nch)
1884: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1885: {
1886: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1887: int **m;
1888:
1889: /* allocate pointers to rows */
1890: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1891: if (!m) nrerror("allocation failure 1 in matrix()");
1892: m += NR_END;
1893: m -= nrl;
1894:
1895:
1896: /* allocate rows and set pointers to them */
1897: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1898: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1899: m[nrl] += NR_END;
1900: m[nrl] -= ncl;
1901:
1902: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1903:
1904: /* return pointer to array of pointers to rows */
1905: return m;
1906: }
1907:
1908: /****************** free_imatrix *************************/
1909: void free_imatrix(m,nrl,nrh,ncl,nch)
1910: int **m;
1911: long nch,ncl,nrh,nrl;
1912: /* free an int matrix allocated by imatrix() */
1913: {
1914: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1915: free((FREE_ARG) (m+nrl-NR_END));
1916: }
1917:
1918: /******************* matrix *******************************/
1919: double **matrix(long nrl, long nrh, long ncl, long nch)
1920: {
1921: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1922: double **m;
1923:
1924: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1925: if (!m) nrerror("allocation failure 1 in matrix()");
1926: m += NR_END;
1927: m -= nrl;
1928:
1929: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1930: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1931: m[nrl] += NR_END;
1932: m[nrl] -= ncl;
1933:
1934: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1935: return m;
1.145 brouard 1936: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1937: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1938: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1939: */
1940: }
1941:
1942: /*************************free matrix ************************/
1943: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1944: {
1945: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1946: free((FREE_ARG)(m+nrl-NR_END));
1947: }
1948:
1949: /******************* ma3x *******************************/
1950: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1951: {
1952: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1953: double ***m;
1954:
1955: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1956: if (!m) nrerror("allocation failure 1 in matrix()");
1957: m += NR_END;
1958: m -= nrl;
1959:
1960: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1961: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1962: m[nrl] += NR_END;
1963: m[nrl] -= ncl;
1964:
1965: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1966:
1967: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1968: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1969: m[nrl][ncl] += NR_END;
1970: m[nrl][ncl] -= nll;
1971: for (j=ncl+1; j<=nch; j++)
1972: m[nrl][j]=m[nrl][j-1]+nlay;
1973:
1974: for (i=nrl+1; i<=nrh; i++) {
1975: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1976: for (j=ncl+1; j<=nch; j++)
1977: m[i][j]=m[i][j-1]+nlay;
1978: }
1979: return m;
1980: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1981: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1982: */
1983: }
1984:
1985: /*************************free ma3x ************************/
1986: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1987: {
1988: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1989: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1990: free((FREE_ARG)(m+nrl-NR_END));
1991: }
1992:
1993: /*************** function subdirf ***********/
1994: char *subdirf(char fileres[])
1995: {
1996: /* Caution optionfilefiname is hidden */
1997: strcpy(tmpout,optionfilefiname);
1998: strcat(tmpout,"/"); /* Add to the right */
1999: strcat(tmpout,fileres);
2000: return tmpout;
2001: }
2002:
2003: /*************** function subdirf2 ***********/
2004: char *subdirf2(char fileres[], char *preop)
2005: {
1.314 brouard 2006: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
2007: Errors in subdirf, 2, 3 while printing tmpout is
1.315 brouard 2008: rewritten within the same printf. Workaround: many printfs */
1.126 brouard 2009: /* Caution optionfilefiname is hidden */
2010: strcpy(tmpout,optionfilefiname);
2011: strcat(tmpout,"/");
2012: strcat(tmpout,preop);
2013: strcat(tmpout,fileres);
2014: return tmpout;
2015: }
2016:
2017: /*************** function subdirf3 ***********/
2018: char *subdirf3(char fileres[], char *preop, char *preop2)
2019: {
2020:
2021: /* Caution optionfilefiname is hidden */
2022: strcpy(tmpout,optionfilefiname);
2023: strcat(tmpout,"/");
2024: strcat(tmpout,preop);
2025: strcat(tmpout,preop2);
2026: strcat(tmpout,fileres);
2027: return tmpout;
2028: }
1.213 brouard 2029:
2030: /*************** function subdirfext ***********/
2031: char *subdirfext(char fileres[], char *preop, char *postop)
2032: {
2033:
2034: strcpy(tmpout,preop);
2035: strcat(tmpout,fileres);
2036: strcat(tmpout,postop);
2037: return tmpout;
2038: }
1.126 brouard 2039:
1.213 brouard 2040: /*************** function subdirfext3 ***********/
2041: char *subdirfext3(char fileres[], char *preop, char *postop)
2042: {
2043:
2044: /* Caution optionfilefiname is hidden */
2045: strcpy(tmpout,optionfilefiname);
2046: strcat(tmpout,"/");
2047: strcat(tmpout,preop);
2048: strcat(tmpout,fileres);
2049: strcat(tmpout,postop);
2050: return tmpout;
2051: }
2052:
1.162 brouard 2053: char *asc_diff_time(long time_sec, char ascdiff[])
2054: {
2055: long sec_left, days, hours, minutes;
2056: days = (time_sec) / (60*60*24);
2057: sec_left = (time_sec) % (60*60*24);
2058: hours = (sec_left) / (60*60) ;
2059: sec_left = (sec_left) %(60*60);
2060: minutes = (sec_left) /60;
2061: sec_left = (sec_left) % (60);
2062: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
2063: return ascdiff;
2064: }
2065:
1.126 brouard 2066: /***************** f1dim *************************/
2067: extern int ncom;
2068: extern double *pcom,*xicom;
2069: extern double (*nrfunc)(double []);
2070:
2071: double f1dim(double x)
2072: {
2073: int j;
2074: double f;
2075: double *xt;
2076:
2077: xt=vector(1,ncom);
2078: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
2079: f=(*nrfunc)(xt);
2080: free_vector(xt,1,ncom);
2081: return f;
2082: }
2083:
2084: /*****************brent *************************/
2085: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 2086: {
2087: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
2088: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
2089: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
2090: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
2091: * returned function value.
2092: */
1.126 brouard 2093: int iter;
2094: double a,b,d,etemp;
1.159 brouard 2095: double fu=0,fv,fw,fx;
1.164 brouard 2096: double ftemp=0.;
1.126 brouard 2097: double p,q,r,tol1,tol2,u,v,w,x,xm;
2098: double e=0.0;
2099:
2100: a=(ax < cx ? ax : cx);
2101: b=(ax > cx ? ax : cx);
2102: x=w=v=bx;
2103: fw=fv=fx=(*f)(x);
2104: for (iter=1;iter<=ITMAX;iter++) {
2105: xm=0.5*(a+b);
2106: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
2107: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
2108: printf(".");fflush(stdout);
2109: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 2110: #ifdef DEBUGBRENT
1.126 brouard 2111: 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);
2112: 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);
2113: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
2114: #endif
2115: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
2116: *xmin=x;
2117: return fx;
2118: }
2119: ftemp=fu;
2120: if (fabs(e) > tol1) {
2121: r=(x-w)*(fx-fv);
2122: q=(x-v)*(fx-fw);
2123: p=(x-v)*q-(x-w)*r;
2124: q=2.0*(q-r);
2125: if (q > 0.0) p = -p;
2126: q=fabs(q);
2127: etemp=e;
2128: e=d;
2129: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2130: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2131: else {
1.224 brouard 2132: d=p/q;
2133: u=x+d;
2134: if (u-a < tol2 || b-u < tol2)
2135: d=SIGN(tol1,xm-x);
1.126 brouard 2136: }
2137: } else {
2138: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2139: }
2140: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2141: fu=(*f)(u);
2142: if (fu <= fx) {
2143: if (u >= x) a=x; else b=x;
2144: SHFT(v,w,x,u)
1.183 brouard 2145: SHFT(fv,fw,fx,fu)
2146: } else {
2147: if (u < x) a=u; else b=u;
2148: if (fu <= fw || w == x) {
1.224 brouard 2149: v=w;
2150: w=u;
2151: fv=fw;
2152: fw=fu;
1.183 brouard 2153: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2154: v=u;
2155: fv=fu;
1.183 brouard 2156: }
2157: }
1.126 brouard 2158: }
2159: nrerror("Too many iterations in brent");
2160: *xmin=x;
2161: return fx;
2162: }
2163:
2164: /****************** mnbrak ***********************/
2165:
2166: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2167: double (*func)(double))
1.183 brouard 2168: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2169: the downhill direction (defined by the function as evaluated at the initial points) and returns
2170: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2171: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2172: */
1.126 brouard 2173: double ulim,u,r,q, dum;
2174: double fu;
1.187 brouard 2175:
2176: double scale=10.;
2177: int iterscale=0;
2178:
2179: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2180: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2181:
2182:
2183: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2184: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2185: /* *bx = *ax - (*ax - *bx)/scale; */
2186: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2187: /* } */
2188:
1.126 brouard 2189: if (*fb > *fa) {
2190: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2191: SHFT(dum,*fb,*fa,dum)
2192: }
1.126 brouard 2193: *cx=(*bx)+GOLD*(*bx-*ax);
2194: *fc=(*func)(*cx);
1.183 brouard 2195: #ifdef DEBUG
1.224 brouard 2196: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2197: 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 2198: #endif
1.224 brouard 2199: 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 2200: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2201: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2202: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2203: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2204: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2205: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2206: fu=(*func)(u);
1.163 brouard 2207: #ifdef DEBUG
2208: /* f(x)=A(x-u)**2+f(u) */
2209: double A, fparabu;
2210: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2211: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2212: 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);
2213: 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 2214: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2215: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2216: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2217: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2218: #endif
1.184 brouard 2219: #ifdef MNBRAKORIGINAL
1.183 brouard 2220: #else
1.191 brouard 2221: /* if (fu > *fc) { */
2222: /* #ifdef DEBUG */
2223: /* printf("mnbrak4 fu > fc \n"); */
2224: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2225: /* #endif */
2226: /* /\* 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 *\\/ *\/ */
2227: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2228: /* dum=u; /\* Shifting c and u *\/ */
2229: /* u = *cx; */
2230: /* *cx = dum; */
2231: /* dum = fu; */
2232: /* fu = *fc; */
2233: /* *fc =dum; */
2234: /* } else { /\* end *\/ */
2235: /* #ifdef DEBUG */
2236: /* printf("mnbrak3 fu < fc \n"); */
2237: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2238: /* #endif */
2239: /* dum=u; /\* Shifting c and u *\/ */
2240: /* u = *cx; */
2241: /* *cx = dum; */
2242: /* dum = fu; */
2243: /* fu = *fc; */
2244: /* *fc =dum; */
2245: /* } */
1.224 brouard 2246: #ifdef DEBUGMNBRAK
2247: double A, fparabu;
2248: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2249: fparabu= *fa - A*(*ax-u)*(*ax-u);
2250: 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);
2251: 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 2252: #endif
1.191 brouard 2253: dum=u; /* Shifting c and u */
2254: u = *cx;
2255: *cx = dum;
2256: dum = fu;
2257: fu = *fc;
2258: *fc =dum;
1.183 brouard 2259: #endif
1.162 brouard 2260: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2261: #ifdef DEBUG
1.224 brouard 2262: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2263: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2264: #endif
1.126 brouard 2265: fu=(*func)(u);
2266: if (fu < *fc) {
1.183 brouard 2267: #ifdef DEBUG
1.224 brouard 2268: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2269: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2270: #endif
2271: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2272: SHFT(*fb,*fc,fu,(*func)(u))
2273: #ifdef DEBUG
2274: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2275: #endif
2276: }
1.162 brouard 2277: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2278: #ifdef DEBUG
1.224 brouard 2279: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2280: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2281: #endif
1.126 brouard 2282: u=ulim;
2283: fu=(*func)(u);
1.183 brouard 2284: } else { /* u could be left to b (if r > q parabola has a maximum) */
2285: #ifdef DEBUG
1.224 brouard 2286: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2287: 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 2288: #endif
1.126 brouard 2289: u=(*cx)+GOLD*(*cx-*bx);
2290: fu=(*func)(u);
1.224 brouard 2291: #ifdef DEBUG
2292: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2293: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2294: #endif
1.183 brouard 2295: } /* end tests */
1.126 brouard 2296: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2297: SHFT(*fa,*fb,*fc,fu)
2298: #ifdef DEBUG
1.224 brouard 2299: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2300: 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 2301: #endif
2302: } /* 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 2303: }
2304:
2305: /*************** linmin ************************/
1.162 brouard 2306: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2307: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2308: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2309: the value of func at the returned location p . This is actually all accomplished by calling the
2310: routines mnbrak and brent .*/
1.126 brouard 2311: int ncom;
2312: double *pcom,*xicom;
2313: double (*nrfunc)(double []);
2314:
1.224 brouard 2315: #ifdef LINMINORIGINAL
1.126 brouard 2316: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2317: #else
2318: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2319: #endif
1.126 brouard 2320: {
2321: double brent(double ax, double bx, double cx,
2322: double (*f)(double), double tol, double *xmin);
2323: double f1dim(double x);
2324: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2325: double *fc, double (*func)(double));
2326: int j;
2327: double xx,xmin,bx,ax;
2328: double fx,fb,fa;
1.187 brouard 2329:
1.203 brouard 2330: #ifdef LINMINORIGINAL
2331: #else
2332: double scale=10., axs, xxs; /* Scale added for infinity */
2333: #endif
2334:
1.126 brouard 2335: ncom=n;
2336: pcom=vector(1,n);
2337: xicom=vector(1,n);
2338: nrfunc=func;
2339: for (j=1;j<=n;j++) {
2340: pcom[j]=p[j];
1.202 brouard 2341: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2342: }
1.187 brouard 2343:
1.203 brouard 2344: #ifdef LINMINORIGINAL
2345: xx=1.;
2346: #else
2347: axs=0.0;
2348: xxs=1.;
2349: do{
2350: xx= xxs;
2351: #endif
1.187 brouard 2352: ax=0.;
2353: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2354: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2355: /* 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)) */
2356: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2357: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2358: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2359: /* 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 2360: #ifdef LINMINORIGINAL
2361: #else
2362: if (fx != fx){
1.224 brouard 2363: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2364: printf("|");
2365: fprintf(ficlog,"|");
1.203 brouard 2366: #ifdef DEBUGLINMIN
1.224 brouard 2367: 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 2368: #endif
2369: }
1.224 brouard 2370: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2371: #endif
2372:
1.191 brouard 2373: #ifdef DEBUGLINMIN
2374: 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 2375: 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 2376: #endif
1.224 brouard 2377: #ifdef LINMINORIGINAL
2378: #else
1.317 brouard 2379: if(fb == fx){ /* Flat function in the direction */
2380: xmin=xx;
1.224 brouard 2381: *flat=1;
1.317 brouard 2382: }else{
1.224 brouard 2383: *flat=0;
2384: #endif
2385: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2386: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2387: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2388: /* fmin = f(p[j] + xmin * xi[j]) */
2389: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2390: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2391: #ifdef DEBUG
1.224 brouard 2392: 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);
2393: 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);
2394: #endif
2395: #ifdef LINMINORIGINAL
2396: #else
2397: }
1.126 brouard 2398: #endif
1.191 brouard 2399: #ifdef DEBUGLINMIN
2400: printf("linmin end ");
1.202 brouard 2401: fprintf(ficlog,"linmin end ");
1.191 brouard 2402: #endif
1.126 brouard 2403: for (j=1;j<=n;j++) {
1.203 brouard 2404: #ifdef LINMINORIGINAL
2405: xi[j] *= xmin;
2406: #else
2407: #ifdef DEBUGLINMIN
2408: if(xxs <1.0)
2409: printf(" before xi[%d]=%12.8f", j,xi[j]);
2410: #endif
2411: 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) */
2412: #ifdef DEBUGLINMIN
2413: if(xxs <1.0)
2414: 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 );
2415: #endif
2416: #endif
1.187 brouard 2417: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2418: }
1.191 brouard 2419: #ifdef DEBUGLINMIN
1.203 brouard 2420: printf("\n");
1.191 brouard 2421: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2422: 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 2423: for (j=1;j<=n;j++) {
1.202 brouard 2424: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2425: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2426: if(j % ncovmodel == 0){
1.191 brouard 2427: printf("\n");
1.202 brouard 2428: fprintf(ficlog,"\n");
2429: }
1.191 brouard 2430: }
1.203 brouard 2431: #else
1.191 brouard 2432: #endif
1.126 brouard 2433: free_vector(xicom,1,n);
2434: free_vector(pcom,1,n);
2435: }
2436:
2437:
2438: /*************** powell ************************/
1.162 brouard 2439: /*
1.317 brouard 2440: Minimization of a function func of n variables. Input consists in an initial starting point
2441: p[1..n] ; an initial matrix xi[1..n][1..n] whose columns contain the initial set of di-
2442: rections (usually the n unit vectors); and ftol, the fractional tolerance in the function value
2443: such that failure to decrease by more than this amount in one iteration signals doneness. On
1.162 brouard 2444: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2445: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2446: */
1.224 brouard 2447: #ifdef LINMINORIGINAL
2448: #else
2449: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2450: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2451: #endif
1.126 brouard 2452: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2453: double (*func)(double []))
2454: {
1.224 brouard 2455: #ifdef LINMINORIGINAL
2456: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2457: double (*func)(double []));
1.224 brouard 2458: #else
1.241 brouard 2459: void linmin(double p[], double xi[], int n, double *fret,
2460: double (*func)(double []),int *flat);
1.224 brouard 2461: #endif
1.239 brouard 2462: int i,ibig,j,jk,k;
1.126 brouard 2463: double del,t,*pt,*ptt,*xit;
1.181 brouard 2464: double directest;
1.126 brouard 2465: double fp,fptt;
2466: double *xits;
2467: int niterf, itmp;
2468:
2469: pt=vector(1,n);
2470: ptt=vector(1,n);
2471: xit=vector(1,n);
2472: xits=vector(1,n);
2473: *fret=(*func)(p);
2474: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2475: rcurr_time = time(NULL);
1.126 brouard 2476: for (*iter=1;;++(*iter)) {
2477: ibig=0;
2478: del=0.0;
1.157 brouard 2479: rlast_time=rcurr_time;
2480: /* (void) gettimeofday(&curr_time,&tzp); */
2481: rcurr_time = time(NULL);
2482: curr_time = *localtime(&rcurr_time);
1.324 brouard 2483: 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);
2484: 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 2485: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.324 brouard 2486: fp=(*fret); /* From former iteration or initial value */
1.192 brouard 2487: for (i=1;i<=n;i++) {
1.126 brouard 2488: fprintf(ficrespow," %.12lf", p[i]);
2489: }
1.239 brouard 2490: fprintf(ficrespow,"\n");fflush(ficrespow);
2491: printf("\n#model= 1 + age ");
2492: fprintf(ficlog,"\n#model= 1 + age ");
2493: if(nagesqr==1){
1.241 brouard 2494: printf(" + age*age ");
2495: fprintf(ficlog," + age*age ");
1.239 brouard 2496: }
2497: for(j=1;j <=ncovmodel-2;j++){
2498: if(Typevar[j]==0) {
2499: printf(" + V%d ",Tvar[j]);
2500: fprintf(ficlog," + V%d ",Tvar[j]);
2501: }else if(Typevar[j]==1) {
2502: printf(" + V%d*age ",Tvar[j]);
2503: fprintf(ficlog," + V%d*age ",Tvar[j]);
2504: }else if(Typevar[j]==2) {
2505: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2506: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2507: }
2508: }
1.126 brouard 2509: printf("\n");
1.239 brouard 2510: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2511: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2512: fprintf(ficlog,"\n");
1.239 brouard 2513: for(i=1,jk=1; i <=nlstate; i++){
2514: for(k=1; k <=(nlstate+ndeath); k++){
2515: if (k != i) {
2516: printf("%d%d ",i,k);
2517: fprintf(ficlog,"%d%d ",i,k);
2518: for(j=1; j <=ncovmodel; j++){
2519: printf("%12.7f ",p[jk]);
2520: fprintf(ficlog,"%12.7f ",p[jk]);
2521: jk++;
2522: }
2523: printf("\n");
2524: fprintf(ficlog,"\n");
2525: }
2526: }
2527: }
1.241 brouard 2528: if(*iter <=3 && *iter >1){
1.157 brouard 2529: tml = *localtime(&rcurr_time);
2530: strcpy(strcurr,asctime(&tml));
2531: rforecast_time=rcurr_time;
1.126 brouard 2532: itmp = strlen(strcurr);
2533: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2534: strcurr[itmp-1]='\0';
1.162 brouard 2535: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2536: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2537: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2538: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2539: forecast_time = *localtime(&rforecast_time);
2540: strcpy(strfor,asctime(&forecast_time));
2541: itmp = strlen(strfor);
2542: if(strfor[itmp-1]=='\n')
2543: strfor[itmp-1]='\0';
2544: 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);
2545: 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 2546: }
2547: }
1.187 brouard 2548: for (i=1;i<=n;i++) { /* For each direction i */
2549: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2550: fptt=(*fret);
2551: #ifdef DEBUG
1.203 brouard 2552: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2553: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2554: #endif
1.203 brouard 2555: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2556: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2557: #ifdef LINMINORIGINAL
1.188 brouard 2558: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2559: #else
2560: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2561: flatdir[i]=flat; /* Function is vanishing in that direction i */
2562: #endif
2563: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2564: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2565: /* because that direction will be replaced unless the gain del is small */
2566: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2567: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2568: /* with the new direction. */
2569: del=fabs(fptt-(*fret));
2570: ibig=i;
1.126 brouard 2571: }
2572: #ifdef DEBUG
2573: printf("%d %.12e",i,(*fret));
2574: fprintf(ficlog,"%d %.12e",i,(*fret));
2575: for (j=1;j<=n;j++) {
1.224 brouard 2576: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2577: printf(" x(%d)=%.12e",j,xit[j]);
2578: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2579: }
2580: for(j=1;j<=n;j++) {
1.225 brouard 2581: printf(" p(%d)=%.12e",j,p[j]);
2582: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2583: }
2584: printf("\n");
2585: fprintf(ficlog,"\n");
2586: #endif
1.187 brouard 2587: } /* end loop on each direction i */
2588: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2589: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2590: /* New value of last point Pn is not computed, P(n-1) */
1.319 brouard 2591: for(j=1;j<=n;j++) {
2592: if(flatdir[j] >0){
2593: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2594: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
1.302 brouard 2595: }
1.319 brouard 2596: /* printf("\n"); */
2597: /* fprintf(ficlog,"\n"); */
2598: }
1.243 brouard 2599: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2600: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2601: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2602: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2603: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2604: /* decreased of more than 3.84 */
2605: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2606: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2607: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2608:
1.188 brouard 2609: /* Starting the program with initial values given by a former maximization will simply change */
2610: /* the scales of the directions and the directions, because the are reset to canonical directions */
2611: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2612: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2613: #ifdef DEBUG
2614: int k[2],l;
2615: k[0]=1;
2616: k[1]=-1;
2617: printf("Max: %.12e",(*func)(p));
2618: fprintf(ficlog,"Max: %.12e",(*func)(p));
2619: for (j=1;j<=n;j++) {
2620: printf(" %.12e",p[j]);
2621: fprintf(ficlog," %.12e",p[j]);
2622: }
2623: printf("\n");
2624: fprintf(ficlog,"\n");
2625: for(l=0;l<=1;l++) {
2626: for (j=1;j<=n;j++) {
2627: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2628: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2629: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2630: }
2631: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2632: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2633: }
2634: #endif
2635:
2636: free_vector(xit,1,n);
2637: free_vector(xits,1,n);
2638: free_vector(ptt,1,n);
2639: free_vector(pt,1,n);
2640: return;
1.192 brouard 2641: } /* enough precision */
1.240 brouard 2642: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2643: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2644: ptt[j]=2.0*p[j]-pt[j];
2645: xit[j]=p[j]-pt[j];
2646: pt[j]=p[j];
2647: }
1.181 brouard 2648: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2649: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2650: if (*iter <=4) {
1.225 brouard 2651: #else
2652: #endif
1.224 brouard 2653: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2654: #else
1.161 brouard 2655: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2656: #endif
1.162 brouard 2657: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2658: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2659: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2660: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2661: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2662: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2663: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2664: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2665: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2666: /* Even if f3 <f1, directest can be negative and t >0 */
2667: /* mu² and del² are equal when f3=f1 */
2668: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2669: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2670: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2671: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2672: #ifdef NRCORIGINAL
2673: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2674: #else
2675: 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 2676: t= t- del*SQR(fp-fptt);
1.183 brouard 2677: #endif
1.202 brouard 2678: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2679: #ifdef DEBUG
1.181 brouard 2680: 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);
2681: 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 2682: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2683: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2684: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2685: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2686: 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);
2687: 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);
2688: #endif
1.183 brouard 2689: #ifdef POWELLORIGINAL
2690: if (t < 0.0) { /* Then we use it for new direction */
2691: #else
1.182 brouard 2692: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2693: 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 2694: 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 2695: 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 2696: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2697: }
1.181 brouard 2698: if (directest < 0.0) { /* Then we use it for new direction */
2699: #endif
1.191 brouard 2700: #ifdef DEBUGLINMIN
1.234 brouard 2701: printf("Before linmin in direction P%d-P0\n",n);
2702: for (j=1;j<=n;j++) {
2703: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2704: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2705: if(j % ncovmodel == 0){
2706: printf("\n");
2707: fprintf(ficlog,"\n");
2708: }
2709: }
1.224 brouard 2710: #endif
2711: #ifdef LINMINORIGINAL
1.234 brouard 2712: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2713: #else
1.234 brouard 2714: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2715: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2716: #endif
1.234 brouard 2717:
1.191 brouard 2718: #ifdef DEBUGLINMIN
1.234 brouard 2719: for (j=1;j<=n;j++) {
2720: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2721: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2722: if(j % ncovmodel == 0){
2723: printf("\n");
2724: fprintf(ficlog,"\n");
2725: }
2726: }
1.224 brouard 2727: #endif
1.234 brouard 2728: for (j=1;j<=n;j++) {
2729: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2730: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2731: }
1.224 brouard 2732: #ifdef LINMINORIGINAL
2733: #else
1.234 brouard 2734: for (j=1, flatd=0;j<=n;j++) {
2735: if(flatdir[j]>0)
2736: flatd++;
2737: }
2738: if(flatd >0){
1.255 brouard 2739: printf("%d flat directions: ",flatd);
2740: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2741: for (j=1;j<=n;j++) {
2742: if(flatdir[j]>0){
2743: printf("%d ",j);
2744: fprintf(ficlog,"%d ",j);
2745: }
2746: }
2747: printf("\n");
2748: fprintf(ficlog,"\n");
1.319 brouard 2749: #ifdef FLATSUP
2750: free_vector(xit,1,n);
2751: free_vector(xits,1,n);
2752: free_vector(ptt,1,n);
2753: free_vector(pt,1,n);
2754: return;
2755: #endif
1.234 brouard 2756: }
1.191 brouard 2757: #endif
1.234 brouard 2758: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2759: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2760:
1.126 brouard 2761: #ifdef DEBUG
1.234 brouard 2762: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2763: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2764: for(j=1;j<=n;j++){
2765: printf(" %lf",xit[j]);
2766: fprintf(ficlog," %lf",xit[j]);
2767: }
2768: printf("\n");
2769: fprintf(ficlog,"\n");
1.126 brouard 2770: #endif
1.192 brouard 2771: } /* end of t or directest negative */
1.224 brouard 2772: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2773: #else
1.234 brouard 2774: } /* end if (fptt < fp) */
1.192 brouard 2775: #endif
1.225 brouard 2776: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2777: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2778: #else
1.224 brouard 2779: #endif
1.234 brouard 2780: } /* loop iteration */
1.126 brouard 2781: }
1.234 brouard 2782:
1.126 brouard 2783: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2784:
1.235 brouard 2785: 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 2786: {
1.279 brouard 2787: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2788: * (and selected quantitative values in nres)
2789: * by left multiplying the unit
2790: * matrix by transitions matrix until convergence is reached with precision ftolpl
2791: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2792: * Wx is row vector: population in state 1, population in state 2, population dead
2793: * or prevalence in state 1, prevalence in state 2, 0
2794: * newm is the matrix after multiplications, its rows are identical at a factor.
2795: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2796: * Output is prlim.
2797: * Initial matrix pimij
2798: */
1.206 brouard 2799: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2800: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2801: /* 0, 0 , 1} */
2802: /*
2803: * and after some iteration: */
2804: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2805: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2806: /* 0, 0 , 1} */
2807: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2808: /* {0.51571254859325999, 0.4842874514067399, */
2809: /* 0.51326036147820708, 0.48673963852179264} */
2810: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2811:
1.332 brouard 2812: int i, ii,j,k, k1;
1.209 brouard 2813: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2814: /* double **matprod2(); */ /* test */
1.218 brouard 2815: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2816: double **newm;
1.209 brouard 2817: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2818: int ncvloop=0;
1.288 brouard 2819: int first=0;
1.169 brouard 2820:
1.209 brouard 2821: min=vector(1,nlstate);
2822: max=vector(1,nlstate);
2823: meandiff=vector(1,nlstate);
2824:
1.218 brouard 2825: /* Starting with matrix unity */
1.126 brouard 2826: for (ii=1;ii<=nlstate+ndeath;ii++)
2827: for (j=1;j<=nlstate+ndeath;j++){
2828: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2829: }
1.169 brouard 2830:
2831: cov[1]=1.;
2832:
2833: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2834: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2835: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2836: ncvloop++;
1.126 brouard 2837: newm=savm;
2838: /* Covariates have to be included here again */
1.138 brouard 2839: cov[2]=agefin;
1.319 brouard 2840: if(nagesqr==1){
2841: cov[3]= agefin*agefin;
2842: }
1.332 brouard 2843: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
2844: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
2845: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
2846: if(Typevar[k1]==1){ /* A product with age */
2847: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
2848: }else{
2849: cov[2+nagesqr+k1]=precov[nres][k1];
2850: }
2851: }/* End of loop on model equation */
2852:
2853: /* Start of old code (replaced by a loop on position in the model equation */
2854: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only of the model *\/ */
2855: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
2856: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; *\/ */
2857: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])]; */
2858: /* /\* model = 1 +age + V1*V3 + age*V1 + V2 + V1 + age*V2 + V3 + V3*age + V1*V2 */
2859: /* * k 1 2 3 4 5 6 7 8 */
2860: /* *cov[] 1 2 3 4 5 6 7 8 9 10 */
2861: /* *TypeVar[k] 2 1 0 0 1 0 1 2 */
2862: /* *Dummy[k] 0 2 0 0 2 0 2 0 */
2863: /* *Tvar[k] 4 1 2 1 2 3 3 5 */
2864: /* *nsd=3 (1) (2) (3) */
2865: /* *TvarsD[nsd] [1]=2 1 3 */
2866: /* *TnsdVar [2]=2 [1]=1 [3]=3 */
2867: /* *TvarsDind[nsd](=k) [1]=3 [2]=4 [3]=6 */
2868: /* *Tage[] [1]=1 [2]=2 [3]=3 */
2869: /* *Tvard[] [1][1]=1 [2][1]=1 */
2870: /* * [1][2]=3 [2][2]=2 */
2871: /* *Tprod[](=k) [1]=1 [2]=8 */
2872: /* *TvarsDp(=Tvar) [1]=1 [2]=2 [3]=3 [4]=5 */
2873: /* *TvarD (=k) [1]=1 [2]=3 [3]=4 [3]=6 [4]=6 */
2874: /* *TvarsDpType */
2875: /* *si model= 1 + age + V3 + V2*age + V2 + V3*age */
2876: /* * nsd=1 (1) (2) */
2877: /* *TvarsD[nsd] 3 2 */
2878: /* *TnsdVar (3)=1 (2)=2 */
2879: /* *TvarsDind[nsd](=k) [1]=1 [2]=3 */
2880: /* *Tage[] [1]=2 [2]= 3 */
2881: /* *\/ */
2882: /* /\* cov[++k1]=nbcode[TvarsD[k]][codtabm(ij,k)]; *\/ */
2883: /* /\* 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)); *\/ */
2884: /* } */
2885: /* for (k=1; k<=nsq;k++) { /\* For single quantitative varying covariates only of the model *\/ */
2886: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
2887: /* /\* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline *\/ */
2888: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
2889: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][resultmodel[nres][k1]] */
2890: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2891: /* /\* 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]); *\/ */
2892: /* } */
2893: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
2894: /* if(Dummy[Tage[k]]==2){ /\* dummy with age *\/ */
2895: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
2896: /* /\* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\/ */
2897: /* } else if(Dummy[Tage[k]]==3){ /\* quantitative with age *\/ */
2898: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
2899: /* /\* cov[++k1]=Tqresult[nres][k]; *\/ */
2900: /* } */
2901: /* /\* 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]); *\/ */
2902: /* } */
2903: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
2904: /* /\* 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]); *\/ */
2905: /* if(Dummy[Tvard[k][1]]==0){ */
2906: /* if(Dummy[Tvard[k][2]]==0){ */
2907: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
2908: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
2909: /* }else{ */
2910: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
2911: /* /\* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; *\/ */
2912: /* } */
2913: /* }else{ */
2914: /* if(Dummy[Tvard[k][2]]==0){ */
2915: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
2916: /* /\* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; *\/ */
2917: /* }else{ */
2918: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
2919: /* /\* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\/ */
2920: /* } */
2921: /* } */
2922: /* } /\* End product without age *\/ */
2923: /* ENd of old code */
1.138 brouard 2924: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2925: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2926: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2927: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2928: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.319 brouard 2929: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2930: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2931:
1.126 brouard 2932: savm=oldm;
2933: oldm=newm;
1.209 brouard 2934:
2935: for(j=1; j<=nlstate; j++){
2936: max[j]=0.;
2937: min[j]=1.;
2938: }
2939: for(i=1;i<=nlstate;i++){
2940: sumnew=0;
2941: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2942: for(j=1; j<=nlstate; j++){
2943: prlim[i][j]= newm[i][j]/(1-sumnew);
2944: max[j]=FMAX(max[j],prlim[i][j]);
2945: min[j]=FMIN(min[j],prlim[i][j]);
2946: }
2947: }
2948:
1.126 brouard 2949: maxmax=0.;
1.209 brouard 2950: for(j=1; j<=nlstate; j++){
2951: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2952: maxmax=FMAX(maxmax,meandiff[j]);
2953: /* 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 2954: } /* j loop */
1.203 brouard 2955: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2956: /* 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 2957: if(maxmax < ftolpl){
1.209 brouard 2958: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2959: free_vector(min,1,nlstate);
2960: free_vector(max,1,nlstate);
2961: free_vector(meandiff,1,nlstate);
1.126 brouard 2962: return prlim;
2963: }
1.288 brouard 2964: } /* agefin loop */
1.208 brouard 2965: /* After some age loop it doesn't converge */
1.288 brouard 2966: if(!first){
2967: first=1;
2968: 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 2969: 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);
2970: }else if (first >=1 && first <10){
2971: 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);
2972: first++;
2973: }else if (first ==10){
2974: 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);
2975: 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");
2976: fprintf(ficlog,"Warning: the stable prevalence no convergence; too many cases, giving up noticing, even in log file\n");
2977: first++;
1.288 brouard 2978: }
2979:
1.209 brouard 2980: /* 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); */
2981: free_vector(min,1,nlstate);
2982: free_vector(max,1,nlstate);
2983: free_vector(meandiff,1,nlstate);
1.208 brouard 2984:
1.169 brouard 2985: return prlim; /* should not reach here */
1.126 brouard 2986: }
2987:
1.217 brouard 2988:
2989: /**** Back Prevalence limit (stable or period prevalence) ****************/
2990:
1.218 brouard 2991: /* 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) */
2992: /* 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 2993: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2994: {
1.264 brouard 2995: /* 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 2996: matrix by transitions matrix until convergence is reached with precision ftolpl */
2997: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2998: /* Wx is row vector: population in state 1, population in state 2, population dead */
2999: /* or prevalence in state 1, prevalence in state 2, 0 */
3000: /* newm is the matrix after multiplications, its rows are identical at a factor */
3001: /* Initial matrix pimij */
3002: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
3003: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
3004: /* 0, 0 , 1} */
3005: /*
3006: * and after some iteration: */
3007: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
3008: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
3009: /* 0, 0 , 1} */
3010: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
3011: /* {0.51571254859325999, 0.4842874514067399, */
3012: /* 0.51326036147820708, 0.48673963852179264} */
3013: /* If we start from prlim again, prlim tends to a constant matrix */
3014:
1.332 brouard 3015: int i, ii,j,k, k1;
1.247 brouard 3016: int first=0;
1.217 brouard 3017: double *min, *max, *meandiff, maxmax,sumnew=0.;
3018: /* double **matprod2(); */ /* test */
3019: double **out, cov[NCOVMAX+1], **bmij();
3020: double **newm;
1.218 brouard 3021: double **dnewm, **doldm, **dsavm; /* for use */
3022: double **oldm, **savm; /* for use */
3023:
1.217 brouard 3024: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
3025: int ncvloop=0;
3026:
3027: min=vector(1,nlstate);
3028: max=vector(1,nlstate);
3029: meandiff=vector(1,nlstate);
3030:
1.266 brouard 3031: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
3032: oldm=oldms; savm=savms;
3033:
3034: /* Starting with matrix unity */
3035: for (ii=1;ii<=nlstate+ndeath;ii++)
3036: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 3037: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3038: }
3039:
3040: cov[1]=1.;
3041:
3042: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3043: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 3044: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 3045: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
3046: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 3047: ncvloop++;
1.218 brouard 3048: newm=savm; /* oldm should be kept from previous iteration or unity at start */
3049: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 3050: /* Covariates have to be included here again */
3051: cov[2]=agefin;
1.319 brouard 3052: if(nagesqr==1){
1.217 brouard 3053: cov[3]= agefin*agefin;;
1.319 brouard 3054: }
1.332 brouard 3055: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3056: if(Typevar[k1]==1){ /* A product with age */
3057: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.242 brouard 3058: }else{
1.332 brouard 3059: cov[2+nagesqr+k1]=precov[nres][k1];
1.242 brouard 3060: }
1.332 brouard 3061: }/* End of loop on model equation */
3062:
3063: /* Old code */
3064:
3065: /* for (k=1; k<=nsd;k++) { /\* For single dummy covariates only *\/ */
3066: /* /\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\/ */
3067: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])]; */
3068: /* /\* 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)); *\/ */
3069: /* } */
3070: /* /\* for (k=1; k<=cptcovn;k++) { *\/ */
3071: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3072: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3073: /* /\* /\\* 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])]); *\\/ *\/ */
3074: /* /\* } *\/ */
3075: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3076: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3077: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3078: /* /\* 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]); *\/ */
3079: /* } */
3080: /* /\* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; *\/ */
3081: /* /\* for (k=1; k<=cptcovprod;k++) /\\* Useless *\\/ *\/ */
3082: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\\/ *\/ */
3083: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3084: /* for (k=1; k<=cptcovage;k++){ /\* For product with age *\/ */
3085: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age *\\/ ERROR ???*\/ */
3086: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3087: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3088: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3089: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3090: /* } */
3091: /* /\* 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]); *\/ */
3092: /* } */
3093: /* for (k=1; k<=cptcovprod;k++){ /\* For product without age *\/ */
3094: /* /\* 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]); *\/ */
3095: /* if(Dummy[Tvard[k][1]]==0){ */
3096: /* if(Dummy[Tvard[k][2]]==0){ */
3097: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3098: /* }else{ */
3099: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3100: /* } */
3101: /* }else{ */
3102: /* if(Dummy[Tvard[k][2]]==0){ */
3103: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3104: /* }else{ */
3105: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3106: /* } */
3107: /* } */
3108: /* } */
1.217 brouard 3109:
3110: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
3111: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
3112: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
3113: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3114: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 3115: /* ij should be linked to the correct index of cov */
3116: /* age and covariate values ij are in 'cov', but we need to pass
3117: * ij for the observed prevalence at age and status and covariate
3118: * number: prevacurrent[(int)agefin][ii][ij]
3119: */
3120: /* 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 *\/ */
3121: /* 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 *\/ */
3122: 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 3123: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 3124: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
3125: /* for(i=1; i<=nlstate+ndeath; i++) { */
3126: /* printf("%d newm= ",i); */
3127: /* for(j=1;j<=nlstate+ndeath;j++) { */
3128: /* printf("%f ",newm[i][j]); */
3129: /* } */
3130: /* printf("oldm * "); */
3131: /* for(j=1;j<=nlstate+ndeath;j++) { */
3132: /* printf("%f ",oldm[i][j]); */
3133: /* } */
1.268 brouard 3134: /* printf(" bmmij "); */
1.266 brouard 3135: /* for(j=1;j<=nlstate+ndeath;j++) { */
3136: /* printf("%f ",pmmij[i][j]); */
3137: /* } */
3138: /* printf("\n"); */
3139: /* } */
3140: /* } */
1.217 brouard 3141: savm=oldm;
3142: oldm=newm;
1.266 brouard 3143:
1.217 brouard 3144: for(j=1; j<=nlstate; j++){
3145: max[j]=0.;
3146: min[j]=1.;
3147: }
3148: for(j=1; j<=nlstate; j++){
3149: for(i=1;i<=nlstate;i++){
1.234 brouard 3150: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
3151: bprlim[i][j]= newm[i][j];
3152: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
3153: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 3154: }
3155: }
1.218 brouard 3156:
1.217 brouard 3157: maxmax=0.;
3158: for(i=1; i<=nlstate; i++){
1.318 brouard 3159: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column, could be nan! */
1.217 brouard 3160: maxmax=FMAX(maxmax,meandiff[i]);
3161: /* 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 3162: } /* i loop */
1.217 brouard 3163: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 3164: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3165: if(maxmax < ftolpl){
1.220 brouard 3166: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 3167: free_vector(min,1,nlstate);
3168: free_vector(max,1,nlstate);
3169: free_vector(meandiff,1,nlstate);
3170: return bprlim;
3171: }
1.288 brouard 3172: } /* agefin loop */
1.217 brouard 3173: /* After some age loop it doesn't converge */
1.288 brouard 3174: if(!first){
1.247 brouard 3175: first=1;
3176: 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\
3177: 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);
3178: }
3179: 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 3180: 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);
3181: /* 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); */
3182: free_vector(min,1,nlstate);
3183: free_vector(max,1,nlstate);
3184: free_vector(meandiff,1,nlstate);
3185:
3186: return bprlim; /* should not reach here */
3187: }
3188:
1.126 brouard 3189: /*************** transition probabilities ***************/
3190:
3191: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3192: {
1.138 brouard 3193: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3194: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3195: model to the ncovmodel covariates (including constant and age).
3196: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3197: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3198: ncth covariate in the global vector x is given by the formula:
3199: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3200: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3201: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3202: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3203: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3204: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3205: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3206: */
3207: double s1, lnpijopii;
1.126 brouard 3208: /*double t34;*/
1.164 brouard 3209: int i,j, nc, ii, jj;
1.126 brouard 3210:
1.223 brouard 3211: for(i=1; i<= nlstate; i++){
3212: for(j=1; j<i;j++){
3213: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3214: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3215: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3216: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3217: }
3218: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3219: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3220: }
3221: for(j=i+1; j<=nlstate+ndeath;j++){
3222: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3223: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3224: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3225: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3226: }
3227: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
1.330 brouard 3228: /* printf("Debug pmij() i=%d j=%d nc=%d s1=%.17f, lnpijopii=%.17f\n",i,j,nc, s1,lnpijopii); */
1.223 brouard 3229: }
3230: }
1.218 brouard 3231:
1.223 brouard 3232: for(i=1; i<= nlstate; i++){
3233: s1=0;
3234: for(j=1; j<i; j++){
3235: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 brouard 3236: /* printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
1.223 brouard 3237: }
3238: for(j=i+1; j<=nlstate+ndeath; j++){
3239: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
1.330 brouard 3240: /* printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
1.223 brouard 3241: }
3242: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3243: ps[i][i]=1./(s1+1.);
3244: /* Computing other pijs */
3245: for(j=1; j<i; j++)
1.325 brouard 3246: ps[i][j]= exp(ps[i][j])*ps[i][i];/* Bug valgrind */
1.223 brouard 3247: for(j=i+1; j<=nlstate+ndeath; j++)
3248: ps[i][j]= exp(ps[i][j])*ps[i][i];
3249: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3250: } /* end i */
1.218 brouard 3251:
1.223 brouard 3252: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3253: for(jj=1; jj<= nlstate+ndeath; jj++){
3254: ps[ii][jj]=0;
3255: ps[ii][ii]=1;
3256: }
3257: }
1.294 brouard 3258:
3259:
1.223 brouard 3260: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3261: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3262: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3263: /* } */
3264: /* printf("\n "); */
3265: /* } */
3266: /* printf("\n ");printf("%lf ",cov[2]);*/
3267: /*
3268: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3269: goto end;*/
1.266 brouard 3270: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3271: }
3272:
1.218 brouard 3273: /*************** backward transition probabilities ***************/
3274:
3275: /* 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 ) */
3276: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3277: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3278: {
1.302 brouard 3279: /* 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 3280: * 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 3281: */
1.218 brouard 3282: int i, ii, j,k;
1.222 brouard 3283:
3284: double **out, **pmij();
3285: double sumnew=0.;
1.218 brouard 3286: double agefin;
1.292 brouard 3287: 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 3288: double **dnewm, **dsavm, **doldm;
3289: double **bbmij;
3290:
1.218 brouard 3291: doldm=ddoldms; /* global pointers */
1.222 brouard 3292: dnewm=ddnewms;
3293: dsavm=ddsavms;
1.318 brouard 3294:
3295: /* Debug */
3296: /* printf("Bmij ij=%d, cov[2}=%f\n", ij, cov[2]); */
1.222 brouard 3297: agefin=cov[2];
1.268 brouard 3298: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3299: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3300: the observed prevalence (with this covariate ij) at beginning of transition */
3301: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3302:
3303: /* P_x */
1.325 brouard 3304: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm *//* Bug valgrind */
1.268 brouard 3305: /* outputs pmmij which is a stochastic matrix in row */
3306:
3307: /* Diag(w_x) */
1.292 brouard 3308: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3309: sumnew=0.;
1.269 brouard 3310: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3311: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3312: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3313: sumnew+=prevacurrent[(int)agefin][ii][ij];
3314: }
3315: if(sumnew >0.01){ /* At least some value in the prevalence */
3316: for (ii=1;ii<=nlstate+ndeath;ii++){
3317: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3318: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3319: }
3320: }else{
3321: for (ii=1;ii<=nlstate+ndeath;ii++){
3322: for (j=1;j<=nlstate+ndeath;j++)
3323: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3324: }
3325: /* if(sumnew <0.9){ */
3326: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3327: /* } */
3328: }
3329: k3=0.0; /* We put the last diagonal to 0 */
3330: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3331: doldm[ii][ii]= k3;
3332: }
3333: /* End doldm, At the end doldm is diag[(w_i)] */
3334:
1.292 brouard 3335: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3336: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3337:
1.292 brouard 3338: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3339: /* 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 3340: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3341: sumnew=0.;
1.222 brouard 3342: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3343: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3344: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3345: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3346: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3347: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3348: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3349: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3350: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3351: /* }else */
1.268 brouard 3352: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3353: } /*End ii */
3354: } /* 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 */
3355:
1.292 brouard 3356: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3357: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3358: /* end bmij */
1.266 brouard 3359: return ps; /*pointer is unchanged */
1.218 brouard 3360: }
1.217 brouard 3361: /*************** transition probabilities ***************/
3362:
1.218 brouard 3363: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3364: {
3365: /* According to parameters values stored in x and the covariate's values stored in cov,
3366: computes the probability to be observed in state j being in state i by appying the
3367: model to the ncovmodel covariates (including constant and age).
3368: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3369: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3370: ncth covariate in the global vector x is given by the formula:
3371: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3372: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3373: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3374: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3375: Outputs ps[i][j] the probability to be observed in j being in j according to
3376: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3377: */
3378: double s1, lnpijopii;
3379: /*double t34;*/
3380: int i,j, nc, ii, jj;
3381:
1.234 brouard 3382: for(i=1; i<= nlstate; i++){
3383: for(j=1; j<i;j++){
3384: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3385: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3386: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3387: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3388: }
3389: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3390: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3391: }
3392: for(j=i+1; j<=nlstate+ndeath;j++){
3393: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3394: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3395: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3396: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3397: }
3398: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3399: }
3400: }
3401:
3402: for(i=1; i<= nlstate; i++){
3403: s1=0;
3404: for(j=1; j<i; j++){
3405: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3406: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3407: }
3408: for(j=i+1; j<=nlstate+ndeath; j++){
3409: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3410: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3411: }
3412: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3413: ps[i][i]=1./(s1+1.);
3414: /* Computing other pijs */
3415: for(j=1; j<i; j++)
3416: ps[i][j]= exp(ps[i][j])*ps[i][i];
3417: for(j=i+1; j<=nlstate+ndeath; j++)
3418: ps[i][j]= exp(ps[i][j])*ps[i][i];
3419: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3420: } /* end i */
3421:
3422: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3423: for(jj=1; jj<= nlstate+ndeath; jj++){
3424: ps[ii][jj]=0;
3425: ps[ii][ii]=1;
3426: }
3427: }
1.296 brouard 3428: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3429: for(jj=1; jj<= nlstate+ndeath; jj++){
3430: s1=0.;
3431: for(ii=1; ii<= nlstate+ndeath; ii++){
3432: s1+=ps[ii][jj];
3433: }
3434: for(ii=1; ii<= nlstate; ii++){
3435: ps[ii][jj]=ps[ii][jj]/s1;
3436: }
3437: }
3438: /* Transposition */
3439: for(jj=1; jj<= nlstate+ndeath; jj++){
3440: for(ii=jj; ii<= nlstate+ndeath; ii++){
3441: s1=ps[ii][jj];
3442: ps[ii][jj]=ps[jj][ii];
3443: ps[jj][ii]=s1;
3444: }
3445: }
3446: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3447: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3448: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3449: /* } */
3450: /* printf("\n "); */
3451: /* } */
3452: /* printf("\n ");printf("%lf ",cov[2]);*/
3453: /*
3454: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3455: goto end;*/
3456: return ps;
1.217 brouard 3457: }
3458:
3459:
1.126 brouard 3460: /**************** Product of 2 matrices ******************/
3461:
1.145 brouard 3462: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3463: {
3464: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3465: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3466: /* in, b, out are matrice of pointers which should have been initialized
3467: before: only the contents of out is modified. The function returns
3468: a pointer to pointers identical to out */
1.145 brouard 3469: int i, j, k;
1.126 brouard 3470: for(i=nrl; i<= nrh; i++)
1.145 brouard 3471: for(k=ncolol; k<=ncoloh; k++){
3472: out[i][k]=0.;
3473: for(j=ncl; j<=nch; j++)
3474: out[i][k] +=in[i][j]*b[j][k];
3475: }
1.126 brouard 3476: return out;
3477: }
3478:
3479:
3480: /************* Higher Matrix Product ***************/
3481:
1.235 brouard 3482: 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 3483: {
1.332 brouard 3484: /* Computes the transition matrix starting at age 'age' and dummies values in each resultline (loop on ij to find the corresponding combination) to over
1.126 brouard 3485: 'nhstepm*hstepm*stepm' months (i.e. until
3486: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3487: nhstepm*hstepm matrices.
3488: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3489: (typically every 2 years instead of every month which is too big
3490: for the memory).
3491: Model is determined by parameters x and covariates have to be
3492: included manually here.
3493:
3494: */
3495:
1.330 brouard 3496: int i, j, d, h, k, k1;
1.131 brouard 3497: double **out, cov[NCOVMAX+1];
1.126 brouard 3498: double **newm;
1.187 brouard 3499: double agexact;
1.214 brouard 3500: double agebegin, ageend;
1.126 brouard 3501:
3502: /* Hstepm could be zero and should return the unit matrix */
3503: for (i=1;i<=nlstate+ndeath;i++)
3504: for (j=1;j<=nlstate+ndeath;j++){
3505: oldm[i][j]=(i==j ? 1.0 : 0.0);
3506: po[i][j][0]=(i==j ? 1.0 : 0.0);
3507: }
3508: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3509: for(h=1; h <=nhstepm; h++){
3510: for(d=1; d <=hstepm; d++){
3511: newm=savm;
3512: /* Covariates have to be included here again */
3513: cov[1]=1.;
1.214 brouard 3514: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3515: cov[2]=agexact;
1.319 brouard 3516: if(nagesqr==1){
1.227 brouard 3517: cov[3]= agexact*agexact;
1.319 brouard 3518: }
1.330 brouard 3519: /* Model(2) V1 + V2 + V3 + V8 + V7*V8 + V5*V6 + V8*age + V3*age + age*age */
3520: /* total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age */
3521: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
1.332 brouard 3522: if(Typevar[k1]==1){ /* A product with age */
3523: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
3524: }else{
3525: cov[2+nagesqr+k1]=precov[nres][k1];
3526: }
3527: }/* End of loop on model equation */
3528: /* Old code */
3529: /* if( Dummy[k1]==0 && Typevar[k1]==0 ){ /\* Single dummy *\/ */
3530: /* /\* V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) *\/ */
3531: /* /\* for (k=1; k<=nsd;k++) { /\\* For single dummy covariates only *\\/ *\/ */
3532: /* /\* /\\* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates *\\/ *\/ */
3533: /* /\* codtabm(ij,k) (1 & (ij-1) >> (k-1))+1 *\/ */
3534: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3535: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3536: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
3537: /* /\* nsd 1 2 3 *\/ /\* Counting single dummies covar fixed or tv *\/ */
3538: /* /\*TvarsD[nsd] 4 3 1 *\/ /\* ID of single dummy cova fixed or timevary*\/ */
3539: /* /\*TvarsDind[k] 2 3 9 *\/ /\* position K of single dummy cova *\/ */
3540: /* /\* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];or [codtabm(ij,TnsdVar[TvarsD[k]] *\/ */
3541: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3542: /* /\* printf("hpxij Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,TnsdVar[TvarsD[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,TnsdVar[TvarsD[k]])],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,TnsdVar[TvarsD[k]])); *\/ */
3543: /* printf("hpxij Dummy combi=%d k1=%d Tvar[%d]=V%d cov[2+%d+%d]=%lf resultmodel[nres][%d]=%d nres/nresult=%d/%d \n",ij,k1,k1, Tvar[k1],nagesqr,k1,cov[2+nagesqr+k1],k1,resultmodel[nres][k1],nres,nresult); */
3544: /* printf("hpxij new Dummy precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3545: /* }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /\* Single quantitative variables *\/ */
3546: /* /\* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline *\/ */
3547: /* cov[2+nagesqr+k1]=Tqresult[nres][resultmodel[nres][k1]]; */
3548: /* /\* for (k=1; k<=nsq;k++) { /\\* For single varying covariates only *\\/ *\/ */
3549: /* /\* /\\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\\/ *\/ */
3550: /* /\* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; *\/ */
3551: /* printf("hPxij Quantitative k1=%d resultmodel[nres][%d]=%d,Tqresult[%d][%d]=%f\n",k1,k1,resultmodel[nres][k1],nres,resultmodel[nres][k1],Tqresult[nres][resultmodel[nres][k1]]); */
3552: /* printf("hpxij new Quanti precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3553: /* }else if( Dummy[k1]==2 ){ /\* For dummy with age product *\/ */
3554: /* /\* Tvar[k1] Variable in the age product age*V1 is 1 *\/ */
3555: /* /\* [Tinvresult[nres][V1] is its value in the resultline nres *\/ */
3556: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvar[k1]]*cov[2]; */
3557: /* printf("DhPxij Dummy with age k1=%d Tvar[%d]=%d TinvDoQresult[nres=%d][%d]=%.f age=%.2f,cov[2+%d+%d]=%.3f\n",k1,k1,Tvar[k1],nres,TinvDoQresult[nres][Tvar[k1]],cov[2],nagesqr,k1,cov[2+nagesqr+k1]); */
3558: /* printf("hpxij new Dummy with age product precov[nres=%d][k1=%d]=%.4f * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3559:
3560: /* /\* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; *\/ */
3561: /* /\* for (k=1; k<=cptcovage;k++){ /\\* For product with age V1+V1*age +V4 +age*V3 *\\/ *\/ */
3562: /* /\* 1+2 Tage[1]=2 TVar[2]=1 Dummy[2]=2, Tage[2]=4 TVar[4]=3 Dummy[4]=3 quant*\/ */
3563: /* /\* *\/ */
1.330 brouard 3564: /* /\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\/ */
3565: /* /\* k 1 2 3 4 5 6 7 8 9 *\/ */
3566: /* /\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\/ */
1.332 brouard 3567: /* /\*cptcovage=2 1 2 *\/ */
3568: /* /\*Tage[k]= 5 8 *\/ */
3569: /* }else if( Dummy[k1]==3 ){ /\* For quant with age product *\/ */
3570: /* cov[2+nagesqr+k1]=Tresult[nres][resultmodel[nres][k1]]; */
3571: /* printf("QhPxij Quant with age k1=%d resultmodel[nres][%d]=%d,Tqresult[%d][%d]=%f\n",k1,k1,resultmodel[nres][k1],nres,resultmodel[nres][k1],Tqresult[nres][resultmodel[nres][k1]]); */
3572: /* printf("hpxij new Quanti with age product precov[nres=%d][k1=%d] * age=%.2f\n", nres, k1, precov[nres][k1], cov[2]); */
3573: /* /\* if(Dummy[Tage[k]]== 2){ /\\* dummy with age *\\/ *\/ */
3574: /* /\* /\\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\\* dummy with age *\\\/ *\\/ *\/ */
3575: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; *\\/ *\/ */
3576: /* /\* /\\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\\/ *\/ */
3577: /* /\* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[TvarsD[Tvar[Tage[k]]]])]*cov[2]; *\/ */
3578: /* /\* printf("hPxij Age combi=%d k=%d cptcovage=%d Tage[%d]=%d Tvar[Tage[%d]]=V%d nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[Tvar[Tage[k]]]])]=%d nres=%d\n",ij,k,cptcovage,k,Tage[k],k,Tvar[Tage[k]], nbcode[Tvar[Tage[k]]][codtabm(ij,TnsdVar[Tvar[Tage[k]]])],nres); *\/ */
3579: /* /\* } else if(Dummy[Tage[k]]== 3){ /\\* quantitative with age *\\/ *\/ */
3580: /* /\* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; *\/ */
3581: /* /\* } *\/ */
3582: /* /\* 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]); *\/ */
3583: /* }else if(Typevar[k1]==2 ){ /\* For product (not with age) *\/ */
3584: /* /\* for (k=1; k<=cptcovprod;k++){ /\\* For product without age *\\/ *\/ */
3585: /* /\* /\\* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 *\\/ *\/ */
3586: /* /\* /\\* k 1 2 3 4 5 6 7 8 9 *\\/ *\/ */
3587: /* /\* /\\*Tvar[k]= 5 4 3 6 5 2 7 1 1 *\\/ *\/ */
3588: /* /\* /\\*cptcovprod=1 1 2 *\\/ *\/ */
3589: /* /\* /\\*Tprod[]= 4 7 *\\/ *\/ */
3590: /* /\* /\\*Tvard[][1] 4 1 *\\/ *\/ */
3591: /* /\* /\\*Tvard[][2] 3 2 *\\/ *\/ */
1.330 brouard 3592:
1.332 brouard 3593: /* /\* printf("hPxij Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]=%d nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]=%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2],nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])],nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]); *\/ */
3594: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3595: /* cov[2+nagesqr+k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]]; */
3596: /* printf("hPxij Prod ij=%d k1=%d cov[2+%d+%d]=%.5f Tvard[%d][1]=V%d * Tvard[%d][2]=V%d ; TinvDoQresult[nres][Tvardk[k1][1]]=%.4f * TinvDoQresult[nres][Tvardk[k1][1]]=%.4f\n",ij,k1,nagesqr,k1,cov[2+nagesqr+k1],k1,Tvardk[k1][1], k1,Tvardk[k1][2], TinvDoQresult[nres][Tvardk[k1][1]], TinvDoQresult[nres][Tvardk[k1][2]]); */
3597: /* printf("hpxij new Product no age product precov[nres=%d][k1=%d]=%.4f\n", nres, k1, precov[nres][k1]); */
3598:
3599: /* /\* if(Dummy[Tvardk[k1][1]]==0){ *\/ */
3600: /* /\* if(Dummy[Tvardk[k1][2]]==0){ /\\* Product of dummies *\\/ *\/ */
3601: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; *\/ */
3602: /* /\* cov[2+nagesqr+k1]=Tinvresult[nres][Tvardk[k1][1]] * Tinvresult[nres][Tvardk[k1][2]]; *\/ */
3603: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])]; *\/ */
3604: /* /\* }else{ /\\* Product of dummy by quantitative *\\/ *\/ */
3605: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,TnsdVar[Tvard[k][1]])] * Tqresult[nres][k]; *\/ */
3606: /* /\* cov[2+nagesqr+k1]=Tresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]]; *\/ */
3607: /* /\* } *\/ */
3608: /* /\* }else{ /\\* Product of quantitative by...*\\/ *\/ */
3609: /* /\* if(Dummy[Tvard[k][2]]==0){ /\\* quant by dummy *\\/ *\/ */
3610: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][Tvard[k][1]]; *\\/ *\/ */
3611: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3612: /* /\* }else{ /\\* Product of two quant *\\/ *\/ */
3613: /* /\* /\\* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; *\\/ *\/ */
3614: /* /\* cov[2+nagesqr+k1]=Tqresult[nres][Tinvresult[nres][Tvardk[k1][1]]] * Tqresult[nres][Tinvresult[nres][Tvardk[k1][2]]] ; *\/ */
3615: /* /\* } *\/ */
3616: /* /\* }/\\*end of products quantitative *\\/ *\/ */
3617: /* }/\*end of products *\/ */
3618: /* } /\* End of loop on model equation *\/ */
1.235 brouard 3619: /* for (k=1; k<=cptcovn;k++) */
3620: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3621: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3622: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3623: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3624: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3625:
3626:
1.126 brouard 3627: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3628: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.319 brouard 3629: /* right multiplication of oldm by the current matrix */
1.126 brouard 3630: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3631: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3632: /* if((int)age == 70){ */
3633: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3634: /* for(i=1; i<=nlstate+ndeath; i++) { */
3635: /* printf("%d pmmij ",i); */
3636: /* for(j=1;j<=nlstate+ndeath;j++) { */
3637: /* printf("%f ",pmmij[i][j]); */
3638: /* } */
3639: /* printf(" oldm "); */
3640: /* for(j=1;j<=nlstate+ndeath;j++) { */
3641: /* printf("%f ",oldm[i][j]); */
3642: /* } */
3643: /* printf("\n"); */
3644: /* } */
3645: /* } */
1.126 brouard 3646: savm=oldm;
3647: oldm=newm;
3648: }
3649: for(i=1; i<=nlstate+ndeath; i++)
3650: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3651: po[i][j][h]=newm[i][j];
3652: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3653: }
1.128 brouard 3654: /*printf("h=%d ",h);*/
1.126 brouard 3655: } /* end h */
1.267 brouard 3656: /* printf("\n H=%d \n",h); */
1.126 brouard 3657: return po;
3658: }
3659:
1.217 brouard 3660: /************* Higher Back Matrix Product ***************/
1.218 brouard 3661: /* 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 3662: 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 3663: {
1.332 brouard 3664: /* For dummy covariates given in each resultline (for historical, computes the corresponding combination ij),
3665: computes the transition matrix starting at age 'age' over
1.217 brouard 3666: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3667: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3668: nhstepm*hstepm matrices.
3669: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3670: (typically every 2 years instead of every month which is too big
1.217 brouard 3671: for the memory).
1.218 brouard 3672: Model is determined by parameters x and covariates have to be
1.266 brouard 3673: included manually here. Then we use a call to bmij(x and cov)
3674: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3675: */
1.217 brouard 3676:
1.332 brouard 3677: int i, j, d, h, k, k1;
1.266 brouard 3678: double **out, cov[NCOVMAX+1], **bmij();
3679: double **newm, ***newmm;
1.217 brouard 3680: double agexact;
3681: double agebegin, ageend;
1.222 brouard 3682: double **oldm, **savm;
1.217 brouard 3683:
1.266 brouard 3684: newmm=po; /* To be saved */
3685: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3686: /* Hstepm could be zero and should return the unit matrix */
3687: for (i=1;i<=nlstate+ndeath;i++)
3688: for (j=1;j<=nlstate+ndeath;j++){
3689: oldm[i][j]=(i==j ? 1.0 : 0.0);
3690: po[i][j][0]=(i==j ? 1.0 : 0.0);
3691: }
3692: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3693: for(h=1; h <=nhstepm; h++){
3694: for(d=1; d <=hstepm; d++){
3695: newm=savm;
3696: /* Covariates have to be included here again */
3697: cov[1]=1.;
1.271 brouard 3698: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3699: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
1.318 brouard 3700: /* Debug */
3701: /* printf("hBxij age=%lf, agexact=%lf\n", age, agexact); */
1.217 brouard 3702: cov[2]=agexact;
1.332 brouard 3703: if(nagesqr==1){
1.222 brouard 3704: cov[3]= agexact*agexact;
1.332 brouard 3705: }
3706: /** New code */
3707: for(k1=1;k1<=cptcovt;k1++){ /* loop on model equation (including products) */
3708: if(Typevar[k1]==1){ /* A product with age */
3709: cov[2+nagesqr+k1]=precov[nres][k1]*cov[2];
1.325 brouard 3710: }else{
1.332 brouard 3711: cov[2+nagesqr+k1]=precov[nres][k1];
1.325 brouard 3712: }
1.332 brouard 3713: }/* End of loop on model equation */
3714: /** End of new code */
3715: /** This was old code */
3716: /* for (k=1; k<=nsd;k++){ /\* For single dummy covariates only *\//\* cptcovn error *\/ */
3717: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; *\/ */
3718: /* /\* /\\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\\/ *\/ */
3719: /* cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,TvarsD[k])];/\* Bug valgrind *\/ */
3720: /* /\* 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)); *\/ */
3721: /* } */
3722: /* for (k=1; k<=nsq;k++) { /\* For single varying covariates only *\/ */
3723: /* /\* Here comes the value of quantitative after renumbering k with single quantitative covariates *\/ */
3724: /* cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k]; */
3725: /* /\* 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]); *\/ */
3726: /* } */
3727: /* for (k=1; k<=cptcovage;k++){ /\* Should start at cptcovn+1 *\//\* For product with age *\/ */
3728: /* /\* if(Dummy[Tvar[Tage[k]]]== 2){ /\\* dummy with age error!!!*\\/ *\/ */
3729: /* if(Dummy[Tage[k]]== 2){ /\* dummy with age *\/ */
3730: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k]])]*cov[2]; */
3731: /* } else if(Dummy[Tage[k]]== 3){ /\* quantitative with age *\/ */
3732: /* cov[2+nagesqr+Tage[k]]=Tqresult[nres][k]; */
3733: /* } */
3734: /* /\* 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]); *\/ */
3735: /* } */
3736: /* for (k=1; k<=cptcovprod;k++){ /\* Useless because included in cptcovn *\/ */
3737: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])]*nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; */
3738: /* if(Dummy[Tvard[k][1]]==0){ */
3739: /* if(Dummy[Tvard[k][2]]==0){ */
3740: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][1])]; */
3741: /* }else{ */
3742: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * Tqresult[nres][k]; */
3743: /* } */
3744: /* }else{ */
3745: /* if(Dummy[Tvard[k][2]]==0){ */
3746: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])] * Tqinvresult[nres][Tvard[k][1]]; */
3747: /* }else{ */
3748: /* cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
3749: /* } */
3750: /* } */
3751: /* } */
3752: /* /\*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*\/ */
3753: /* /\*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*\/ */
3754: /** End of old code */
3755:
1.218 brouard 3756: /* Careful transposed matrix */
1.266 brouard 3757: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3758: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3759: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3760: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.325 brouard 3761: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);/* Bug valgrind */
1.217 brouard 3762: /* if((int)age == 70){ */
3763: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3764: /* for(i=1; i<=nlstate+ndeath; i++) { */
3765: /* printf("%d pmmij ",i); */
3766: /* for(j=1;j<=nlstate+ndeath;j++) { */
3767: /* printf("%f ",pmmij[i][j]); */
3768: /* } */
3769: /* printf(" oldm "); */
3770: /* for(j=1;j<=nlstate+ndeath;j++) { */
3771: /* printf("%f ",oldm[i][j]); */
3772: /* } */
3773: /* printf("\n"); */
3774: /* } */
3775: /* } */
3776: savm=oldm;
3777: oldm=newm;
3778: }
3779: for(i=1; i<=nlstate+ndeath; i++)
3780: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3781: po[i][j][h]=newm[i][j];
1.268 brouard 3782: /* if(h==nhstepm) */
3783: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3784: }
1.268 brouard 3785: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3786: } /* end h */
1.268 brouard 3787: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3788: return po;
3789: }
3790:
3791:
1.162 brouard 3792: #ifdef NLOPT
3793: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3794: double fret;
3795: double *xt;
3796: int j;
3797: myfunc_data *d2 = (myfunc_data *) pd;
3798: /* xt = (p1-1); */
3799: xt=vector(1,n);
3800: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3801:
3802: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3803: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3804: printf("Function = %.12lf ",fret);
3805: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3806: printf("\n");
3807: free_vector(xt,1,n);
3808: return fret;
3809: }
3810: #endif
1.126 brouard 3811:
3812: /*************** log-likelihood *************/
3813: double func( double *x)
3814: {
1.226 brouard 3815: int i, ii, j, k, mi, d, kk;
3816: int ioffset=0;
3817: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3818: double **out;
3819: double lli; /* Individual log likelihood */
3820: int s1, s2;
1.228 brouard 3821: 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 3822: double bbh, survp;
3823: long ipmx;
3824: double agexact;
3825: /*extern weight */
3826: /* We are differentiating ll according to initial status */
3827: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3828: /*for(i=1;i<imx;i++)
3829: printf(" %d\n",s[4][i]);
3830: */
1.162 brouard 3831:
1.226 brouard 3832: ++countcallfunc;
1.162 brouard 3833:
1.226 brouard 3834: cov[1]=1.;
1.126 brouard 3835:
1.226 brouard 3836: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3837: ioffset=0;
1.226 brouard 3838: if(mle==1){
3839: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3840: /* Computes the values of the ncovmodel covariates of the model
3841: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3842: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3843: to be observed in j being in i according to the model.
3844: */
1.243 brouard 3845: ioffset=2+nagesqr ;
1.233 brouard 3846: /* Fixed */
1.319 brouard 3847: for (k=1; k<=ncovf;k++){ /* For each fixed covariate dummu or quant or prod */
3848: /* # V1=sex, V2=raedyrs Quant Fixed, State=livarnb4..livarnb11, V3=iadl4..iald11, V4=adlw4..adlw11, V5=r4bmi..r11bmi */
3849: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
3850: /* 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 3851: /* TvarFind; TvarFind[1]=6, TvarFind[2]=7, TvarFind[3]=9 *//* Inverse V2(6) is first fixed (single or prod) */
1.319 brouard 3852: 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)*/
3853: /* V1*V2 (7) TvarFind[2]=7, TvarFind[3]=9 */
1.234 brouard 3854: }
1.226 brouard 3855: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
1.319 brouard 3856: is 5, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]=6
1.226 brouard 3857: has been calculated etc */
3858: /* For an individual i, wav[i] gives the number of effective waves */
3859: /* We compute the contribution to Likelihood of each effective transition
3860: mw[mi][i] is real wave of the mi th effectve wave */
3861: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3862: s2=s[mw[mi+1][i]][i];
3863: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3864: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3865: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3866: */
3867: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3868: 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*/
3869: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; but where is the crossproduct? */
1.242 brouard 3870: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3871: }
3872: for (ii=1;ii<=nlstate+ndeath;ii++)
3873: for (j=1;j<=nlstate+ndeath;j++){
3874: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3875: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3876: }
3877: for(d=0; d<dh[mi][i]; d++){
3878: newm=savm;
3879: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3880: cov[2]=agexact;
3881: if(nagesqr==1)
3882: cov[3]= agexact*agexact; /* Should be changed here */
3883: for (kk=1; kk<=cptcovage;kk++) {
1.318 brouard 3884: if(!FixedV[Tvar[Tage[kk]]])
3885: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
3886: else
3887: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3888: }
3889: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3890: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3891: savm=oldm;
3892: oldm=newm;
3893: } /* end mult */
3894:
3895: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3896: /* But now since version 0.9 we anticipate for bias at large stepm.
3897: * If stepm is larger than one month (smallest stepm) and if the exact delay
3898: * (in months) between two waves is not a multiple of stepm, we rounded to
3899: * the nearest (and in case of equal distance, to the lowest) interval but now
3900: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3901: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3902: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3903: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3904: * -stepm/2 to stepm/2 .
3905: * For stepm=1 the results are the same as for previous versions of Imach.
3906: * For stepm > 1 the results are less biased than in previous versions.
3907: */
1.234 brouard 3908: s1=s[mw[mi][i]][i];
3909: s2=s[mw[mi+1][i]][i];
3910: bbh=(double)bh[mi][i]/(double)stepm;
3911: /* bias bh is positive if real duration
3912: * is higher than the multiple of stepm and negative otherwise.
3913: */
3914: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3915: if( s2 > nlstate){
3916: /* i.e. if s2 is a death state and if the date of death is known
3917: then the contribution to the likelihood is the probability to
3918: die between last step unit time and current step unit time,
3919: which is also equal to probability to die before dh
3920: minus probability to die before dh-stepm .
3921: In version up to 0.92 likelihood was computed
3922: as if date of death was unknown. Death was treated as any other
3923: health state: the date of the interview describes the actual state
3924: and not the date of a change in health state. The former idea was
3925: to consider that at each interview the state was recorded
3926: (healthy, disable or death) and IMaCh was corrected; but when we
3927: introduced the exact date of death then we should have modified
3928: the contribution of an exact death to the likelihood. This new
3929: contribution is smaller and very dependent of the step unit
3930: stepm. It is no more the probability to die between last interview
3931: and month of death but the probability to survive from last
3932: interview up to one month before death multiplied by the
3933: probability to die within a month. Thanks to Chris
3934: Jackson for correcting this bug. Former versions increased
3935: mortality artificially. The bad side is that we add another loop
3936: which slows down the processing. The difference can be up to 10%
3937: lower mortality.
3938: */
3939: /* If, at the beginning of the maximization mostly, the
3940: cumulative probability or probability to be dead is
3941: constant (ie = 1) over time d, the difference is equal to
3942: 0. out[s1][3] = savm[s1][3]: probability, being at state
3943: s1 at precedent wave, to be dead a month before current
3944: wave is equal to probability, being at state s1 at
3945: precedent wave, to be dead at mont of the current
3946: wave. Then the observed probability (that this person died)
3947: is null according to current estimated parameter. In fact,
3948: it should be very low but not zero otherwise the log go to
3949: infinity.
3950: */
1.183 brouard 3951: /* #ifdef INFINITYORIGINAL */
3952: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3953: /* #else */
3954: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3955: /* lli=log(mytinydouble); */
3956: /* else */
3957: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3958: /* #endif */
1.226 brouard 3959: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3960:
1.226 brouard 3961: } else if ( s2==-1 ) { /* alive */
3962: for (j=1,survp=0. ; j<=nlstate; j++)
3963: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3964: /*survp += out[s1][j]; */
3965: lli= log(survp);
3966: }
3967: else if (s2==-4) {
3968: for (j=3,survp=0. ; j<=nlstate; j++)
3969: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3970: lli= log(survp);
3971: }
3972: else if (s2==-5) {
3973: for (j=1,survp=0. ; j<=2; j++)
3974: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3975: lli= log(survp);
3976: }
3977: else{
3978: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3979: /* 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 */
3980: }
3981: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3982: /*if(lli ==000.0)*/
3983: /*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); */
3984: ipmx +=1;
3985: sw += weight[i];
3986: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3987: /* if (lli < log(mytinydouble)){ */
3988: /* 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); */
3989: /* 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]); */
3990: /* } */
3991: } /* end of wave */
3992: } /* end of individual */
3993: } else if(mle==2){
3994: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.319 brouard 3995: ioffset=2+nagesqr ;
3996: for (k=1; k<=ncovf;k++)
3997: cov[ioffset+TvarFind[k]]=covar[Tvar[TvarFind[k]]][i];
1.226 brouard 3998: for(mi=1; mi<= wav[i]-1; mi++){
1.319 brouard 3999: for(k=1; k <= ncovv ; k++){
4000: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4001: }
1.226 brouard 4002: for (ii=1;ii<=nlstate+ndeath;ii++)
4003: for (j=1;j<=nlstate+ndeath;j++){
4004: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4005: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4006: }
4007: for(d=0; d<=dh[mi][i]; d++){
4008: newm=savm;
4009: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4010: cov[2]=agexact;
4011: if(nagesqr==1)
4012: cov[3]= agexact*agexact;
4013: for (kk=1; kk<=cptcovage;kk++) {
4014: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4015: }
4016: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4017: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4018: savm=oldm;
4019: oldm=newm;
4020: } /* end mult */
4021:
4022: s1=s[mw[mi][i]][i];
4023: s2=s[mw[mi+1][i]][i];
4024: bbh=(double)bh[mi][i]/(double)stepm;
4025: 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 */
4026: ipmx +=1;
4027: sw += weight[i];
4028: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4029: } /* end of wave */
4030: } /* end of individual */
4031: } else if(mle==3){ /* exponential inter-extrapolation */
4032: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4033: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4034: for(mi=1; mi<= wav[i]-1; mi++){
4035: for (ii=1;ii<=nlstate+ndeath;ii++)
4036: for (j=1;j<=nlstate+ndeath;j++){
4037: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4038: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4039: }
4040: for(d=0; d<dh[mi][i]; d++){
4041: newm=savm;
4042: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4043: cov[2]=agexact;
4044: if(nagesqr==1)
4045: cov[3]= agexact*agexact;
4046: for (kk=1; kk<=cptcovage;kk++) {
4047: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4048: }
4049: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4050: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4051: savm=oldm;
4052: oldm=newm;
4053: } /* end mult */
4054:
4055: s1=s[mw[mi][i]][i];
4056: s2=s[mw[mi+1][i]][i];
4057: bbh=(double)bh[mi][i]/(double)stepm;
4058: 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 */
4059: ipmx +=1;
4060: sw += weight[i];
4061: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4062: } /* end of wave */
4063: } /* end of individual */
4064: }else if (mle==4){ /* ml=4 no inter-extrapolation */
4065: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4066: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4067: for(mi=1; mi<= wav[i]-1; mi++){
4068: for (ii=1;ii<=nlstate+ndeath;ii++)
4069: for (j=1;j<=nlstate+ndeath;j++){
4070: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4071: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4072: }
4073: for(d=0; d<dh[mi][i]; d++){
4074: newm=savm;
4075: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4076: cov[2]=agexact;
4077: if(nagesqr==1)
4078: cov[3]= agexact*agexact;
4079: for (kk=1; kk<=cptcovage;kk++) {
4080: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4081: }
1.126 brouard 4082:
1.226 brouard 4083: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4084: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4085: savm=oldm;
4086: oldm=newm;
4087: } /* end mult */
4088:
4089: s1=s[mw[mi][i]][i];
4090: s2=s[mw[mi+1][i]][i];
4091: if( s2 > nlstate){
4092: lli=log(out[s1][s2] - savm[s1][s2]);
4093: } else if ( s2==-1 ) { /* alive */
4094: for (j=1,survp=0. ; j<=nlstate; j++)
4095: survp += out[s1][j];
4096: lli= log(survp);
4097: }else{
4098: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4099: }
4100: ipmx +=1;
4101: sw += weight[i];
4102: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 4103: /* 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 4104: } /* end of wave */
4105: } /* end of individual */
4106: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
4107: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
4108: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
4109: for(mi=1; mi<= wav[i]-1; mi++){
4110: for (ii=1;ii<=nlstate+ndeath;ii++)
4111: for (j=1;j<=nlstate+ndeath;j++){
4112: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4113: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4114: }
4115: for(d=0; d<dh[mi][i]; d++){
4116: newm=savm;
4117: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
4118: cov[2]=agexact;
4119: if(nagesqr==1)
4120: cov[3]= agexact*agexact;
4121: for (kk=1; kk<=cptcovage;kk++) {
4122: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4123: }
1.126 brouard 4124:
1.226 brouard 4125: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4126: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4127: savm=oldm;
4128: oldm=newm;
4129: } /* end mult */
4130:
4131: s1=s[mw[mi][i]][i];
4132: s2=s[mw[mi+1][i]][i];
4133: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
4134: ipmx +=1;
4135: sw += weight[i];
4136: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
4137: /*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]);*/
4138: } /* end of wave */
4139: } /* end of individual */
4140: } /* End of if */
4141: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4142: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4143: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4144: return -l;
1.126 brouard 4145: }
4146:
4147: /*************** log-likelihood *************/
4148: double funcone( double *x)
4149: {
1.228 brouard 4150: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 4151: int i, ii, j, k, mi, d, kk;
1.228 brouard 4152: int ioffset=0;
1.131 brouard 4153: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 4154: double **out;
4155: double lli; /* Individual log likelihood */
4156: double llt;
4157: int s1, s2;
1.228 brouard 4158: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
4159:
1.126 brouard 4160: double bbh, survp;
1.187 brouard 4161: double agexact;
1.214 brouard 4162: double agebegin, ageend;
1.126 brouard 4163: /*extern weight */
4164: /* We are differentiating ll according to initial status */
4165: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
4166: /*for(i=1;i<imx;i++)
4167: printf(" %d\n",s[4][i]);
4168: */
4169: cov[1]=1.;
4170:
4171: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 4172: ioffset=0;
4173: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 4174: /* ioffset=2+nagesqr+cptcovage; */
4175: ioffset=2+nagesqr;
1.232 brouard 4176: /* Fixed */
1.224 brouard 4177: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 4178: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.311 brouard 4179: 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 4180: 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)*/
4181: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
4182: /* cov[2+6]=covar[Tvar[6]][i]; */
4183: /* cov[2+6]=covar[2][i]; V2 */
4184: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
4185: /* cov[2+7]=covar[Tvar[7]][i]; */
4186: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
4187: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
4188: /* cov[2+9]=covar[Tvar[9]][i]; */
4189: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 4190: }
1.232 brouard 4191: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
4192: /* 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?)*\/ */
4193: /* } */
1.231 brouard 4194: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
4195: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
4196: /* } */
1.225 brouard 4197:
1.233 brouard 4198:
4199: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 4200: /* Wave varying (but not age varying) */
4201: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 4202: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
4203: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
4204: }
1.232 brouard 4205: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 4206: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4207: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
4208: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
4209: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
4210: /* 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 4211: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 4212: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
4213: /* /\* 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]); *\/ */
4214: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 4215: /* } */
1.126 brouard 4216: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 4217: for (j=1;j<=nlstate+ndeath;j++){
4218: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
4219: savm[ii][j]=(ii==j ? 1.0 : 0.0);
4220: }
1.214 brouard 4221:
4222: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
4223: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
4224: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 4225: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 4226: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
4227: and mw[mi+1][i]. dh depends on stepm.*/
4228: newm=savm;
1.247 brouard 4229: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 4230: cov[2]=agexact;
4231: if(nagesqr==1)
4232: cov[3]= agexact*agexact;
4233: for (kk=1; kk<=cptcovage;kk++) {
4234: if(!FixedV[Tvar[Tage[kk]]])
4235: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
4236: else
4237: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
4238: }
4239: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
4240: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
4241: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
4242: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
4243: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
4244: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
4245: savm=oldm;
4246: oldm=newm;
1.126 brouard 4247: } /* end mult */
4248:
4249: s1=s[mw[mi][i]][i];
4250: s2=s[mw[mi+1][i]][i];
1.217 brouard 4251: /* if(s2==-1){ */
1.268 brouard 4252: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 4253: /* /\* exit(1); *\/ */
4254: /* } */
1.126 brouard 4255: bbh=(double)bh[mi][i]/(double)stepm;
4256: /* bias is positive if real duration
4257: * is higher than the multiple of stepm and negative otherwise.
4258: */
4259: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 4260: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 4261: } else if ( s2==-1 ) { /* alive */
1.242 brouard 4262: for (j=1,survp=0. ; j<=nlstate; j++)
4263: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
4264: lli= log(survp);
1.126 brouard 4265: }else if (mle==1){
1.242 brouard 4266: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 4267: } else if(mle==2){
1.242 brouard 4268: 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 4269: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 4270: 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 4271: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 4272: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 4273: } else{ /* mle=0 back to 1 */
1.242 brouard 4274: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
4275: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 4276: } /* End of if */
4277: ipmx +=1;
4278: sw += weight[i];
4279: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 4280: /*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 4281: if(globpr){
1.246 brouard 4282: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 4283: %11.6f %11.6f %11.6f ", \
1.242 brouard 4284: 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 4285: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 4286: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
4287: llt +=ll[k]*gipmx/gsw;
4288: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
4289: }
4290: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 4291: }
1.232 brouard 4292: } /* end of wave */
4293: } /* end of individual */
4294: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
4295: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
4296: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
4297: if(globpr==0){ /* First time we count the contributions and weights */
4298: gipmx=ipmx;
4299: gsw=sw;
4300: }
4301: return -l;
1.126 brouard 4302: }
4303:
4304:
4305: /*************** function likelione ***********/
1.292 brouard 4306: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 4307: {
4308: /* This routine should help understanding what is done with
4309: the selection of individuals/waves and
4310: to check the exact contribution to the likelihood.
4311: Plotting could be done.
4312: */
4313: int k;
4314:
4315: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4316: strcpy(fileresilk,"ILK_");
1.202 brouard 4317: strcat(fileresilk,fileresu);
1.126 brouard 4318: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4319: printf("Problem with resultfile: %s\n", fileresilk);
4320: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4321: }
1.214 brouard 4322: 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");
4323: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4324: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4325: for(k=1; k<=nlstate; k++)
4326: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4327: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4328: }
4329:
1.292 brouard 4330: *fretone=(*func)(p);
1.126 brouard 4331: if(*globpri !=0){
4332: fclose(ficresilk);
1.205 brouard 4333: if (mle ==0)
4334: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4335: else if(mle >=1)
4336: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4337: 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 4338: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4339:
4340: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4341: 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 4342: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4343: }
1.207 brouard 4344: 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 4345: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4346: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4347: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4348: fflush(fichtm);
1.205 brouard 4349: }
1.126 brouard 4350: return;
4351: }
4352:
4353:
4354: /*********** Maximum Likelihood Estimation ***************/
4355:
4356: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4357: {
1.319 brouard 4358: int i,j,k, jk, jkk=0, iter=0;
1.126 brouard 4359: double **xi;
4360: double fret;
4361: double fretone; /* Only one call to likelihood */
4362: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4363:
4364: #ifdef NLOPT
4365: int creturn;
4366: nlopt_opt opt;
4367: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4368: double *lb;
4369: double minf; /* the minimum objective value, upon return */
4370: double * p1; /* Shifted parameters from 0 instead of 1 */
4371: myfunc_data dinst, *d = &dinst;
4372: #endif
4373:
4374:
1.126 brouard 4375: xi=matrix(1,npar,1,npar);
4376: for (i=1;i<=npar;i++)
4377: for (j=1;j<=npar;j++)
4378: xi[i][j]=(i==j ? 1.0 : 0.0);
4379: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4380: strcpy(filerespow,"POW_");
1.126 brouard 4381: strcat(filerespow,fileres);
4382: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4383: printf("Problem with resultfile: %s\n", filerespow);
4384: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4385: }
4386: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4387: for (i=1;i<=nlstate;i++)
4388: for(j=1;j<=nlstate+ndeath;j++)
4389: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4390: fprintf(ficrespow,"\n");
1.162 brouard 4391: #ifdef POWELL
1.319 brouard 4392: #ifdef LINMINORIGINAL
4393: #else /* LINMINORIGINAL */
4394:
4395: flatdir=ivector(1,npar);
4396: for (j=1;j<=npar;j++) flatdir[j]=0;
4397: #endif /*LINMINORIGINAL */
4398:
4399: #ifdef FLATSUP
4400: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4401: /* reorganizing p by suppressing flat directions */
4402: for(i=1, jk=1; i <=nlstate; i++){
4403: for(k=1; k <=(nlstate+ndeath); k++){
4404: if (k != i) {
4405: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4406: if(flatdir[jk]==1){
4407: printf(" To be skipped %d%d flatdir[%d]=%d ",i,k,jk, flatdir[jk]);
4408: }
4409: for(j=1; j <=ncovmodel; j++){
4410: printf("%12.7f ",p[jk]);
4411: jk++;
4412: }
4413: printf("\n");
4414: }
4415: }
4416: }
4417: /* skipping */
4418: /* for(i=1, jk=1, jkk=1;(flatdir[jk]==0)&& (i <=nlstate); i++){ */
4419: for(i=1, jk=1, jkk=1;i <=nlstate; i++){
4420: for(k=1; k <=(nlstate+ndeath); k++){
4421: if (k != i) {
4422: printf("%d%d flatdir[%d]=%d",i,k,jk, flatdir[jk]);
4423: if(flatdir[jk]==1){
4424: printf(" To be skipped %d%d flatdir[%d]=%d jk=%d p[%d] ",i,k,jk, flatdir[jk],jk, jk);
4425: for(j=1; j <=ncovmodel; jk++,j++){
4426: printf(" p[%d]=%12.7f",jk, p[jk]);
4427: /*q[jjk]=p[jk];*/
4428: }
4429: }else{
4430: printf(" To be kept %d%d flatdir[%d]=%d jk=%d q[%d]=p[%d] ",i,k,jk, flatdir[jk],jk, jkk, jk);
4431: for(j=1; j <=ncovmodel; jk++,jkk++,j++){
4432: printf(" p[%d]=%12.7f=q[%d]",jk, p[jk],jkk);
4433: /*q[jjk]=p[jk];*/
4434: }
4435: }
4436: printf("\n");
4437: }
4438: fflush(stdout);
4439: }
4440: }
4441: powell(p,xi,npar,ftol,&iter,&fret,flatdir,func);
4442: #else /* FLATSUP */
1.126 brouard 4443: powell(p,xi,npar,ftol,&iter,&fret,func);
1.319 brouard 4444: #endif /* FLATSUP */
4445:
4446: #ifdef LINMINORIGINAL
4447: #else
4448: free_ivector(flatdir,1,npar);
4449: #endif /* LINMINORIGINAL*/
4450: #endif /* POWELL */
1.126 brouard 4451:
1.162 brouard 4452: #ifdef NLOPT
4453: #ifdef NEWUOA
4454: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4455: #else
4456: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4457: #endif
4458: lb=vector(0,npar-1);
4459: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4460: nlopt_set_lower_bounds(opt, lb);
4461: nlopt_set_initial_step1(opt, 0.1);
4462:
4463: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4464: d->function = func;
4465: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4466: nlopt_set_min_objective(opt, myfunc, d);
4467: nlopt_set_xtol_rel(opt, ftol);
4468: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4469: printf("nlopt failed! %d\n",creturn);
4470: }
4471: else {
4472: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4473: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4474: iter=1; /* not equal */
4475: }
4476: nlopt_destroy(opt);
4477: #endif
1.319 brouard 4478: #ifdef FLATSUP
4479: /* npared = npar -flatd/ncovmodel; */
4480: /* xired= matrix(1,npared,1,npared); */
4481: /* paramred= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel); */
4482: /* powell(pred,xired,npared,ftol,&iter,&fret,flatdir,func); */
4483: /* free_matrix(xire,1,npared,1,npared); */
4484: #else /* FLATSUP */
4485: #endif /* FLATSUP */
1.126 brouard 4486: free_matrix(xi,1,npar,1,npar);
4487: fclose(ficrespow);
1.203 brouard 4488: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4489: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4490: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4491:
4492: }
4493:
4494: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4495: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4496: {
4497: double **a,**y,*x,pd;
1.203 brouard 4498: /* double **hess; */
1.164 brouard 4499: int i, j;
1.126 brouard 4500: int *indx;
4501:
4502: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4503: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4504: void lubksb(double **a, int npar, int *indx, double b[]) ;
4505: void ludcmp(double **a, int npar, int *indx, double *d) ;
4506: double gompertz(double p[]);
1.203 brouard 4507: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4508:
4509: printf("\nCalculation of the hessian matrix. Wait...\n");
4510: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4511: for (i=1;i<=npar;i++){
1.203 brouard 4512: printf("%d-",i);fflush(stdout);
4513: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4514:
4515: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4516:
4517: /* printf(" %f ",p[i]);
4518: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4519: }
4520:
4521: for (i=1;i<=npar;i++) {
4522: for (j=1;j<=npar;j++) {
4523: if (j>i) {
1.203 brouard 4524: printf(".%d-%d",i,j);fflush(stdout);
4525: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4526: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4527:
4528: hess[j][i]=hess[i][j];
4529: /*printf(" %lf ",hess[i][j]);*/
4530: }
4531: }
4532: }
4533: printf("\n");
4534: fprintf(ficlog,"\n");
4535:
4536: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4537: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4538:
4539: a=matrix(1,npar,1,npar);
4540: y=matrix(1,npar,1,npar);
4541: x=vector(1,npar);
4542: indx=ivector(1,npar);
4543: for (i=1;i<=npar;i++)
4544: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4545: ludcmp(a,npar,indx,&pd);
4546:
4547: for (j=1;j<=npar;j++) {
4548: for (i=1;i<=npar;i++) x[i]=0;
4549: x[j]=1;
4550: lubksb(a,npar,indx,x);
4551: for (i=1;i<=npar;i++){
4552: matcov[i][j]=x[i];
4553: }
4554: }
4555:
4556: printf("\n#Hessian matrix#\n");
4557: fprintf(ficlog,"\n#Hessian matrix#\n");
4558: for (i=1;i<=npar;i++) {
4559: for (j=1;j<=npar;j++) {
1.203 brouard 4560: printf("%.6e ",hess[i][j]);
4561: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4562: }
4563: printf("\n");
4564: fprintf(ficlog,"\n");
4565: }
4566:
1.203 brouard 4567: /* printf("\n#Covariance matrix#\n"); */
4568: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4569: /* for (i=1;i<=npar;i++) { */
4570: /* for (j=1;j<=npar;j++) { */
4571: /* printf("%.6e ",matcov[i][j]); */
4572: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4573: /* } */
4574: /* printf("\n"); */
4575: /* fprintf(ficlog,"\n"); */
4576: /* } */
4577:
1.126 brouard 4578: /* Recompute Inverse */
1.203 brouard 4579: /* for (i=1;i<=npar;i++) */
4580: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4581: /* ludcmp(a,npar,indx,&pd); */
4582:
4583: /* printf("\n#Hessian matrix recomputed#\n"); */
4584:
4585: /* for (j=1;j<=npar;j++) { */
4586: /* for (i=1;i<=npar;i++) x[i]=0; */
4587: /* x[j]=1; */
4588: /* lubksb(a,npar,indx,x); */
4589: /* for (i=1;i<=npar;i++){ */
4590: /* y[i][j]=x[i]; */
4591: /* printf("%.3e ",y[i][j]); */
4592: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4593: /* } */
4594: /* printf("\n"); */
4595: /* fprintf(ficlog,"\n"); */
4596: /* } */
4597:
4598: /* Verifying the inverse matrix */
4599: #ifdef DEBUGHESS
4600: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4601:
1.203 brouard 4602: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4603: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4604:
4605: for (j=1;j<=npar;j++) {
4606: for (i=1;i<=npar;i++){
1.203 brouard 4607: printf("%.2f ",y[i][j]);
4608: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4609: }
4610: printf("\n");
4611: fprintf(ficlog,"\n");
4612: }
1.203 brouard 4613: #endif
1.126 brouard 4614:
4615: free_matrix(a,1,npar,1,npar);
4616: free_matrix(y,1,npar,1,npar);
4617: free_vector(x,1,npar);
4618: free_ivector(indx,1,npar);
1.203 brouard 4619: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4620:
4621:
4622: }
4623:
4624: /*************** hessian matrix ****************/
4625: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4626: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4627: int i;
4628: int l=1, lmax=20;
1.203 brouard 4629: double k1,k2, res, fx;
1.132 brouard 4630: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4631: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4632: int k=0,kmax=10;
4633: double l1;
4634:
4635: fx=func(x);
4636: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4637: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4638: l1=pow(10,l);
4639: delts=delt;
4640: for(k=1 ; k <kmax; k=k+1){
4641: delt = delta*(l1*k);
4642: p2[theta]=x[theta] +delt;
1.145 brouard 4643: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4644: p2[theta]=x[theta]-delt;
4645: k2=func(p2)-fx;
4646: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4647: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4648:
1.203 brouard 4649: #ifdef DEBUGHESSII
1.126 brouard 4650: 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);
4651: 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);
4652: #endif
4653: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4654: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4655: k=kmax;
4656: }
4657: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4658: k=kmax; l=lmax*10;
1.126 brouard 4659: }
4660: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4661: delts=delt;
4662: }
1.203 brouard 4663: } /* End loop k */
1.126 brouard 4664: }
4665: delti[theta]=delts;
4666: return res;
4667:
4668: }
4669:
1.203 brouard 4670: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4671: {
4672: int i;
1.164 brouard 4673: int l=1, lmax=20;
1.126 brouard 4674: double k1,k2,k3,k4,res,fx;
1.132 brouard 4675: double p2[MAXPARM+1];
1.203 brouard 4676: int k, kmax=1;
4677: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4678:
4679: int firstime=0;
1.203 brouard 4680:
1.126 brouard 4681: fx=func(x);
1.203 brouard 4682: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4683: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4684: p2[thetai]=x[thetai]+delti[thetai]*k;
4685: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4686: k1=func(p2)-fx;
4687:
1.203 brouard 4688: p2[thetai]=x[thetai]+delti[thetai]*k;
4689: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4690: k2=func(p2)-fx;
4691:
1.203 brouard 4692: p2[thetai]=x[thetai]-delti[thetai]*k;
4693: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4694: k3=func(p2)-fx;
4695:
1.203 brouard 4696: p2[thetai]=x[thetai]-delti[thetai]*k;
4697: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4698: k4=func(p2)-fx;
1.203 brouard 4699: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4700: if(k1*k2*k3*k4 <0.){
1.208 brouard 4701: firstime=1;
1.203 brouard 4702: kmax=kmax+10;
1.208 brouard 4703: }
4704: if(kmax >=10 || firstime ==1){
1.246 brouard 4705: 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);
4706: 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 4707: 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);
4708: 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);
4709: }
4710: #ifdef DEBUGHESSIJ
4711: v1=hess[thetai][thetai];
4712: v2=hess[thetaj][thetaj];
4713: cv12=res;
4714: /* Computing eigen value of Hessian matrix */
4715: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4716: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4717: if ((lc2 <0) || (lc1 <0) ){
4718: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4719: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4720: 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);
4721: 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);
4722: }
1.126 brouard 4723: #endif
4724: }
4725: return res;
4726: }
4727:
1.203 brouard 4728: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4729: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4730: /* { */
4731: /* int i; */
4732: /* int l=1, lmax=20; */
4733: /* double k1,k2,k3,k4,res,fx; */
4734: /* double p2[MAXPARM+1]; */
4735: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4736: /* int k=0,kmax=10; */
4737: /* double l1; */
4738:
4739: /* fx=func(x); */
4740: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4741: /* l1=pow(10,l); */
4742: /* delts=delt; */
4743: /* for(k=1 ; k <kmax; k=k+1){ */
4744: /* delt = delti*(l1*k); */
4745: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4746: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4747: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4748: /* k1=func(p2)-fx; */
4749:
4750: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4751: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4752: /* k2=func(p2)-fx; */
4753:
4754: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4755: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4756: /* k3=func(p2)-fx; */
4757:
4758: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4759: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4760: /* k4=func(p2)-fx; */
4761: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4762: /* #ifdef DEBUGHESSIJ */
4763: /* 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); */
4764: /* 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); */
4765: /* #endif */
4766: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4767: /* k=kmax; */
4768: /* } */
4769: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4770: /* k=kmax; l=lmax*10; */
4771: /* } */
4772: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4773: /* delts=delt; */
4774: /* } */
4775: /* } /\* End loop k *\/ */
4776: /* } */
4777: /* delti[theta]=delts; */
4778: /* return res; */
4779: /* } */
4780:
4781:
1.126 brouard 4782: /************** Inverse of matrix **************/
4783: void ludcmp(double **a, int n, int *indx, double *d)
4784: {
4785: int i,imax,j,k;
4786: double big,dum,sum,temp;
4787: double *vv;
4788:
4789: vv=vector(1,n);
4790: *d=1.0;
4791: for (i=1;i<=n;i++) {
4792: big=0.0;
4793: for (j=1;j<=n;j++)
4794: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4795: if (big == 0.0){
4796: printf(" Singular Hessian matrix at row %d:\n",i);
4797: for (j=1;j<=n;j++) {
4798: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4799: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4800: }
4801: fflush(ficlog);
4802: fclose(ficlog);
4803: nrerror("Singular matrix in routine ludcmp");
4804: }
1.126 brouard 4805: vv[i]=1.0/big;
4806: }
4807: for (j=1;j<=n;j++) {
4808: for (i=1;i<j;i++) {
4809: sum=a[i][j];
4810: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4811: a[i][j]=sum;
4812: }
4813: big=0.0;
4814: for (i=j;i<=n;i++) {
4815: sum=a[i][j];
4816: for (k=1;k<j;k++)
4817: sum -= a[i][k]*a[k][j];
4818: a[i][j]=sum;
4819: if ( (dum=vv[i]*fabs(sum)) >= big) {
4820: big=dum;
4821: imax=i;
4822: }
4823: }
4824: if (j != imax) {
4825: for (k=1;k<=n;k++) {
4826: dum=a[imax][k];
4827: a[imax][k]=a[j][k];
4828: a[j][k]=dum;
4829: }
4830: *d = -(*d);
4831: vv[imax]=vv[j];
4832: }
4833: indx[j]=imax;
4834: if (a[j][j] == 0.0) a[j][j]=TINY;
4835: if (j != n) {
4836: dum=1.0/(a[j][j]);
4837: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4838: }
4839: }
4840: free_vector(vv,1,n); /* Doesn't work */
4841: ;
4842: }
4843:
4844: void lubksb(double **a, int n, int *indx, double b[])
4845: {
4846: int i,ii=0,ip,j;
4847: double sum;
4848:
4849: for (i=1;i<=n;i++) {
4850: ip=indx[i];
4851: sum=b[ip];
4852: b[ip]=b[i];
4853: if (ii)
4854: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4855: else if (sum) ii=i;
4856: b[i]=sum;
4857: }
4858: for (i=n;i>=1;i--) {
4859: sum=b[i];
4860: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4861: b[i]=sum/a[i][i];
4862: }
4863: }
4864:
4865: void pstamp(FILE *fichier)
4866: {
1.196 brouard 4867: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4868: }
4869:
1.297 brouard 4870: void date2dmy(double date,double *day, double *month, double *year){
4871: double yp=0., yp1=0., yp2=0.;
4872:
4873: yp1=modf(date,&yp);/* extracts integral of date in yp and
4874: fractional in yp1 */
4875: *year=yp;
4876: yp2=modf((yp1*12),&yp);
4877: *month=yp;
4878: yp1=modf((yp2*30.5),&yp);
4879: *day=yp;
4880: if(*day==0) *day=1;
4881: if(*month==0) *month=1;
4882: }
4883:
1.253 brouard 4884:
4885:
1.126 brouard 4886: /************ Frequencies ********************/
1.251 brouard 4887: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4888: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4889: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4890: { /* Some frequencies as well as proposing some starting values */
1.332 brouard 4891: /* Frequencies of any combination of dummy covariate used in the model equation */
1.265 brouard 4892: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4893: int iind=0, iage=0;
4894: int mi; /* Effective wave */
4895: int first;
4896: double ***freq; /* Frequencies */
1.268 brouard 4897: 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 */
4898: 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 4899: double *meanq, *stdq, *idq;
1.226 brouard 4900: double **meanqt;
4901: double *pp, **prop, *posprop, *pospropt;
4902: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4903: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4904: double agebegin, ageend;
4905:
4906: pp=vector(1,nlstate);
1.251 brouard 4907: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4908: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4909: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4910: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4911: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4912: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4913: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4914: meanqt=matrix(1,lastpass,1,nqtveff);
4915: strcpy(fileresp,"P_");
4916: strcat(fileresp,fileresu);
4917: /*strcat(fileresphtm,fileresu);*/
4918: if((ficresp=fopen(fileresp,"w"))==NULL) {
4919: printf("Problem with prevalence resultfile: %s\n", fileresp);
4920: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4921: exit(0);
4922: }
1.240 brouard 4923:
1.226 brouard 4924: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4925: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4926: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4927: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4928: fflush(ficlog);
4929: exit(70);
4930: }
4931: else{
4932: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4933: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4934: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4935: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4936: }
1.319 brouard 4937: 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 4938:
1.226 brouard 4939: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4940: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4941: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4942: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4943: fflush(ficlog);
4944: exit(70);
1.240 brouard 4945: } else{
1.226 brouard 4946: 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 4947: ,<hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4948: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4949: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4950: }
1.319 brouard 4951: 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 4952:
1.253 brouard 4953: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4954: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4955: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4956: j1=0;
1.126 brouard 4957:
1.227 brouard 4958: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4959: j=cptcoveff; /* Only dummy covariates of the model */
1.330 brouard 4960: /* j=cptcovn; /\* Only dummy covariates of the model *\/ */
1.226 brouard 4961: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4962:
4963:
1.226 brouard 4964: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4965: reference=low_education V1=0,V2=0
4966: med_educ V1=1 V2=0,
4967: high_educ V1=0 V2=1
1.330 brouard 4968: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcovn
1.226 brouard 4969: */
1.249 brouard 4970: dateintsum=0;
4971: k2cpt=0;
4972:
1.253 brouard 4973: if(cptcoveff == 0 )
1.265 brouard 4974: nl=1; /* Constant and age model only */
1.253 brouard 4975: else
4976: nl=2;
1.265 brouard 4977:
4978: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4979: /* Loop on nj=1 or 2 if dummy covariates j!=0
1.330 brouard 4980: * Loop on j1(1 to 2**cptcovn) covariate combination
1.265 brouard 4981: * freq[s1][s2][iage] =0.
4982: * Loop on iind
4983: * ++freq[s1][s2][iage] weighted
4984: * end iind
4985: * if covariate and j!0
4986: * headers Variable on one line
4987: * endif cov j!=0
4988: * header of frequency table by age
4989: * Loop on age
4990: * pp[s1]+=freq[s1][s2][iage] weighted
4991: * pos+=freq[s1][s2][iage] weighted
4992: * Loop on s1 initial state
4993: * fprintf(ficresp
4994: * end s1
4995: * end age
4996: * if j!=0 computes starting values
4997: * end compute starting values
4998: * end j1
4999: * end nl
5000: */
1.253 brouard 5001: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
5002: if(nj==1)
5003: j=0; /* First pass for the constant */
1.265 brouard 5004: else{
1.330 brouard 5005: j=cptcovs; /* Other passes for the covariate values */
1.265 brouard 5006: }
1.251 brouard 5007: first=1;
1.332 brouard 5008: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on all dummy covariates combination of the model, ie excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
1.251 brouard 5009: posproptt=0.;
1.330 brouard 5010: /*printf("cptcovn=%d Tvaraff=%d", cptcovn,Tvaraff[1]);
1.251 brouard 5011: scanf("%d", i);*/
5012: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 5013: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 5014: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 5015: freq[i][s2][m]=0;
1.251 brouard 5016:
5017: for (i=1; i<=nlstate; i++) {
1.240 brouard 5018: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 5019: prop[i][m]=0;
5020: posprop[i]=0;
5021: pospropt[i]=0;
5022: }
1.283 brouard 5023: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 5024: idq[z1]=0.;
5025: meanq[z1]=0.;
5026: stdq[z1]=0.;
1.283 brouard 5027: }
5028: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 5029: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 5030: /* meanqt[m][z1]=0.; */
5031: /* } */
5032: /* } */
1.251 brouard 5033: /* dateintsum=0; */
5034: /* k2cpt=0; */
5035:
1.265 brouard 5036: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 5037: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
5038: bool=1;
5039: if(j !=0){
5040: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
1.330 brouard 5041: if (cptcovn >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
5042: for (z1=1; z1<=cptcovn; z1++) { /* loops on covariates in the model */
1.251 brouard 5043: /* if(Tvaraff[z1] ==-20){ */
5044: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
5045: /* }else if(Tvaraff[z1] ==-10){ */
5046: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
1.330 brouard 5047: /* }else */ /* TODO TODO codtabm(j1,z1) or codtabm(j1,Tvaraff[z1]]z1)*/
1.332 brouard 5048: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]){ /* for combination j1 of covariates */
1.265 brouard 5049: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 5050: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
1.332 brouard 5051: /* 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", */
5052: /* bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),*/
5053: /* j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
1.251 brouard 5054: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
5055: } /* Onlyf fixed */
5056: } /* end z1 */
5057: } /* cptcovn > 0 */
5058: } /* end any */
5059: }/* end j==0 */
1.265 brouard 5060: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 5061: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 5062: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 5063: m=mw[mi][iind];
5064: if(j!=0){
5065: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
1.330 brouard 5066: for (z1=1; z1<=cptcovn; z1++) {
1.251 brouard 5067: if( Fixed[Tmodelind[z1]]==1){
5068: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.332 brouard 5069: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality. If covariate's
1.251 brouard 5070: value is -1, we don't select. It differs from the
5071: constant and age model which counts them. */
5072: bool=0; /* not selected */
5073: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
1.332 brouard 5074: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.251 brouard 5075: bool=0;
5076: }
5077: }
5078: }
5079: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
5080: } /* end j==0 */
5081: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 5082: if(bool==1){ /*Selected */
1.251 brouard 5083: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
5084: and mw[mi+1][iind]. dh depends on stepm. */
5085: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
5086: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
5087: if(m >=firstpass && m <=lastpass){
5088: k2=anint[m][iind]+(mint[m][iind]/12.);
5089: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
5090: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
5091: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
5092: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
5093: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
5094: if (m<lastpass) {
5095: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
5096: /* 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]); */
5097: if(s[m][iind]==-1)
5098: 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.));
5099: 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 5100: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
5101: if(!isnan(covar[ncovcol+z1][iind])){
1.332 brouard 5102: idq[z1]=idq[z1]+weight[iind];
5103: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
5104: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
5105: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
1.311 brouard 5106: }
1.284 brouard 5107: }
1.251 brouard 5108: /* if((int)agev[m][iind] == 55) */
5109: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
5110: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
5111: 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 5112: }
1.251 brouard 5113: } /* end if between passes */
5114: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
5115: dateintsum=dateintsum+k2; /* on all covariates ?*/
5116: k2cpt++;
5117: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 5118: }
1.251 brouard 5119: }else{
5120: bool=1;
5121: }/* end bool 2 */
5122: } /* end m */
1.284 brouard 5123: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
5124: /* idq[z1]=idq[z1]+weight[iind]; */
5125: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
5126: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
5127: /* } */
1.251 brouard 5128: } /* end bool */
5129: } /* end iind = 1 to imx */
1.319 brouard 5130: /* prop[s][age] is fed for any initial and valid live state as well as
1.251 brouard 5131: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
5132:
5133:
5134: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.330 brouard 5135: if(cptcovn==0 && nj==1) /* no covariate and first pass */
1.265 brouard 5136: pstamp(ficresp);
1.330 brouard 5137: if (cptcovn>0 && j!=0){
1.265 brouard 5138: pstamp(ficresp);
1.251 brouard 5139: printf( "\n#********** Variable ");
5140: fprintf(ficresp, "\n#********** Variable ");
5141: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
5142: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
5143: fprintf(ficlog, "\n#********** Variable ");
1.330 brouard 5144: for (z1=1; z1<=cptcovs; z1++){
1.251 brouard 5145: if(!FixedV[Tvaraff[z1]]){
1.330 brouard 5146: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5147: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5148: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5149: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5150: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.250 brouard 5151: }else{
1.330 brouard 5152: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5153: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5154: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5155: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
5156: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5157: }
5158: }
5159: printf( "**********\n#");
5160: fprintf(ficresp, "**********\n#");
5161: fprintf(ficresphtm, "**********</h3>\n");
5162: fprintf(ficresphtmfr, "**********</h3>\n");
5163: fprintf(ficlog, "**********\n");
5164: }
1.284 brouard 5165: /*
5166: Printing means of quantitative variables if any
5167: */
5168: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 5169: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 5170: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 5171: if(weightopt==1){
5172: printf(" Weighted mean and standard deviation of");
5173: fprintf(ficlog," Weighted mean and standard deviation of");
5174: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
5175: }
1.311 brouard 5176: /* mu = \frac{w x}{\sum w}
5177: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
5178: */
5179: 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]));
5180: 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]));
5181: 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 5182: }
5183: /* for (z1=1; z1<= nqtveff; z1++) { */
5184: /* for(m=1;m<=lastpass;m++){ */
5185: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
5186: /* } */
5187: /* } */
1.283 brouard 5188:
1.251 brouard 5189: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.330 brouard 5190: if((cptcovn==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
1.265 brouard 5191: fprintf(ficresp, " Age");
1.332 brouard 5192: if(nj==2) for (z1=1; z1<=cptcovn; z1++) fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.251 brouard 5193: for(i=1; i<=nlstate;i++) {
1.330 brouard 5194: if((cptcovn==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 5195: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
5196: }
1.330 brouard 5197: if((cptcovn==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 5198: fprintf(ficresphtm, "\n");
5199:
5200: /* Header of frequency table by age */
5201: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
5202: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 5203: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 5204: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5205: if(s2!=0 && m!=0)
5206: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 5207: }
1.226 brouard 5208: }
1.251 brouard 5209: fprintf(ficresphtmfr, "\n");
5210:
5211: /* For each age */
5212: for(iage=iagemin; iage <= iagemax+3; iage++){
5213: fprintf(ficresphtm,"<tr>");
5214: if(iage==iagemax+1){
5215: fprintf(ficlog,"1");
5216: fprintf(ficresphtmfr,"<tr><th>0</th> ");
5217: }else if(iage==iagemax+2){
5218: fprintf(ficlog,"0");
5219: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
5220: }else if(iage==iagemax+3){
5221: fprintf(ficlog,"Total");
5222: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
5223: }else{
1.240 brouard 5224: if(first==1){
1.251 brouard 5225: first=0;
5226: printf("See log file for details...\n");
5227: }
5228: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
5229: fprintf(ficlog,"Age %d", iage);
5230: }
1.265 brouard 5231: for(s1=1; s1 <=nlstate ; s1++){
5232: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
5233: pp[s1] += freq[s1][m][iage];
1.251 brouard 5234: }
1.265 brouard 5235: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5236: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 5237: pos += freq[s1][m][iage];
5238: if(pp[s1]>=1.e-10){
1.251 brouard 5239: if(first==1){
1.265 brouard 5240: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5241: }
1.265 brouard 5242: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 5243: }else{
5244: if(first==1)
1.265 brouard 5245: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
5246: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 5247: }
5248: }
5249:
1.265 brouard 5250: for(s1=1; s1 <=nlstate ; s1++){
5251: /* posprop[s1]=0; */
5252: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
5253: pp[s1] += freq[s1][m][iage];
5254: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
5255:
5256: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
5257: pos += pp[s1]; /* pos is the total number of transitions until this age */
5258: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
5259: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5260: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
5261: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
5262: }
5263:
5264: /* Writing ficresp */
1.330 brouard 5265: if(cptcovn==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5266: if( iage <= iagemax){
5267: fprintf(ficresp," %d",iage);
5268: }
5269: }else if( nj==2){
5270: if( iage <= iagemax){
5271: fprintf(ficresp," %d",iage);
1.332 brouard 5272: for (z1=1; z1<=cptcovn; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.265 brouard 5273: }
1.240 brouard 5274: }
1.265 brouard 5275: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 5276: if(pos>=1.e-5){
1.251 brouard 5277: if(first==1)
1.265 brouard 5278: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
5279: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 5280: }else{
5281: if(first==1)
1.265 brouard 5282: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
5283: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 5284: }
5285: if( iage <= iagemax){
5286: if(pos>=1.e-5){
1.330 brouard 5287: if(cptcovn==0 && nj==1){ /* no covariate and first pass */
1.265 brouard 5288: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5289: }else if( nj==2){
5290: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5291: }
5292: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
5293: /*probs[iage][s1][j1]= pp[s1]/pos;*/
5294: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
5295: } else{
1.330 brouard 5296: if((cptcovn==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
1.265 brouard 5297: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 5298: }
1.240 brouard 5299: }
1.265 brouard 5300: pospropt[s1] +=posprop[s1];
5301: } /* end loop s1 */
1.251 brouard 5302: /* pospropt=0.; */
1.265 brouard 5303: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 5304: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 5305: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 5306: if(first==1){
1.265 brouard 5307: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5308: }
1.265 brouard 5309: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
5310: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 5311: }
1.265 brouard 5312: if(s1!=0 && m!=0)
5313: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 5314: }
1.265 brouard 5315: } /* end loop s1 */
1.251 brouard 5316: posproptt=0.;
1.265 brouard 5317: for(s1=1; s1 <=nlstate; s1++){
5318: posproptt += pospropt[s1];
1.251 brouard 5319: }
5320: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 5321: fprintf(ficresphtm,"</tr>\n");
1.330 brouard 5322: if((cptcovn==0 && nj==1)|| nj==2 ) {
1.265 brouard 5323: if(iage <= iagemax)
5324: fprintf(ficresp,"\n");
1.240 brouard 5325: }
1.251 brouard 5326: if(first==1)
5327: printf("Others in log...\n");
5328: fprintf(ficlog,"\n");
5329: } /* end loop age iage */
1.265 brouard 5330:
1.251 brouard 5331: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 5332: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 5333: if(posproptt < 1.e-5){
1.265 brouard 5334: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 5335: }else{
1.265 brouard 5336: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 5337: }
1.226 brouard 5338: }
1.251 brouard 5339: fprintf(ficresphtm,"</tr>\n");
5340: fprintf(ficresphtm,"</table>\n");
5341: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 5342: if(posproptt < 1.e-5){
1.251 brouard 5343: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
5344: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 5345: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
5346: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 5347: invalidvarcomb[j1]=1;
1.226 brouard 5348: }else{
1.251 brouard 5349: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
5350: invalidvarcomb[j1]=0;
1.226 brouard 5351: }
1.251 brouard 5352: fprintf(ficresphtmfr,"</table>\n");
5353: fprintf(ficlog,"\n");
5354: if(j!=0){
5355: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 5356: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5357: for(k=1; k <=(nlstate+ndeath); k++){
5358: if (k != i) {
1.265 brouard 5359: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 5360: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 5361: if(j1==1){ /* All dummy covariates to zero */
5362: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
5363: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 5364: printf("%d%d ",i,k);
5365: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5366: 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]));
5367: 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]));
5368: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 5369: }
1.253 brouard 5370: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
5371: for(iage=iagemin; iage <= iagemax+3; iage++){
5372: x[iage]= (double)iage;
5373: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 5374: /* 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 5375: }
1.268 brouard 5376: /* Some are not finite, but linreg will ignore these ages */
5377: no=0;
1.253 brouard 5378: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5379: pstart[s1]=b;
5380: pstart[s1-1]=a;
1.252 brouard 5381: }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 */
5382: 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]);
5383: 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 5384: 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 5385: printf("%d%d ",i,k);
5386: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5387: 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 5388: }else{ /* Other cases, like quantitative fixed or varying covariates */
5389: ;
5390: }
5391: /* printf("%12.7f )", param[i][jj][k]); */
5392: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5393: s1++;
1.251 brouard 5394: } /* end jj */
5395: } /* end k!= i */
5396: } /* end k */
1.265 brouard 5397: } /* end i, s1 */
1.251 brouard 5398: } /* end j !=0 */
5399: } /* end selected combination of covariate j1 */
5400: if(j==0){ /* We can estimate starting values from the occurences in each case */
5401: printf("#Freqsummary: Starting values for the constants:\n");
5402: fprintf(ficlog,"\n");
1.265 brouard 5403: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5404: for(k=1; k <=(nlstate+ndeath); k++){
5405: if (k != i) {
5406: printf("%d%d ",i,k);
5407: fprintf(ficlog,"%d%d ",i,k);
5408: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5409: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5410: if(jj==1){ /* Age has to be done */
1.265 brouard 5411: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5412: 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]));
5413: 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 5414: }
5415: /* printf("%12.7f )", param[i][jj][k]); */
5416: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5417: s1++;
1.250 brouard 5418: }
1.251 brouard 5419: printf("\n");
5420: fprintf(ficlog,"\n");
1.250 brouard 5421: }
5422: }
1.284 brouard 5423: } /* end of state i */
1.251 brouard 5424: printf("#Freqsummary\n");
5425: fprintf(ficlog,"\n");
1.265 brouard 5426: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5427: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5428: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5429: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5430: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5431: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5432: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5433: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5434: /* } */
5435: }
1.265 brouard 5436: } /* end loop s1 */
1.251 brouard 5437:
5438: printf("\n");
5439: fprintf(ficlog,"\n");
5440: } /* end j=0 */
1.249 brouard 5441: } /* end j */
1.252 brouard 5442:
1.253 brouard 5443: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5444: for(i=1, jk=1; i <=nlstate; i++){
5445: for(j=1; j <=nlstate+ndeath; j++){
5446: if(j!=i){
5447: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5448: printf("%1d%1d",i,j);
5449: fprintf(ficparo,"%1d%1d",i,j);
5450: for(k=1; k<=ncovmodel;k++){
5451: /* printf(" %lf",param[i][j][k]); */
5452: /* fprintf(ficparo," %lf",param[i][j][k]); */
5453: p[jk]=pstart[jk];
5454: printf(" %f ",pstart[jk]);
5455: fprintf(ficparo," %f ",pstart[jk]);
5456: jk++;
5457: }
5458: printf("\n");
5459: fprintf(ficparo,"\n");
5460: }
5461: }
5462: }
5463: } /* end mle=-2 */
1.226 brouard 5464: dateintmean=dateintsum/k2cpt;
1.296 brouard 5465: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5466:
1.226 brouard 5467: fclose(ficresp);
5468: fclose(ficresphtm);
5469: fclose(ficresphtmfr);
1.283 brouard 5470: free_vector(idq,1,nqfveff);
1.226 brouard 5471: free_vector(meanq,1,nqfveff);
1.284 brouard 5472: free_vector(stdq,1,nqfveff);
1.226 brouard 5473: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5474: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5475: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5476: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5477: free_vector(pospropt,1,nlstate);
5478: free_vector(posprop,1,nlstate);
1.251 brouard 5479: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5480: free_vector(pp,1,nlstate);
5481: /* End of freqsummary */
5482: }
1.126 brouard 5483:
1.268 brouard 5484: /* Simple linear regression */
5485: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5486:
5487: /* y=a+bx regression */
5488: double sumx = 0.0; /* sum of x */
5489: double sumx2 = 0.0; /* sum of x**2 */
5490: double sumxy = 0.0; /* sum of x * y */
5491: double sumy = 0.0; /* sum of y */
5492: double sumy2 = 0.0; /* sum of y**2 */
5493: double sume2 = 0.0; /* sum of square or residuals */
5494: double yhat;
5495:
5496: double denom=0;
5497: int i;
5498: int ne=*no;
5499:
5500: for ( i=ifi, ne=0;i<=ila;i++) {
5501: if(!isfinite(x[i]) || !isfinite(y[i])){
5502: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5503: continue;
5504: }
5505: ne=ne+1;
5506: sumx += x[i];
5507: sumx2 += x[i]*x[i];
5508: sumxy += x[i] * y[i];
5509: sumy += y[i];
5510: sumy2 += y[i]*y[i];
5511: denom = (ne * sumx2 - sumx*sumx);
5512: /* 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); */
5513: }
5514:
5515: denom = (ne * sumx2 - sumx*sumx);
5516: if (denom == 0) {
5517: // vertical, slope m is infinity
5518: *b = INFINITY;
5519: *a = 0;
5520: if (r) *r = 0;
5521: return 1;
5522: }
5523:
5524: *b = (ne * sumxy - sumx * sumy) / denom;
5525: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5526: if (r!=NULL) {
5527: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5528: sqrt((sumx2 - sumx*sumx/ne) *
5529: (sumy2 - sumy*sumy/ne));
5530: }
5531: *no=ne;
5532: for ( i=ifi, ne=0;i<=ila;i++) {
5533: if(!isfinite(x[i]) || !isfinite(y[i])){
5534: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5535: continue;
5536: }
5537: ne=ne+1;
5538: yhat = y[i] - *a -*b* x[i];
5539: sume2 += yhat * yhat ;
5540:
5541: denom = (ne * sumx2 - sumx*sumx);
5542: /* 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); */
5543: }
5544: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5545: *sa= *sb * sqrt(sumx2/ne);
5546:
5547: return 0;
5548: }
5549:
1.126 brouard 5550: /************ Prevalence ********************/
1.227 brouard 5551: 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)
5552: {
5553: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5554: in each health status at the date of interview (if between dateprev1 and dateprev2).
5555: We still use firstpass and lastpass as another selection.
5556: */
1.126 brouard 5557:
1.227 brouard 5558: int i, m, jk, j1, bool, z1,j, iv;
5559: int mi; /* Effective wave */
5560: int iage;
5561: double agebegin, ageend;
5562:
5563: double **prop;
5564: double posprop;
5565: double y2; /* in fractional years */
5566: int iagemin, iagemax;
5567: int first; /** to stop verbosity which is redirected to log file */
5568:
5569: iagemin= (int) agemin;
5570: iagemax= (int) agemax;
5571: /*pp=vector(1,nlstate);*/
1.251 brouard 5572: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5573: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5574: j1=0;
1.222 brouard 5575:
1.227 brouard 5576: /*j=cptcoveff;*/
5577: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5578:
1.288 brouard 5579: first=0;
1.227 brouard 5580: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5581: for (i=1; i<=nlstate; i++)
1.251 brouard 5582: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5583: prop[i][iage]=0.0;
5584: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5585: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5586: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5587:
5588: for (i=1; i<=imx; i++) { /* Each individual */
5589: bool=1;
5590: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5591: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5592: m=mw[mi][i];
5593: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5594: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5595: for (z1=1; z1<=cptcoveff; z1++){
5596: if( Fixed[Tmodelind[z1]]==1){
5597: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
1.332 brouard 5598: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) /* iv=1 to ntv, right modality */
1.227 brouard 5599: bool=0;
5600: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
1.332 brouard 5601: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]) {
1.227 brouard 5602: bool=0;
5603: }
5604: }
5605: if(bool==1){ /* Otherwise we skip that wave/person */
5606: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5607: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5608: if(m >=firstpass && m <=lastpass){
5609: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5610: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5611: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5612: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5613: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5614: 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);
5615: exit(1);
5616: }
5617: if (s[m][i]>0 && s[m][i]<=nlstate) {
5618: /*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]]);*/
5619: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5620: prop[s[m][i]][iagemax+3] += weight[i];
5621: } /* end valid statuses */
5622: } /* end selection of dates */
5623: } /* end selection of waves */
5624: } /* end bool */
5625: } /* end wave */
5626: } /* end individual */
5627: for(i=iagemin; i <= iagemax+3; i++){
5628: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5629: posprop += prop[jk][i];
5630: }
5631:
5632: for(jk=1; jk <=nlstate ; jk++){
5633: if( i <= iagemax){
5634: if(posprop>=1.e-5){
5635: probs[i][jk][j1]= prop[jk][i]/posprop;
5636: } else{
1.288 brouard 5637: if(!first){
5638: first=1;
1.266 brouard 5639: 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]);
5640: }else{
1.288 brouard 5641: 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 5642: }
5643: }
5644: }
5645: }/* end jk */
5646: }/* end i */
1.222 brouard 5647: /*} *//* end i1 */
1.227 brouard 5648: } /* end j1 */
1.222 brouard 5649:
1.227 brouard 5650: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5651: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5652: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5653: } /* End of prevalence */
1.126 brouard 5654:
5655: /************* Waves Concatenation ***************/
5656:
5657: 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)
5658: {
1.298 brouard 5659: /* 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 5660: Death is a valid wave (if date is known).
5661: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5662: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5663: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5664: */
1.126 brouard 5665:
1.224 brouard 5666: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5667: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5668: double sum=0., jmean=0.;*/
1.224 brouard 5669: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5670: int j, k=0,jk, ju, jl;
5671: double sum=0.;
5672: first=0;
1.214 brouard 5673: firstwo=0;
1.217 brouard 5674: firsthree=0;
1.218 brouard 5675: firstfour=0;
1.164 brouard 5676: jmin=100000;
1.126 brouard 5677: jmax=-1;
5678: jmean=0.;
1.224 brouard 5679:
5680: /* Treating live states */
1.214 brouard 5681: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5682: mi=0; /* First valid wave */
1.227 brouard 5683: mli=0; /* Last valid wave */
1.309 brouard 5684: m=firstpass; /* Loop on waves */
5685: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5686: 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 */
5687: mli=m-1;/* mw[++mi][i]=m-1; */
5688: }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 5689: 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 5690: mli=m;
1.224 brouard 5691: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5692: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5693: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5694: }
1.309 brouard 5695: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5696: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5697: break;
1.224 brouard 5698: #else
1.317 brouard 5699: 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 5700: if(firsthree == 0){
1.302 brouard 5701: 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 5702: firsthree=1;
1.317 brouard 5703: }else if(firsthree >=1 && firsthree < 10){
5704: 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);
5705: firsthree++;
5706: }else if(firsthree == 10){
5707: printf("Information, too many Information flags: no more reported to log either\n");
5708: fprintf(ficlog,"Information, too many Information flags: no more reported to log either\n");
5709: firsthree++;
5710: }else{
5711: firsthree++;
1.227 brouard 5712: }
1.309 brouard 5713: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5714: mli=m;
5715: }
5716: if(s[m][i]==-2){ /* Vital status is really unknown */
5717: nbwarn++;
1.309 brouard 5718: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5719: 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);
5720: 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);
5721: }
5722: break;
5723: }
5724: break;
1.224 brouard 5725: #endif
1.227 brouard 5726: }/* End m >= lastpass */
1.126 brouard 5727: }/* end while */
1.224 brouard 5728:
1.227 brouard 5729: /* 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 5730: /* After last pass */
1.224 brouard 5731: /* Treating death states */
1.214 brouard 5732: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5733: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5734: /* } */
1.126 brouard 5735: mi++; /* Death is another wave */
5736: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5737: /* Only death is a correct wave */
1.126 brouard 5738: mw[mi][i]=m;
1.257 brouard 5739: } /* else not in a death state */
1.224 brouard 5740: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5741: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5742: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5743: 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 5744: nbwarn++;
5745: if(firstfiv==0){
1.309 brouard 5746: 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 5747: firstfiv=1;
5748: }else{
1.309 brouard 5749: 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 5750: }
1.309 brouard 5751: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5752: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5753: nberr++;
5754: if(firstwo==0){
1.309 brouard 5755: 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 5756: firstwo=1;
5757: }
1.309 brouard 5758: 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 5759: }
1.257 brouard 5760: }else{ /* if date of interview is unknown */
1.227 brouard 5761: /* death is known but not confirmed by death status at any wave */
5762: if(firstfour==0){
1.309 brouard 5763: 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 5764: firstfour=1;
5765: }
1.309 brouard 5766: 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 5767: }
1.224 brouard 5768: } /* end if date of death is known */
5769: #endif
1.309 brouard 5770: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5771: /* wav[i]=mw[mi][i]; */
1.126 brouard 5772: if(mi==0){
5773: nbwarn++;
5774: if(first==0){
1.227 brouard 5775: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5776: first=1;
1.126 brouard 5777: }
5778: if(first==1){
1.227 brouard 5779: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5780: }
5781: } /* end mi==0 */
5782: } /* End individuals */
1.214 brouard 5783: /* wav and mw are no more changed */
1.223 brouard 5784:
1.317 brouard 5785: printf("Information, you have to check %d informations which haven't been logged!\n",firsthree);
5786: fprintf(ficlog,"Information, you have to check %d informations which haven't been logged!\n",firsthree);
5787:
5788:
1.126 brouard 5789: for(i=1; i<=imx; i++){
5790: for(mi=1; mi<wav[i];mi++){
5791: if (stepm <=0)
1.227 brouard 5792: dh[mi][i]=1;
1.126 brouard 5793: else{
1.260 brouard 5794: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5795: if (agedc[i] < 2*AGESUP) {
5796: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5797: if(j==0) j=1; /* Survives at least one month after exam */
5798: else if(j<0){
5799: nberr++;
5800: 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]);
5801: j=1; /* Temporary Dangerous patch */
5802: 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);
5803: 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]);
5804: 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);
5805: }
5806: k=k+1;
5807: if (j >= jmax){
5808: jmax=j;
5809: ijmax=i;
5810: }
5811: if (j <= jmin){
5812: jmin=j;
5813: ijmin=i;
5814: }
5815: sum=sum+j;
5816: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5817: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5818: }
5819: }
5820: else{
5821: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5822: /* 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 5823:
1.227 brouard 5824: k=k+1;
5825: if (j >= jmax) {
5826: jmax=j;
5827: ijmax=i;
5828: }
5829: else if (j <= jmin){
5830: jmin=j;
5831: ijmin=i;
5832: }
5833: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5834: /*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]);*/
5835: if(j<0){
5836: nberr++;
5837: 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]);
5838: 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]);
5839: }
5840: sum=sum+j;
5841: }
5842: jk= j/stepm;
5843: jl= j -jk*stepm;
5844: ju= j -(jk+1)*stepm;
5845: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5846: if(jl==0){
5847: dh[mi][i]=jk;
5848: bh[mi][i]=0;
5849: }else{ /* We want a negative bias in order to only have interpolation ie
5850: * to avoid the price of an extra matrix product in likelihood */
5851: dh[mi][i]=jk+1;
5852: bh[mi][i]=ju;
5853: }
5854: }else{
5855: if(jl <= -ju){
5856: dh[mi][i]=jk;
5857: bh[mi][i]=jl; /* bias is positive if real duration
5858: * is higher than the multiple of stepm and negative otherwise.
5859: */
5860: }
5861: else{
5862: dh[mi][i]=jk+1;
5863: bh[mi][i]=ju;
5864: }
5865: if(dh[mi][i]==0){
5866: dh[mi][i]=1; /* At least one step */
5867: bh[mi][i]=ju; /* At least one step */
5868: /* 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);*/
5869: }
5870: } /* end if mle */
1.126 brouard 5871: }
5872: } /* end wave */
5873: }
5874: jmean=sum/k;
5875: 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 5876: 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 5877: }
1.126 brouard 5878:
5879: /*********** Tricode ****************************/
1.220 brouard 5880: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5881: {
5882: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5883: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5884: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5885: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5886: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5887: */
1.130 brouard 5888:
1.242 brouard 5889: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5890: int modmaxcovj=0; /* Modality max of covariates j */
5891: int cptcode=0; /* Modality max of covariates j */
5892: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5893:
5894:
1.242 brouard 5895: /* cptcoveff=0; */
5896: /* *cptcov=0; */
1.126 brouard 5897:
1.242 brouard 5898: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5899: for (k=1; k <= maxncov; k++)
5900: for(j=1; j<=2; j++)
5901: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5902:
1.242 brouard 5903: /* Loop on covariates without age and products and no quantitative variable */
5904: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5905: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5906: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5907: switch(Fixed[k]) {
5908: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5909: modmaxcovj=0;
5910: modmincovj=0;
1.242 brouard 5911: 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*/
5912: ij=(int)(covar[Tvar[k]][i]);
5913: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5914: * If product of Vn*Vm, still boolean *:
5915: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5916: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5917: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5918: modality of the nth covariate of individual i. */
5919: if (ij > modmaxcovj)
5920: modmaxcovj=ij;
5921: else if (ij < modmincovj)
5922: modmincovj=ij;
1.287 brouard 5923: if (ij <0 || ij >1 ){
1.311 brouard 5924: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5925: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5926: fflush(ficlog);
5927: exit(1);
1.287 brouard 5928: }
5929: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5930: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5931: exit(1);
5932: }else
5933: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5934: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5935: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5936: /* getting the maximum value of the modality of the covariate
5937: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5938: female ies 1, then modmaxcovj=1.
5939: */
5940: } /* end for loop on individuals i */
5941: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5942: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5943: cptcode=modmaxcovj;
5944: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5945: /*for (i=0; i<=cptcode; i++) {*/
5946: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5947: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5948: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5949: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5950: if( j != -1){
5951: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5952: covariate for which somebody answered excluding
5953: undefined. Usually 2: 0 and 1. */
5954: }
5955: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5956: covariate for which somebody answered including
5957: undefined. Usually 3: -1, 0 and 1. */
5958: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5959: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5960: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5961:
1.242 brouard 5962: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5963: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5964: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5965: /* modmincovj=3; modmaxcovj = 7; */
5966: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5967: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5968: /* defining two dummy variables: variables V1_1 and V1_2.*/
5969: /* nbcode[Tvar[j]][ij]=k; */
5970: /* nbcode[Tvar[j]][1]=0; */
5971: /* nbcode[Tvar[j]][2]=1; */
5972: /* nbcode[Tvar[j]][3]=2; */
5973: /* To be continued (not working yet). */
5974: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5975:
5976: /* 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*/
5977: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5978: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5979: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5980: /*, could be restored in the future */
5981: 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 5982: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5983: break;
5984: }
5985: ij++;
1.287 brouard 5986: 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 5987: cptcode = ij; /* New max modality for covar j */
5988: } /* end of loop on modality i=-1 to 1 or more */
5989: break;
5990: case 1: /* Testing on varying covariate, could be simple and
5991: * should look at waves or product of fixed *
5992: * varying. No time to test -1, assuming 0 and 1 only */
5993: ij=0;
5994: for(i=0; i<=1;i++){
5995: nbcode[Tvar[k]][++ij]=i;
5996: }
5997: break;
5998: default:
5999: break;
6000: } /* end switch */
6001: } /* end dummy test */
1.311 brouard 6002: if(Dummy[k]==1 && Typevar[k] !=1){ /* Dummy covariate and not age product */
6003: 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*/
6004: if(isnan(covar[Tvar[k]][i])){
6005: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6006: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
6007: fflush(ficlog);
6008: exit(1);
6009: }
6010: }
6011: }
1.287 brouard 6012: } /* 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 6013:
6014: for (k=-1; k< maxncov; k++) Ndum[k]=0;
6015: /* Look at fixed dummy (single or product) covariates to check empty modalities */
6016: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
6017: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
6018: 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 */
6019: 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 */
6020: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
6021: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
6022:
6023: ij=0;
6024: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
6025: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
6026: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
6027: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
6028: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
6029: /* If product not in single variable we don't print results */
6030: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
6031: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
6032: 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*/
6033: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
6034: 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 */
6035: if(Fixed[k]!=0)
6036: anyvaryingduminmodel=1;
6037: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
6038: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
6039: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
6040: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
6041: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
6042: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
6043: }
6044: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
6045: /* ij--; */
6046: /* cptcoveff=ij; /\*Number of total covariates*\/ */
1.330 brouard 6047: *cptcov=ij; /* cptcov= Number of total real effective covariates: effective (used as cptcoveff in other functions)
1.242 brouard 6048: * because they can be excluded from the model and real
6049: * if in the model but excluded because missing values, but how to get k from ij?*/
6050: for(j=ij+1; j<= cptcovt; j++){
6051: Tvaraff[j]=0;
6052: Tmodelind[j]=0;
6053: }
6054: for(j=ntveff+1; j<= cptcovt; j++){
6055: TmodelInvind[j]=0;
6056: }
6057: /* To be sorted */
6058: ;
6059: }
1.126 brouard 6060:
1.145 brouard 6061:
1.126 brouard 6062: /*********** Health Expectancies ****************/
6063:
1.235 brouard 6064: 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 6065:
6066: {
6067: /* Health expectancies, no variances */
1.329 brouard 6068: /* cij is the combination in the list of combination of dummy covariates */
6069: /* strstart is a string of time at start of computing */
1.164 brouard 6070: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 6071: int nhstepma, nstepma; /* Decreasing with age */
6072: double age, agelim, hf;
6073: double ***p3mat;
6074: double eip;
6075:
1.238 brouard 6076: /* pstamp(ficreseij); */
1.126 brouard 6077: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
6078: fprintf(ficreseij,"# Age");
6079: for(i=1; i<=nlstate;i++){
6080: for(j=1; j<=nlstate;j++){
6081: fprintf(ficreseij," e%1d%1d ",i,j);
6082: }
6083: fprintf(ficreseij," e%1d. ",i);
6084: }
6085: fprintf(ficreseij,"\n");
6086:
6087:
6088: if(estepm < stepm){
6089: printf ("Problem %d lower than %d\n",estepm, stepm);
6090: }
6091: else hstepm=estepm;
6092: /* We compute the life expectancy from trapezoids spaced every estepm months
6093: * This is mainly to measure the difference between two models: for example
6094: * if stepm=24 months pijx are given only every 2 years and by summing them
6095: * we are calculating an estimate of the Life Expectancy assuming a linear
6096: * progression in between and thus overestimating or underestimating according
6097: * to the curvature of the survival function. If, for the same date, we
6098: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6099: * to compare the new estimate of Life expectancy with the same linear
6100: * hypothesis. A more precise result, taking into account a more precise
6101: * curvature will be obtained if estepm is as small as stepm. */
6102:
6103: /* For example we decided to compute the life expectancy with the smallest unit */
6104: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6105: nhstepm is the number of hstepm from age to agelim
6106: nstepm is the number of stepm from age to agelin.
1.270 brouard 6107: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 6108: and note for a fixed period like estepm months */
6109: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6110: survival function given by stepm (the optimization length). Unfortunately it
6111: means that if the survival funtion is printed only each two years of age and if
6112: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6113: results. So we changed our mind and took the option of the best precision.
6114: */
6115: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6116:
6117: agelim=AGESUP;
6118: /* If stepm=6 months */
6119: /* Computed by stepm unit matrices, product of hstepm matrices, stored
6120: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
6121:
6122: /* nhstepm age range expressed in number of stepm */
6123: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6124: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6125: /* if (stepm >= YEARM) hstepm=1;*/
6126: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6127: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6128:
6129: for (age=bage; age<=fage; age ++){
6130: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6131: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6132: /* if (stepm >= YEARM) hstepm=1;*/
6133: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
6134:
6135: /* If stepm=6 months */
6136: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6137: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
1.330 brouard 6138: /* printf("HELLO evsij Entering hpxij age=%d cij=%d hstepm=%d x[1]=%f nres=%d\n",(int) age, cij, hstepm, x[1], nres); */
1.235 brouard 6139: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 6140:
6141: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6142:
6143: printf("%d|",(int)age);fflush(stdout);
6144: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6145:
6146: /* Computing expectancies */
6147: for(i=1; i<=nlstate;i++)
6148: for(j=1; j<=nlstate;j++)
6149: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6150: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
6151:
6152: /* 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]);*/
6153:
6154: }
6155:
6156: fprintf(ficreseij,"%3.0f",age );
6157: for(i=1; i<=nlstate;i++){
6158: eip=0;
6159: for(j=1; j<=nlstate;j++){
6160: eip +=eij[i][j][(int)age];
6161: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
6162: }
6163: fprintf(ficreseij,"%9.4f", eip );
6164: }
6165: fprintf(ficreseij,"\n");
6166:
6167: }
6168: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6169: printf("\n");
6170: fprintf(ficlog,"\n");
6171:
6172: }
6173:
1.235 brouard 6174: 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 6175:
6176: {
6177: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 6178: to initial status i, ei. .
1.126 brouard 6179: */
6180: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
6181: int nhstepma, nstepma; /* Decreasing with age */
6182: double age, agelim, hf;
6183: double ***p3matp, ***p3matm, ***varhe;
6184: double **dnewm,**doldm;
6185: double *xp, *xm;
6186: double **gp, **gm;
6187: double ***gradg, ***trgradg;
6188: int theta;
6189:
6190: double eip, vip;
6191:
6192: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
6193: xp=vector(1,npar);
6194: xm=vector(1,npar);
6195: dnewm=matrix(1,nlstate*nlstate,1,npar);
6196: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
6197:
6198: pstamp(ficresstdeij);
6199: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
6200: fprintf(ficresstdeij,"# Age");
6201: for(i=1; i<=nlstate;i++){
6202: for(j=1; j<=nlstate;j++)
6203: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
6204: fprintf(ficresstdeij," e%1d. ",i);
6205: }
6206: fprintf(ficresstdeij,"\n");
6207:
6208: pstamp(ficrescveij);
6209: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
6210: fprintf(ficrescveij,"# Age");
6211: for(i=1; i<=nlstate;i++)
6212: for(j=1; j<=nlstate;j++){
6213: cptj= (j-1)*nlstate+i;
6214: for(i2=1; i2<=nlstate;i2++)
6215: for(j2=1; j2<=nlstate;j2++){
6216: cptj2= (j2-1)*nlstate+i2;
6217: if(cptj2 <= cptj)
6218: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
6219: }
6220: }
6221: fprintf(ficrescveij,"\n");
6222:
6223: if(estepm < stepm){
6224: printf ("Problem %d lower than %d\n",estepm, stepm);
6225: }
6226: else hstepm=estepm;
6227: /* We compute the life expectancy from trapezoids spaced every estepm months
6228: * This is mainly to measure the difference between two models: for example
6229: * if stepm=24 months pijx are given only every 2 years and by summing them
6230: * we are calculating an estimate of the Life Expectancy assuming a linear
6231: * progression in between and thus overestimating or underestimating according
6232: * to the curvature of the survival function. If, for the same date, we
6233: * estimate the model with stepm=1 month, we can keep estepm to 24 months
6234: * to compare the new estimate of Life expectancy with the same linear
6235: * hypothesis. A more precise result, taking into account a more precise
6236: * curvature will be obtained if estepm is as small as stepm. */
6237:
6238: /* For example we decided to compute the life expectancy with the smallest unit */
6239: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6240: nhstepm is the number of hstepm from age to agelim
6241: nstepm is the number of stepm from age to agelin.
6242: Look at hpijx to understand the reason of that which relies in memory size
6243: and note for a fixed period like estepm months */
6244: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
6245: survival function given by stepm (the optimization length). Unfortunately it
6246: means that if the survival funtion is printed only each two years of age and if
6247: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6248: results. So we changed our mind and took the option of the best precision.
6249: */
6250: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6251:
6252: /* If stepm=6 months */
6253: /* nhstepm age range expressed in number of stepm */
6254: agelim=AGESUP;
6255: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
6256: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6257: /* if (stepm >= YEARM) hstepm=1;*/
6258: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6259:
6260: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6261: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6262: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
6263: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
6264: gp=matrix(0,nhstepm,1,nlstate*nlstate);
6265: gm=matrix(0,nhstepm,1,nlstate*nlstate);
6266:
6267: for (age=bage; age<=fage; age ++){
6268: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
6269: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
6270: /* if (stepm >= YEARM) hstepm=1;*/
6271: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 6272:
1.126 brouard 6273: /* If stepm=6 months */
6274: /* Computed by stepm unit matrices, product of hstepma matrices, stored
6275: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
6276:
6277: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 6278:
1.126 brouard 6279: /* Computing Variances of health expectancies */
6280: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
6281: decrease memory allocation */
6282: for(theta=1; theta <=npar; theta++){
6283: for(i=1; i<=npar; i++){
1.222 brouard 6284: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6285: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 6286: }
1.235 brouard 6287: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
6288: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 6289:
1.126 brouard 6290: for(j=1; j<= nlstate; j++){
1.222 brouard 6291: for(i=1; i<=nlstate; i++){
6292: for(h=0; h<=nhstepm-1; h++){
6293: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
6294: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
6295: }
6296: }
1.126 brouard 6297: }
1.218 brouard 6298:
1.126 brouard 6299: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 6300: for(h=0; h<=nhstepm-1; h++){
6301: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
6302: }
1.126 brouard 6303: }/* End theta */
6304:
6305:
6306: for(h=0; h<=nhstepm-1; h++)
6307: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 6308: for(theta=1; theta <=npar; theta++)
6309: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 6310:
1.218 brouard 6311:
1.222 brouard 6312: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 6313: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 6314: varhe[ij][ji][(int)age] =0.;
1.218 brouard 6315:
1.222 brouard 6316: printf("%d|",(int)age);fflush(stdout);
6317: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
6318: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 6319: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 6320: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
6321: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
6322: for(ij=1;ij<=nlstate*nlstate;ij++)
6323: for(ji=1;ji<=nlstate*nlstate;ji++)
6324: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 6325: }
6326: }
1.320 brouard 6327: /* if((int)age ==50){ */
6328: /* printf(" age=%d cij=%d nres=%d varhe[%d][%d]=%f ",(int)age, cij, nres, 1,2,varhe[1][2]); */
6329: /* } */
1.126 brouard 6330: /* Computing expectancies */
1.235 brouard 6331: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 6332: for(i=1; i<=nlstate;i++)
6333: for(j=1; j<=nlstate;j++)
1.222 brouard 6334: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
6335: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 6336:
1.222 brouard 6337: /* 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 6338:
1.222 brouard 6339: }
1.269 brouard 6340:
6341: /* Standard deviation of expectancies ij */
1.126 brouard 6342: fprintf(ficresstdeij,"%3.0f",age );
6343: for(i=1; i<=nlstate;i++){
6344: eip=0.;
6345: vip=0.;
6346: for(j=1; j<=nlstate;j++){
1.222 brouard 6347: eip += eij[i][j][(int)age];
6348: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
6349: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
6350: 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 6351: }
6352: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
6353: }
6354: fprintf(ficresstdeij,"\n");
1.218 brouard 6355:
1.269 brouard 6356: /* Variance of expectancies ij */
1.126 brouard 6357: fprintf(ficrescveij,"%3.0f",age );
6358: for(i=1; i<=nlstate;i++)
6359: for(j=1; j<=nlstate;j++){
1.222 brouard 6360: cptj= (j-1)*nlstate+i;
6361: for(i2=1; i2<=nlstate;i2++)
6362: for(j2=1; j2<=nlstate;j2++){
6363: cptj2= (j2-1)*nlstate+i2;
6364: if(cptj2 <= cptj)
6365: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
6366: }
1.126 brouard 6367: }
6368: fprintf(ficrescveij,"\n");
1.218 brouard 6369:
1.126 brouard 6370: }
6371: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
6372: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
6373: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
6374: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
6375: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6376: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6377: printf("\n");
6378: fprintf(ficlog,"\n");
1.218 brouard 6379:
1.126 brouard 6380: free_vector(xm,1,npar);
6381: free_vector(xp,1,npar);
6382: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
6383: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
6384: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
6385: }
1.218 brouard 6386:
1.126 brouard 6387: /************ Variance ******************/
1.235 brouard 6388: 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 6389: {
1.279 brouard 6390: /** Variance of health expectancies
6391: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
6392: * double **newm;
6393: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6394: */
1.218 brouard 6395:
6396: /* int movingaverage(); */
6397: double **dnewm,**doldm;
6398: double **dnewmp,**doldmp;
6399: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6400: int first=0;
1.218 brouard 6401: int k;
6402: double *xp;
1.279 brouard 6403: double **gp, **gm; /**< for var eij */
6404: double ***gradg, ***trgradg; /**< for var eij */
6405: double **gradgp, **trgradgp; /**< for var p point j */
6406: double *gpp, *gmp; /**< for var p point j */
6407: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6408: double ***p3mat;
6409: double age,agelim, hf;
6410: /* double ***mobaverage; */
6411: int theta;
6412: char digit[4];
6413: char digitp[25];
6414:
6415: char fileresprobmorprev[FILENAMELENGTH];
6416:
6417: if(popbased==1){
6418: if(mobilav!=0)
6419: strcpy(digitp,"-POPULBASED-MOBILAV_");
6420: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6421: }
6422: else
6423: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6424:
1.218 brouard 6425: /* if (mobilav!=0) { */
6426: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6427: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6428: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6429: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6430: /* } */
6431: /* } */
6432:
6433: strcpy(fileresprobmorprev,"PRMORPREV-");
6434: sprintf(digit,"%-d",ij);
6435: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6436: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6437: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6438: strcat(fileresprobmorprev,fileresu);
6439: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6440: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6441: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6442: }
6443: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6444: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6445: pstamp(ficresprobmorprev);
6446: 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 6447: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
6448: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 6449: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 6450: }
6451: for(j=1;j<=cptcoveff;j++)
1.332 brouard 6452: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,TnsdVar[Tvaraff[j]])]);
1.238 brouard 6453: fprintf(ficresprobmorprev,"\n");
6454:
1.218 brouard 6455: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6456: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6457: fprintf(ficresprobmorprev," p.%-d SE",j);
6458: for(i=1; i<=nlstate;i++)
6459: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6460: }
6461: fprintf(ficresprobmorprev,"\n");
6462:
6463: fprintf(ficgp,"\n# Routine varevsij");
6464: fprintf(ficgp,"\nunset title \n");
6465: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6466: 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");
6467: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6468:
1.218 brouard 6469: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6470: pstamp(ficresvij);
6471: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6472: if(popbased==1)
6473: 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);
6474: else
6475: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6476: fprintf(ficresvij,"# Age");
6477: for(i=1; i<=nlstate;i++)
6478: for(j=1; j<=nlstate;j++)
6479: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6480: fprintf(ficresvij,"\n");
6481:
6482: xp=vector(1,npar);
6483: dnewm=matrix(1,nlstate,1,npar);
6484: doldm=matrix(1,nlstate,1,nlstate);
6485: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6486: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6487:
6488: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6489: gpp=vector(nlstate+1,nlstate+ndeath);
6490: gmp=vector(nlstate+1,nlstate+ndeath);
6491: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6492:
1.218 brouard 6493: if(estepm < stepm){
6494: printf ("Problem %d lower than %d\n",estepm, stepm);
6495: }
6496: else hstepm=estepm;
6497: /* For example we decided to compute the life expectancy with the smallest unit */
6498: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6499: nhstepm is the number of hstepm from age to agelim
6500: nstepm is the number of stepm from age to agelim.
6501: Look at function hpijx to understand why because of memory size limitations,
6502: we decided (b) to get a life expectancy respecting the most precise curvature of the
6503: survival function given by stepm (the optimization length). Unfortunately it
6504: means that if the survival funtion is printed every two years of age and if
6505: you sum them up and add 1 year (area under the trapezoids) you won't get the same
6506: results. So we changed our mind and took the option of the best precision.
6507: */
6508: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
6509: agelim = AGESUP;
6510: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6511: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6512: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6513: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6514: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6515: gp=matrix(0,nhstepm,1,nlstate);
6516: gm=matrix(0,nhstepm,1,nlstate);
6517:
6518:
6519: for(theta=1; theta <=npar; theta++){
6520: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6521: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6522: }
1.279 brouard 6523: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6524: * returns into prlim .
1.288 brouard 6525: */
1.242 brouard 6526: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6527:
6528: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6529: if (popbased==1) {
6530: if(mobilav ==0){
6531: for(i=1; i<=nlstate;i++)
6532: prlim[i][i]=probs[(int)age][i][ij];
6533: }else{ /* mobilav */
6534: for(i=1; i<=nlstate;i++)
6535: prlim[i][i]=mobaverage[(int)age][i][ij];
6536: }
6537: }
1.295 brouard 6538: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6539: */
6540: 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 6541: /**< 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 6542: * at horizon h in state j including mortality.
6543: */
1.218 brouard 6544: for(j=1; j<= nlstate; j++){
6545: for(h=0; h<=nhstepm; h++){
6546: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6547: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6548: }
6549: }
1.279 brouard 6550: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6551: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6552: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6553: */
6554: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6555: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6556: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6557: }
6558:
6559: /* Again with minus shift */
1.218 brouard 6560:
6561: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6562: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6563:
1.242 brouard 6564: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6565:
6566: if (popbased==1) {
6567: if(mobilav ==0){
6568: for(i=1; i<=nlstate;i++)
6569: prlim[i][i]=probs[(int)age][i][ij];
6570: }else{ /* mobilav */
6571: for(i=1; i<=nlstate;i++)
6572: prlim[i][i]=mobaverage[(int)age][i][ij];
6573: }
6574: }
6575:
1.235 brouard 6576: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6577:
6578: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6579: for(h=0; h<=nhstepm; h++){
6580: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6581: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6582: }
6583: }
6584: /* This for computing probability of death (h=1 means
6585: computed over hstepm matrices product = hstepm*stepm months)
6586: as a weighted average of prlim.
6587: */
6588: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6589: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6590: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6591: }
1.279 brouard 6592: /* end shifting computations */
6593:
6594: /**< Computing gradient matrix at horizon h
6595: */
1.218 brouard 6596: for(j=1; j<= nlstate; j++) /* vareij */
6597: for(h=0; h<=nhstepm; h++){
6598: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6599: }
1.279 brouard 6600: /**< Gradient of overall mortality p.3 (or p.j)
6601: */
6602: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6603: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6604: }
6605:
6606: } /* End theta */
1.279 brouard 6607:
6608: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6609: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6610:
6611: for(h=0; h<=nhstepm; h++) /* veij */
6612: for(j=1; j<=nlstate;j++)
6613: for(theta=1; theta <=npar; theta++)
6614: trgradg[h][j][theta]=gradg[h][theta][j];
6615:
6616: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6617: for(theta=1; theta <=npar; theta++)
6618: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6619: /**< as well as its transposed matrix
6620: */
1.218 brouard 6621:
6622: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6623: for(i=1;i<=nlstate;i++)
6624: for(j=1;j<=nlstate;j++)
6625: vareij[i][j][(int)age] =0.;
1.279 brouard 6626:
6627: /* Computing trgradg by matcov by gradg at age and summing over h
6628: * and k (nhstepm) formula 15 of article
6629: * Lievre-Brouard-Heathcote
6630: */
6631:
1.218 brouard 6632: for(h=0;h<=nhstepm;h++){
6633: for(k=0;k<=nhstepm;k++){
6634: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6635: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6636: for(i=1;i<=nlstate;i++)
6637: for(j=1;j<=nlstate;j++)
6638: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6639: }
6640: }
6641:
1.279 brouard 6642: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6643: * p.j overall mortality formula 49 but computed directly because
6644: * we compute the grad (wix pijx) instead of grad (pijx),even if
6645: * wix is independent of theta.
6646: */
1.218 brouard 6647: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6648: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6649: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6650: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6651: varppt[j][i]=doldmp[j][i];
6652: /* end ppptj */
6653: /* x centered again */
6654:
1.242 brouard 6655: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6656:
6657: if (popbased==1) {
6658: if(mobilav ==0){
6659: for(i=1; i<=nlstate;i++)
6660: prlim[i][i]=probs[(int)age][i][ij];
6661: }else{ /* mobilav */
6662: for(i=1; i<=nlstate;i++)
6663: prlim[i][i]=mobaverage[(int)age][i][ij];
6664: }
6665: }
6666:
6667: /* This for computing probability of death (h=1 means
6668: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6669: as a weighted average of prlim.
6670: */
1.235 brouard 6671: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6672: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6673: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6674: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6675: }
6676: /* end probability of death */
6677:
6678: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6679: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6680: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6681: for(i=1; i<=nlstate;i++){
6682: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6683: }
6684: }
6685: fprintf(ficresprobmorprev,"\n");
6686:
6687: fprintf(ficresvij,"%.0f ",age );
6688: for(i=1; i<=nlstate;i++)
6689: for(j=1; j<=nlstate;j++){
6690: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6691: }
6692: fprintf(ficresvij,"\n");
6693: free_matrix(gp,0,nhstepm,1,nlstate);
6694: free_matrix(gm,0,nhstepm,1,nlstate);
6695: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6696: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6697: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6698: } /* End age */
6699: free_vector(gpp,nlstate+1,nlstate+ndeath);
6700: free_vector(gmp,nlstate+1,nlstate+ndeath);
6701: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6702: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6703: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6704: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6705: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6706: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6707: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6708: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6709: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6710: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6711: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6712: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6713: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6714: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6715: 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);
6716: /* 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 6717: */
1.218 brouard 6718: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6719: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6720:
1.218 brouard 6721: free_vector(xp,1,npar);
6722: free_matrix(doldm,1,nlstate,1,nlstate);
6723: free_matrix(dnewm,1,nlstate,1,npar);
6724: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6725: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6726: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6727: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6728: fclose(ficresprobmorprev);
6729: fflush(ficgp);
6730: fflush(fichtm);
6731: } /* end varevsij */
1.126 brouard 6732:
6733: /************ Variance of prevlim ******************/
1.269 brouard 6734: 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 6735: {
1.205 brouard 6736: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6737: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6738:
1.268 brouard 6739: double **dnewmpar,**doldm;
1.126 brouard 6740: int i, j, nhstepm, hstepm;
6741: double *xp;
6742: double *gp, *gm;
6743: double **gradg, **trgradg;
1.208 brouard 6744: double **mgm, **mgp;
1.126 brouard 6745: double age,agelim;
6746: int theta;
6747:
6748: pstamp(ficresvpl);
1.288 brouard 6749: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6750: fprintf(ficresvpl,"# Age ");
6751: if(nresult >=1)
6752: fprintf(ficresvpl," Result# ");
1.126 brouard 6753: for(i=1; i<=nlstate;i++)
6754: fprintf(ficresvpl," %1d-%1d",i,i);
6755: fprintf(ficresvpl,"\n");
6756:
6757: xp=vector(1,npar);
1.268 brouard 6758: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6759: doldm=matrix(1,nlstate,1,nlstate);
6760:
6761: hstepm=1*YEARM; /* Every year of age */
6762: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6763: agelim = AGESUP;
6764: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6765: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6766: if (stepm >= YEARM) hstepm=1;
6767: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6768: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6769: mgp=matrix(1,npar,1,nlstate);
6770: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6771: gp=vector(1,nlstate);
6772: gm=vector(1,nlstate);
6773:
6774: for(theta=1; theta <=npar; theta++){
6775: for(i=1; i<=npar; i++){ /* Computes gradient */
6776: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6777: }
1.288 brouard 6778: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6779: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6780: /* else */
6781: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6782: for(i=1;i<=nlstate;i++){
1.126 brouard 6783: gp[i] = prlim[i][i];
1.208 brouard 6784: mgp[theta][i] = prlim[i][i];
6785: }
1.126 brouard 6786: for(i=1; i<=npar; i++) /* Computes gradient */
6787: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6788: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6789: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6790: /* else */
6791: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6792: for(i=1;i<=nlstate;i++){
1.126 brouard 6793: gm[i] = prlim[i][i];
1.208 brouard 6794: mgm[theta][i] = prlim[i][i];
6795: }
1.126 brouard 6796: for(i=1;i<=nlstate;i++)
6797: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6798: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6799: } /* End theta */
6800:
6801: trgradg =matrix(1,nlstate,1,npar);
6802:
6803: for(j=1; j<=nlstate;j++)
6804: for(theta=1; theta <=npar; theta++)
6805: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6806: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6807: /* printf("\nmgm mgp %d ",(int)age); */
6808: /* for(j=1; j<=nlstate;j++){ */
6809: /* printf(" %d ",j); */
6810: /* for(theta=1; theta <=npar; theta++) */
6811: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6812: /* printf("\n "); */
6813: /* } */
6814: /* } */
6815: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6816: /* printf("\n gradg %d ",(int)age); */
6817: /* for(j=1; j<=nlstate;j++){ */
6818: /* printf("%d ",j); */
6819: /* for(theta=1; theta <=npar; theta++) */
6820: /* printf("%d %lf ",theta,gradg[theta][j]); */
6821: /* printf("\n "); */
6822: /* } */
6823: /* } */
1.126 brouard 6824:
6825: for(i=1;i<=nlstate;i++)
6826: varpl[i][(int)age] =0.;
1.209 brouard 6827: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6828: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6829: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6830: }else{
1.268 brouard 6831: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6832: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6833: }
1.126 brouard 6834: for(i=1;i<=nlstate;i++)
6835: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6836:
6837: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6838: if(nresult >=1)
6839: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6840: for(i=1; i<=nlstate;i++){
1.126 brouard 6841: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6842: /* for(j=1;j<=nlstate;j++) */
6843: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6844: }
1.126 brouard 6845: fprintf(ficresvpl,"\n");
6846: free_vector(gp,1,nlstate);
6847: free_vector(gm,1,nlstate);
1.208 brouard 6848: free_matrix(mgm,1,npar,1,nlstate);
6849: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6850: free_matrix(gradg,1,npar,1,nlstate);
6851: free_matrix(trgradg,1,nlstate,1,npar);
6852: } /* End age */
6853:
6854: free_vector(xp,1,npar);
6855: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6856: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6857:
6858: }
6859:
6860:
6861: /************ Variance of backprevalence limit ******************/
1.269 brouard 6862: 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 6863: {
6864: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6865: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6866:
6867: double **dnewmpar,**doldm;
6868: int i, j, nhstepm, hstepm;
6869: double *xp;
6870: double *gp, *gm;
6871: double **gradg, **trgradg;
6872: double **mgm, **mgp;
6873: double age,agelim;
6874: int theta;
6875:
6876: pstamp(ficresvbl);
6877: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6878: fprintf(ficresvbl,"# Age ");
6879: if(nresult >=1)
6880: fprintf(ficresvbl," Result# ");
6881: for(i=1; i<=nlstate;i++)
6882: fprintf(ficresvbl," %1d-%1d",i,i);
6883: fprintf(ficresvbl,"\n");
6884:
6885: xp=vector(1,npar);
6886: dnewmpar=matrix(1,nlstate,1,npar);
6887: doldm=matrix(1,nlstate,1,nlstate);
6888:
6889: hstepm=1*YEARM; /* Every year of age */
6890: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6891: agelim = AGEINF;
6892: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6893: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6894: if (stepm >= YEARM) hstepm=1;
6895: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6896: gradg=matrix(1,npar,1,nlstate);
6897: mgp=matrix(1,npar,1,nlstate);
6898: mgm=matrix(1,npar,1,nlstate);
6899: gp=vector(1,nlstate);
6900: gm=vector(1,nlstate);
6901:
6902: for(theta=1; theta <=npar; theta++){
6903: for(i=1; i<=npar; i++){ /* Computes gradient */
6904: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6905: }
6906: if(mobilavproj > 0 )
6907: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6908: else
6909: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6910: for(i=1;i<=nlstate;i++){
6911: gp[i] = bprlim[i][i];
6912: mgp[theta][i] = bprlim[i][i];
6913: }
6914: for(i=1; i<=npar; i++) /* Computes gradient */
6915: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6916: if(mobilavproj > 0 )
6917: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6918: else
6919: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6920: for(i=1;i<=nlstate;i++){
6921: gm[i] = bprlim[i][i];
6922: mgm[theta][i] = bprlim[i][i];
6923: }
6924: for(i=1;i<=nlstate;i++)
6925: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6926: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6927: } /* End theta */
6928:
6929: trgradg =matrix(1,nlstate,1,npar);
6930:
6931: for(j=1; j<=nlstate;j++)
6932: for(theta=1; theta <=npar; theta++)
6933: trgradg[j][theta]=gradg[theta][j];
6934: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6935: /* printf("\nmgm mgp %d ",(int)age); */
6936: /* for(j=1; j<=nlstate;j++){ */
6937: /* printf(" %d ",j); */
6938: /* for(theta=1; theta <=npar; theta++) */
6939: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6940: /* printf("\n "); */
6941: /* } */
6942: /* } */
6943: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6944: /* printf("\n gradg %d ",(int)age); */
6945: /* for(j=1; j<=nlstate;j++){ */
6946: /* printf("%d ",j); */
6947: /* for(theta=1; theta <=npar; theta++) */
6948: /* printf("%d %lf ",theta,gradg[theta][j]); */
6949: /* printf("\n "); */
6950: /* } */
6951: /* } */
6952:
6953: for(i=1;i<=nlstate;i++)
6954: varbpl[i][(int)age] =0.;
6955: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6956: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6957: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6958: }else{
6959: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6960: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6961: }
6962: for(i=1;i<=nlstate;i++)
6963: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6964:
6965: fprintf(ficresvbl,"%.0f ",age );
6966: if(nresult >=1)
6967: fprintf(ficresvbl,"%d ",nres );
6968: for(i=1; i<=nlstate;i++)
6969: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6970: fprintf(ficresvbl,"\n");
6971: free_vector(gp,1,nlstate);
6972: free_vector(gm,1,nlstate);
6973: free_matrix(mgm,1,npar,1,nlstate);
6974: free_matrix(mgp,1,npar,1,nlstate);
6975: free_matrix(gradg,1,npar,1,nlstate);
6976: free_matrix(trgradg,1,nlstate,1,npar);
6977: } /* End age */
6978:
6979: free_vector(xp,1,npar);
6980: free_matrix(doldm,1,nlstate,1,npar);
6981: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6982:
6983: }
6984:
6985: /************ Variance of one-step probabilities ******************/
6986: 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 6987: {
6988: int i, j=0, k1, l1, tj;
6989: int k2, l2, j1, z1;
6990: int k=0, l;
6991: int first=1, first1, first2;
1.326 brouard 6992: int nres=0; /* New */
1.222 brouard 6993: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6994: double **dnewm,**doldm;
6995: double *xp;
6996: double *gp, *gm;
6997: double **gradg, **trgradg;
6998: double **mu;
6999: double age, cov[NCOVMAX+1];
7000: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
7001: int theta;
7002: char fileresprob[FILENAMELENGTH];
7003: char fileresprobcov[FILENAMELENGTH];
7004: char fileresprobcor[FILENAMELENGTH];
7005: double ***varpij;
7006:
7007: strcpy(fileresprob,"PROB_");
7008: strcat(fileresprob,fileres);
7009: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
7010: printf("Problem with resultfile: %s\n", fileresprob);
7011: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
7012: }
7013: strcpy(fileresprobcov,"PROBCOV_");
7014: strcat(fileresprobcov,fileresu);
7015: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
7016: printf("Problem with resultfile: %s\n", fileresprobcov);
7017: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
7018: }
7019: strcpy(fileresprobcor,"PROBCOR_");
7020: strcat(fileresprobcor,fileresu);
7021: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
7022: printf("Problem with resultfile: %s\n", fileresprobcor);
7023: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
7024: }
7025: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7026: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
7027: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7028: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
7029: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7030: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
7031: pstamp(ficresprob);
7032: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
7033: fprintf(ficresprob,"# Age");
7034: pstamp(ficresprobcov);
7035: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
7036: fprintf(ficresprobcov,"# Age");
7037: pstamp(ficresprobcor);
7038: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
7039: fprintf(ficresprobcor,"# Age");
1.126 brouard 7040:
7041:
1.222 brouard 7042: for(i=1; i<=nlstate;i++)
7043: for(j=1; j<=(nlstate+ndeath);j++){
7044: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
7045: fprintf(ficresprobcov," p%1d-%1d ",i,j);
7046: fprintf(ficresprobcor," p%1d-%1d ",i,j);
7047: }
7048: /* fprintf(ficresprob,"\n");
7049: fprintf(ficresprobcov,"\n");
7050: fprintf(ficresprobcor,"\n");
7051: */
7052: xp=vector(1,npar);
7053: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7054: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7055: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
7056: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
7057: first=1;
7058: fprintf(ficgp,"\n# Routine varprob");
7059: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
7060: fprintf(fichtm,"\n");
7061:
1.288 brouard 7062: 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 7063: 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);
7064: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 7065: and drawn. It helps understanding how is the covariance between two incidences.\
7066: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 7067: 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 7068: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
7069: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
7070: standard deviations wide on each axis. <br>\
7071: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
7072: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
7073: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
7074:
1.222 brouard 7075: cov[1]=1;
7076: /* tj=cptcoveff; */
1.225 brouard 7077: tj = (int) pow(2,cptcoveff);
1.222 brouard 7078: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
7079: j1=0;
1.332 brouard 7080:
7081: for(nres=1;nres <=nresult; nres++){ /* For each resultline */
7082: for(j1=1; j1<=tj;j1++){ /* For any combination of dummy covariates, fixed and varying */
7083: printf("Varprob TKresult[nres]=%d j1=%d, nres=%d, cptcovn=%d, cptcoveff=%d tj=%d \n", TKresult[nres], j1, nres, cptcovn, cptcoveff, tj);
7084: if(tj != 1 && TKresult[nres]!= j1)
7085: continue;
7086:
7087: /* for(j1=1; j1<=tj;j1++){ /\* For each valid combination of covariates or only once*\/ */
7088: /* for(nres=1;nres <=1; nres++){ /\* For each resultline *\/ */
7089: /* /\* for(nres=1;nres <=nresult; nres++){ /\\* For each resultline *\\/ *\/ */
1.222 brouard 7090: if (cptcovn>0) {
7091: fprintf(ficresprob, "\n#********** Variable ");
1.332 brouard 7092: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.222 brouard 7093: fprintf(ficresprob, "**********\n#\n");
7094: fprintf(ficresprobcov, "\n#********** Variable ");
1.332 brouard 7095: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.222 brouard 7096: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 7097:
1.222 brouard 7098: fprintf(ficgp, "\n#********** Variable ");
1.332 brouard 7099: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.222 brouard 7100: fprintf(ficgp, "**********\n#\n");
1.220 brouard 7101:
7102:
1.222 brouard 7103: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.319 brouard 7104: /* for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]); */
1.332 brouard 7105: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtmcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.222 brouard 7106: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7107:
1.222 brouard 7108: fprintf(ficresprobcor, "\n#********** Variable ");
1.332 brouard 7109: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,TnsdVar[Tvaraff[z1]])]);
1.222 brouard 7110: fprintf(ficresprobcor, "**********\n#");
7111: if(invalidvarcomb[j1]){
7112: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
7113: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
7114: continue;
7115: }
7116: }
7117: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
7118: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
7119: gp=vector(1,(nlstate)*(nlstate+ndeath));
7120: gm=vector(1,(nlstate)*(nlstate+ndeath));
7121: for (age=bage; age<=fage; age ++){
7122: cov[2]=age;
7123: if(nagesqr==1)
7124: cov[3]= age*age;
1.326 brouard 7125: /* for (k=1; k<=cptcovn;k++) { */
7126: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)]; */
7127: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
7128: /* Here comes the value of the covariate 'j1' after renumbering k with single dummy covariates */
1.332 brouard 7129: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(j1,TnsdVar[TvarsD[k]])];
1.222 brouard 7130: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
7131: * 1 1 1 1 1
7132: * 2 2 1 1 1
7133: * 3 1 2 1 1
7134: */
7135: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
7136: }
1.319 brouard 7137: /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1, Tage[1]=2 */
7138: /* ) p nbcode[Tvar[Tage[k]]][(1 & (ij-1) >> (k-1))+1] */
7139: /*for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
1.326 brouard 7140: for (k=1; k<=cptcovage;k++){ /* For product with age */
7141: if(Dummy[Tage[k]]==2){ /* dummy with age */
1.332 brouard 7142: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(j1,TnsdVar[Tvar[Tage[k]]])]*cov[2];
1.326 brouard 7143: /* cov[++k1]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
7144: } else if(Dummy[Tage[k]]==3){ /* quantitative with age */
1.327 brouard 7145: 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]);
1.332 brouard 7146: /* cov[2+nagesqr+Tage[k]]=meanq[k]/idq[k]*cov[2];/\* Using the mean of quantitative variable Tvar[Tage[k]] /\* Tqresult[nres][k]; *\/ */
7147: /* exit(1); */
1.326 brouard 7148: /* cov[++k1]=Tqresult[nres][k]; */
7149: }
7150: /* cov[2+Tage[k]+nagesqr]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
7151: }
7152: for (k=1; k<=cptcovprod;k++){/* For product without age */
1.329 brouard 7153: if(Dummy[Tvard[k][1]]==0){
7154: if(Dummy[Tvard[k][2]]==0){
1.332 brouard 7155: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])];
1.326 brouard 7156: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
7157: }else{ /* Should we use the mean of the quantitative variables? */
1.332 brouard 7158: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(j1,TnsdVar[Tvard[k][1]])] * Tqresult[nres][resultmodel[nres][k]];
1.326 brouard 7159: /* cov[++k1]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k]; */
7160: }
7161: }else{
1.329 brouard 7162: if(Dummy[Tvard[k][2]]==0){
1.332 brouard 7163: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(j1,TnsdVar[Tvard[k][2]])] * Tqinvresult[nres][TnsdVar[Tvard[k][1]]];
1.326 brouard 7164: /* cov[++k1]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]]; */
7165: }else{
1.332 brouard 7166: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][TnsdVar[Tvard[k][1]]]* Tqinvresult[nres][TnsdVar[Tvard[k][2]]];
1.326 brouard 7167: /* cov[++k1]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]]; */
7168: }
7169: }
7170: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
7171: }
7172: /* For each age and combination of dummy covariates we slightly move the parameters of delti in order to get the gradient*/
1.222 brouard 7173: for(theta=1; theta <=npar; theta++){
7174: for(i=1; i<=npar; i++)
7175: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 7176:
1.222 brouard 7177: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 7178:
1.222 brouard 7179: k=0;
7180: for(i=1; i<= (nlstate); i++){
7181: for(j=1; j<=(nlstate+ndeath);j++){
7182: k=k+1;
7183: gp[k]=pmmij[i][j];
7184: }
7185: }
1.220 brouard 7186:
1.222 brouard 7187: for(i=1; i<=npar; i++)
7188: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 7189:
1.222 brouard 7190: pmij(pmmij,cov,ncovmodel,xp,nlstate);
7191: k=0;
7192: for(i=1; i<=(nlstate); i++){
7193: for(j=1; j<=(nlstate+ndeath);j++){
7194: k=k+1;
7195: gm[k]=pmmij[i][j];
7196: }
7197: }
1.220 brouard 7198:
1.222 brouard 7199: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
7200: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
7201: }
1.126 brouard 7202:
1.222 brouard 7203: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
7204: for(theta=1; theta <=npar; theta++)
7205: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 7206:
1.222 brouard 7207: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
7208: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 7209:
1.222 brouard 7210: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 7211:
1.222 brouard 7212: k=0;
7213: for(i=1; i<=(nlstate); i++){
7214: for(j=1; j<=(nlstate+ndeath);j++){
7215: k=k+1;
7216: mu[k][(int) age]=pmmij[i][j];
7217: }
7218: }
7219: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
7220: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
7221: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 7222:
1.222 brouard 7223: /*printf("\n%d ",(int)age);
7224: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7225: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7226: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
7227: }*/
1.220 brouard 7228:
1.222 brouard 7229: fprintf(ficresprob,"\n%d ",(int)age);
7230: fprintf(ficresprobcov,"\n%d ",(int)age);
7231: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 7232:
1.222 brouard 7233: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
7234: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
7235: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
7236: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
7237: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
7238: }
7239: i=0;
7240: for (k=1; k<=(nlstate);k++){
7241: for (l=1; l<=(nlstate+ndeath);l++){
7242: i++;
7243: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
7244: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
7245: for (j=1; j<=i;j++){
7246: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
7247: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
7248: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
7249: }
7250: }
7251: }/* end of loop for state */
7252: } /* end of loop for age */
7253: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
7254: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
7255: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7256: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
7257:
7258: /* Confidence intervalle of pij */
7259: /*
7260: fprintf(ficgp,"\nunset parametric;unset label");
7261: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
7262: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
7263: 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);
7264: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
7265: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
7266: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
7267: */
7268:
7269: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
7270: first1=1;first2=2;
7271: for (k2=1; k2<=(nlstate);k2++){
7272: for (l2=1; l2<=(nlstate+ndeath);l2++){
7273: if(l2==k2) continue;
7274: j=(k2-1)*(nlstate+ndeath)+l2;
7275: for (k1=1; k1<=(nlstate);k1++){
7276: for (l1=1; l1<=(nlstate+ndeath);l1++){
7277: if(l1==k1) continue;
7278: i=(k1-1)*(nlstate+ndeath)+l1;
7279: if(i<=j) continue;
7280: for (age=bage; age<=fage; age ++){
7281: if ((int)age %5==0){
7282: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
7283: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
7284: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
7285: mu1=mu[i][(int) age]/stepm*YEARM ;
7286: mu2=mu[j][(int) age]/stepm*YEARM;
7287: c12=cv12/sqrt(v1*v2);
7288: /* Computing eigen value of matrix of covariance */
7289: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7290: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
7291: if ((lc2 <0) || (lc1 <0) ){
7292: if(first2==1){
7293: first1=0;
7294: 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);
7295: }
7296: 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);
7297: /* lc1=fabs(lc1); */ /* If we want to have them positive */
7298: /* lc2=fabs(lc2); */
7299: }
1.220 brouard 7300:
1.222 brouard 7301: /* Eigen vectors */
1.280 brouard 7302: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
7303: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7304: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
7305: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
7306: }else
7307: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 7308: /*v21=sqrt(1.-v11*v11); *//* error */
7309: v21=(lc1-v1)/cv12*v11;
7310: v12=-v21;
7311: v22=v11;
7312: tnalp=v21/v11;
7313: if(first1==1){
7314: first1=0;
7315: 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);
7316: }
7317: 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);
7318: /*printf(fignu*/
7319: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
7320: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
7321: if(first==1){
7322: first=0;
7323: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
7324: fprintf(ficgp,"\nset parametric;unset label");
7325: 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);
7326: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 7327: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 7328: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 7329: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 7330: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
7331: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7332: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7333: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
7334: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7335: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7336: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7337: 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 7338: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
7339: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 7340: }else{
7341: first=0;
7342: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
7343: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
7344: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
7345: 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 7346: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
7347: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 7348: }/* if first */
7349: } /* age mod 5 */
7350: } /* end loop age */
7351: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
7352: first=1;
7353: } /*l12 */
7354: } /* k12 */
7355: } /*l1 */
7356: }/* k1 */
1.332 brouard 7357: } /* loop on combination of covariates j1 */
1.326 brouard 7358: } /* loop on nres */
1.222 brouard 7359: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
7360: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
7361: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
7362: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
7363: free_vector(xp,1,npar);
7364: fclose(ficresprob);
7365: fclose(ficresprobcov);
7366: fclose(ficresprobcor);
7367: fflush(ficgp);
7368: fflush(fichtmcov);
7369: }
1.126 brouard 7370:
7371:
7372: /******************* Printing html file ***********/
1.201 brouard 7373: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 7374: int lastpass, int stepm, int weightopt, char model[],\
7375: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 7376: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
7377: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
7378: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 7379: int jj1, k1, i1, cpt, k4, nres;
1.319 brouard 7380: /* In fact some results are already printed in fichtm which is open */
1.126 brouard 7381: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
7382: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
7383: </ul>");
1.319 brouard 7384: /* fprintf(fichtm,"<ul><li> model=1+age+%s\n \ */
7385: /* </ul>", model); */
1.214 brouard 7386: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
7387: 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",
7388: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
1.332 brouard 7389: fprintf(fichtm,"<li> - Observed prevalence (cross-sectional prevalence) in each state (during the period defined between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf): <a href=\"%s\">%s</a> (html file) ",
1.213 brouard 7390: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
7391: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 7392: fprintf(fichtm,"\
7393: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 7394: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 7395: fprintf(fichtm,"\
1.217 brouard 7396: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
7397: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
7398: fprintf(fichtm,"\
1.288 brouard 7399: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7400: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 7401: fprintf(fichtm,"\
1.288 brouard 7402: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 7403: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
7404: fprintf(fichtm,"\
1.211 brouard 7405: - (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 7406: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 7407: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 7408: if(prevfcast==1){
7409: fprintf(fichtm,"\
7410: - Prevalence projections by age and states: \
1.201 brouard 7411: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 7412: }
1.126 brouard 7413:
7414:
1.225 brouard 7415: m=pow(2,cptcoveff);
1.222 brouard 7416: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7417:
1.317 brouard 7418: fprintf(fichtm," \n<ul><li><b>Graphs (first order)</b></li><p>");
1.264 brouard 7419:
7420: jj1=0;
7421:
7422: fprintf(fichtm," \n<ul>");
7423: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7424: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7425: if(m != 1 && TKresult[nres]!= k1)
7426: continue;
7427: jj1++;
7428: if (cptcovn > 0) {
7429: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
7430: for (cpt=1; cpt<=cptcoveff;cpt++){
7431: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7432: }
7433: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7434: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7435: }
7436: fprintf(fichtm,"\">");
7437:
7438: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7439: fprintf(fichtm,"************ Results for covariates");
7440: for (cpt=1; cpt<=cptcoveff;cpt++){
7441: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7442: }
7443: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7444: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7445: }
7446: if(invalidvarcomb[k1]){
7447: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7448: continue;
7449: }
7450: fprintf(fichtm,"</a></li>");
7451: } /* cptcovn >0 */
7452: }
1.317 brouard 7453: fprintf(fichtm," \n</ul>");
1.264 brouard 7454:
1.222 brouard 7455: jj1=0;
1.237 brouard 7456:
7457: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7458: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7459: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7460: continue;
1.220 brouard 7461:
1.222 brouard 7462: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7463: jj1++;
7464: if (cptcovn > 0) {
1.264 brouard 7465: fprintf(fichtm,"\n<p><a name=\"rescov");
7466: for (cpt=1; cpt<=cptcoveff;cpt++){
7467: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7468: }
7469: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7470: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7471: }
7472: fprintf(fichtm,"\"</a>");
7473:
1.222 brouard 7474: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7475: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7476: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7477: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7478: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7479: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7480: }
1.237 brouard 7481: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7482: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7483: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7484: }
7485:
1.230 brouard 7486: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.321 brouard 7487: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.222 brouard 7488: if(invalidvarcomb[k1]){
7489: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7490: printf("\nCombination (%d) ignored because no cases \n",k1);
7491: continue;
7492: }
7493: }
7494: /* aij, bij */
1.259 brouard 7495: 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 7496: <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 7497: /* Pij */
1.241 brouard 7498: 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> \
7499: <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 7500: /* Quasi-incidences */
7501: 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 7502: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7503: 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 7504: 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> \
7505: <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 7506: /* Survival functions (period) in state j */
7507: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7508: fprintf(fichtm,"<br>\n- Survival functions in state %d. And probability to be observed in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, cpt, nlstate, subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
7509: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7510: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7511: }
7512: /* State specific survival functions (period) */
7513: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7514: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7515: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.329 brouard 7516: <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> ", cpt, nlstate, cpt, subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
7517: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"PIJ_"),subdirf2(optionfilefiname,"PIJ_"));
7518: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7519: }
1.288 brouard 7520: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7521: for(cpt=1; cpt<=nlstate;cpt++){
1.329 brouard 7522: fprintf(fichtm,"<br>\n- Convergence to period (stable) prevalence in state %d. Or probability for a person being in state (1 to %d) at different ages, to be in state %d some years after. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br>", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
7523: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"P_"),subdirf2(optionfilefiname,"P_"));
7524: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">" ,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7525: }
1.296 brouard 7526: if(prevbcast==1){
1.288 brouard 7527: /* Backward prevalence in each health state */
1.222 brouard 7528: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7529: 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 7530: <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 7531: }
1.217 brouard 7532: }
1.222 brouard 7533: if(prevfcast==1){
1.288 brouard 7534: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7535: for(cpt=1; cpt<=nlstate;cpt++){
1.314 brouard 7536: 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);
7537: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
7538: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
7539: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7540: }
7541: }
1.296 brouard 7542: if(prevbcast==1){
1.268 brouard 7543: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7544: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7545: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7546: 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 \
7547: 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 7548: 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);
7549: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
7550: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7551: }
7552: }
1.220 brouard 7553:
1.222 brouard 7554: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 brouard 7555: 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);
7556: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
7557: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7558: }
7559: /* } /\* end i1 *\/ */
7560: }/* End k1 */
7561: fprintf(fichtm,"</ul>");
1.126 brouard 7562:
1.222 brouard 7563: fprintf(fichtm,"\
1.126 brouard 7564: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7565: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7566: - 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 7567: But because parameters are usually highly correlated (a higher incidence of disability \
7568: and a higher incidence of recovery can give very close observed transition) it might \
7569: be very useful to look not only at linear confidence intervals estimated from the \
7570: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7571: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7572: covariance matrix of the one-step probabilities. \
7573: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7574:
1.222 brouard 7575: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7576: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7577: fprintf(fichtm,"\
1.126 brouard 7578: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7579: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7580:
1.222 brouard 7581: fprintf(fichtm,"\
1.126 brouard 7582: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7583: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7584: fprintf(fichtm,"\
1.126 brouard 7585: - 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): \
7586: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7587: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7588: fprintf(fichtm,"\
1.126 brouard 7589: - (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): \
7590: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7591: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7592: fprintf(fichtm,"\
1.288 brouard 7593: - 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 7594: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7595: fprintf(fichtm,"\
1.128 brouard 7596: - 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 7597: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7598: fprintf(fichtm,"\
1.288 brouard 7599: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7600: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7601:
7602: /* if(popforecast==1) fprintf(fichtm,"\n */
7603: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7604: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7605: /* <br>",fileres,fileres,fileres,fileres); */
7606: /* else */
7607: /* 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 7608: fflush(fichtm);
1.126 brouard 7609:
1.225 brouard 7610: m=pow(2,cptcoveff);
1.222 brouard 7611: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7612:
1.317 brouard 7613: fprintf(fichtm," <ul><li><b>Graphs (second order)</b></li><p>");
7614:
7615: jj1=0;
7616:
7617: fprintf(fichtm," \n<ul>");
7618: for(nres=1; nres <= nresult; nres++) /* For each resultline */
7619: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
7620: if(m != 1 && TKresult[nres]!= k1)
7621: continue;
7622: jj1++;
7623: if (cptcovn > 0) {
7624: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescovsecond");
7625: for (cpt=1; cpt<=cptcoveff;cpt++){
7626: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7627: }
7628: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7629: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7630: }
7631: fprintf(fichtm,"\">");
7632:
7633: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7634: fprintf(fichtm,"************ Results for covariates");
7635: for (cpt=1; cpt<=cptcoveff;cpt++){
7636: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7637: }
7638: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7639: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7640: }
7641: if(invalidvarcomb[k1]){
7642: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7643: continue;
7644: }
7645: fprintf(fichtm,"</a></li>");
7646: } /* cptcovn >0 */
7647: }
7648: fprintf(fichtm," \n</ul>");
7649:
1.222 brouard 7650: jj1=0;
1.237 brouard 7651:
1.241 brouard 7652: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7653: for(k1=1; k1<=m;k1++){
1.253 brouard 7654: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7655: continue;
1.222 brouard 7656: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7657: jj1++;
1.126 brouard 7658: if (cptcovn > 0) {
1.317 brouard 7659: fprintf(fichtm,"\n<p><a name=\"rescovsecond");
7660: for (cpt=1; cpt<=cptcoveff;cpt++){
7661: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7662: }
7663: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7664: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7665: }
7666: fprintf(fichtm,"\"</a>");
7667:
1.126 brouard 7668: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.317 brouard 7669: for (cpt=1; cpt<=cptcoveff;cpt++){ /**< cptcoveff number of variables */
1.237 brouard 7670: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
1.317 brouard 7671: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
1.237 brouard 7672: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
1.317 brouard 7673: }
1.237 brouard 7674: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7675: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7676: }
7677:
1.321 brouard 7678: fprintf(fichtm," (model=%s) ************\n<hr size=\"2\" color=\"#EC5E5E\">",model);
1.220 brouard 7679:
1.222 brouard 7680: if(invalidvarcomb[k1]){
7681: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7682: continue;
7683: }
1.126 brouard 7684: }
7685: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7686: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 brouard 7687: 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);
7688: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
7689: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7690: }
7691: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 brouard 7692: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7693: true period expectancies (those weighted with period prevalences are also\
7694: drawn in addition to the population based expectancies computed using\
1.314 brouard 7695: 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);
7696: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
7697: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7698: /* } /\* end i1 *\/ */
7699: }/* End k1 */
1.241 brouard 7700: }/* End nres */
1.222 brouard 7701: fprintf(fichtm,"</ul>");
7702: fflush(fichtm);
1.126 brouard 7703: }
7704:
7705: /******************* Gnuplot file **************/
1.296 brouard 7706: 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 7707:
7708: char dirfileres[132],optfileres[132];
1.264 brouard 7709: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7710: 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 7711: int lv=0, vlv=0, kl=0;
1.130 brouard 7712: int ng=0;
1.201 brouard 7713: int vpopbased;
1.223 brouard 7714: int ioffset; /* variable offset for columns */
1.270 brouard 7715: int iyearc=1; /* variable column for year of projection */
7716: int iagec=1; /* variable column for age of projection */
1.235 brouard 7717: int nres=0; /* Index of resultline */
1.266 brouard 7718: int istart=1; /* For starting graphs in projections */
1.219 brouard 7719:
1.126 brouard 7720: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7721: /* printf("Problem with file %s",optionfilegnuplot); */
7722: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7723: /* } */
7724:
7725: /*#ifdef windows */
7726: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7727: /*#endif */
1.225 brouard 7728: m=pow(2,cptcoveff);
1.126 brouard 7729:
1.274 brouard 7730: /* diagram of the model */
7731: fprintf(ficgp,"\n#Diagram of the model \n");
7732: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7733: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7734: 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);
7735:
7736: 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);
7737: fprintf(ficgp,"\n#show arrow\nunset label\n");
7738: 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);
7739: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7740: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7741: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7742: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7743:
1.202 brouard 7744: /* Contribution to likelihood */
7745: /* Plot the probability implied in the likelihood */
1.223 brouard 7746: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7747: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7748: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7749: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7750: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7751: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7752: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7753: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7754: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7755: 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));
7756: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7757: 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));
7758: for (i=1; i<= nlstate ; i ++) {
7759: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7760: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7761: 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);
7762: for (j=2; j<= nlstate+ndeath ; j ++) {
7763: 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);
7764: }
7765: fprintf(ficgp,";\nset out; unset ylabel;\n");
7766: }
7767: /* 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 */
7768: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7769: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7770: fprintf(ficgp,"\nset out;unset log\n");
7771: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7772:
1.126 brouard 7773: strcpy(dirfileres,optionfilefiname);
7774: strcpy(optfileres,"vpl");
1.223 brouard 7775: /* 1eme*/
1.238 brouard 7776: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7777: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7778: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7779: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7780: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7781: continue;
7782: /* We are interested in selected combination by the resultline */
1.246 brouard 7783: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7784: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7785: strcpy(gplotlabel,"(");
1.238 brouard 7786: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
1.332 brouard 7787: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the value of the covariate corresponding to k1 combination *\/ */
7788: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 7789: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7790: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7791: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7792: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7793: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7794: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7795: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7796: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7797: }
7798: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7799: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7800: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7801: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7802: }
7803: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7804: /* printf("\n#\n"); */
1.238 brouard 7805: fprintf(ficgp,"\n#\n");
7806: if(invalidvarcomb[k1]){
1.260 brouard 7807: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7808: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7809: continue;
7810: }
1.235 brouard 7811:
1.241 brouard 7812: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7813: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7814: /* 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 7815: fprintf(ficgp,"set title \"Alive state %d %s model=%s\" font \"Helvetica,12\"\n",cpt,gplotlabel,model);
1.260 brouard 7816: 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);
7817: /* 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); */
7818: /* k1-1 error should be nres-1*/
1.238 brouard 7819: for (i=1; i<= nlstate ; i ++) {
7820: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7821: else fprintf(ficgp," %%*lf (%%*lf)");
7822: }
1.288 brouard 7823: 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 7824: for (i=1; i<= nlstate ; i ++) {
7825: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7826: else fprintf(ficgp," %%*lf (%%*lf)");
7827: }
1.260 brouard 7828: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7829: for (i=1; i<= nlstate ; i ++) {
7830: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7831: else fprintf(ficgp," %%*lf (%%*lf)");
7832: }
1.265 brouard 7833: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7834:
7835: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7836: if(cptcoveff ==0){
1.271 brouard 7837: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7838: }else{
7839: kl=0;
7840: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 7841: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
7842: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.265 brouard 7843: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7844: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7845: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7846: vlv= nbcode[Tvaraff[k]][lv];
7847: kl++;
7848: /* 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 *\/ */
7849: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7850: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7851: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7852: if(k==cptcoveff){
7853: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7854: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7855: }else{
7856: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7857: kl++;
7858: }
7859: } /* end covariate */
7860: } /* end if no covariate */
7861:
1.296 brouard 7862: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7863: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7864: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7865: if(cptcoveff ==0){
1.245 brouard 7866: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7867: }else{
7868: kl=0;
7869: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
1.332 brouard 7870: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
7871: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.238 brouard 7872: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7873: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7874: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 7875: /* vlv= nbcode[Tvaraff[k]][lv]; */
7876: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 7877: kl++;
1.238 brouard 7878: /* 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 *\/ */
7879: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7880: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7881: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7882: if(k==cptcoveff){
1.245 brouard 7883: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7884: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7885: }else{
1.332 brouard 7886: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])]);
1.238 brouard 7887: kl++;
7888: }
7889: } /* end covariate */
7890: } /* end if no covariate */
1.296 brouard 7891: if(prevbcast == 1){
1.268 brouard 7892: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7893: /* k1-1 error should be nres-1*/
7894: for (i=1; i<= nlstate ; i ++) {
7895: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7896: else fprintf(ficgp," %%*lf (%%*lf)");
7897: }
1.271 brouard 7898: 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 7899: for (i=1; i<= nlstate ; i ++) {
7900: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7901: else fprintf(ficgp," %%*lf (%%*lf)");
7902: }
1.276 brouard 7903: 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 7904: for (i=1; i<= nlstate ; i ++) {
7905: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7906: else fprintf(ficgp," %%*lf (%%*lf)");
7907: }
1.274 brouard 7908: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7909: } /* end if backprojcast */
1.296 brouard 7910: } /* end if prevbcast */
1.276 brouard 7911: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7912: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7913: } /* nres */
1.201 brouard 7914: } /* k1 */
7915: } /* cpt */
1.235 brouard 7916:
7917:
1.126 brouard 7918: /*2 eme*/
1.238 brouard 7919: for (k1=1; k1<= m ; k1 ++){
7920: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7921: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7922: continue;
7923: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7924: strcpy(gplotlabel,"(");
1.238 brouard 7925: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 7926: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
7927: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.223 brouard 7928: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7929: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7930: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 7931: /* vlv= nbcode[Tvaraff[k]][lv]; */
7932: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.223 brouard 7933: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7934: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7935: }
1.237 brouard 7936: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7937: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7938: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7939: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7940: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7941: }
1.264 brouard 7942: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7943: fprintf(ficgp,"\n#\n");
1.223 brouard 7944: if(invalidvarcomb[k1]){
7945: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7946: continue;
7947: }
1.219 brouard 7948:
1.241 brouard 7949: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7950: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7951: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7952: if(vpopbased==0){
1.238 brouard 7953: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7954: }else
1.238 brouard 7955: fprintf(ficgp,"\nreplot ");
7956: for (i=1; i<= nlstate+1 ; i ++) {
7957: k=2*i;
1.261 brouard 7958: 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 7959: for (j=1; j<= nlstate+1 ; j ++) {
7960: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7961: else fprintf(ficgp," %%*lf (%%*lf)");
7962: }
7963: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7964: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7965: 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 7966: for (j=1; j<= nlstate+1 ; j ++) {
7967: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7968: else fprintf(ficgp," %%*lf (%%*lf)");
7969: }
7970: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7971: 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 7972: for (j=1; j<= nlstate+1 ; j ++) {
7973: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7974: else fprintf(ficgp," %%*lf (%%*lf)");
7975: }
7976: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7977: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7978: } /* state */
7979: } /* vpopbased */
1.264 brouard 7980: 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 7981: } /* end nres */
7982: } /* k1 end 2 eme*/
7983:
7984:
7985: /*3eme*/
7986: for (k1=1; k1<= m ; k1 ++){
7987: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7988: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7989: continue;
7990:
1.332 brouard 7991: for (cpt=1; cpt<= nlstate ; cpt ++) { /* Fragile no verification of covariate values */
1.261 brouard 7992: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7993: strcpy(gplotlabel,"(");
1.238 brouard 7994: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 7995: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
7996: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.238 brouard 7997: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7998: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7999: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8000: /* vlv= nbcode[Tvaraff[k]][lv]; */
8001: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.238 brouard 8002: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8003: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 8004: }
8005: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8006: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]);
8007: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][resultmodel[nres][k4]]);
1.238 brouard 8008: }
1.264 brouard 8009: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8010: fprintf(ficgp,"\n#\n");
8011: if(invalidvarcomb[k1]){
8012: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8013: continue;
8014: }
8015:
8016: /* k=2+nlstate*(2*cpt-2); */
8017: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 8018: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 8019: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 8020: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 8021: 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 8022: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8023: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8024: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
8025: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
8026: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
8027: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 8028:
1.238 brouard 8029: */
8030: for (i=1; i< nlstate ; i ++) {
1.261 brouard 8031: 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 8032: /* 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 8033:
1.238 brouard 8034: }
1.261 brouard 8035: 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 8036: }
1.264 brouard 8037: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 8038: } /* end nres */
8039: } /* end kl 3eme */
1.126 brouard 8040:
1.223 brouard 8041: /* 4eme */
1.201 brouard 8042: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 8043: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
8044: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8045: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 8046: continue;
1.238 brouard 8047: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 8048: strcpy(gplotlabel,"(");
1.238 brouard 8049: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
8050: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8051: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
8052: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
1.238 brouard 8053: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8054: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8055: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8056: /* vlv= nbcode[Tvaraff[k]][lv]; */
8057: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.238 brouard 8058: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8059: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 8060: }
8061: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8062: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8063: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.238 brouard 8064: }
1.264 brouard 8065: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8066: fprintf(ficgp,"\n#\n");
8067: if(invalidvarcomb[k1]){
8068: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8069: continue;
1.223 brouard 8070: }
1.238 brouard 8071:
1.241 brouard 8072: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 8073: 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 8074: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8075: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8076: k=3;
8077: for (i=1; i<= nlstate ; i ++){
8078: if(i==1){
8079: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8080: }else{
8081: fprintf(ficgp,", '' ");
8082: }
8083: l=(nlstate+ndeath)*(i-1)+1;
8084: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8085: for (j=2; j<= nlstate+ndeath ; j ++)
8086: fprintf(ficgp,"+$%d",k+l+j-1);
8087: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
8088: } /* nlstate */
1.264 brouard 8089: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8090: } /* end cpt state*/
8091: } /* end nres */
8092: } /* end covariate k1 */
8093:
1.220 brouard 8094: /* 5eme */
1.201 brouard 8095: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 8096: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
8097: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8098: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 8099: continue;
1.238 brouard 8100: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 8101: strcpy(gplotlabel,"(");
1.238 brouard 8102: 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);
8103: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8104: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
8105: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
1.238 brouard 8106: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8107: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8108: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8109: /* vlv= nbcode[Tvaraff[k]][lv]; */
8110: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.238 brouard 8111: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8112: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 8113: }
8114: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8115: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8116: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.238 brouard 8117: }
1.264 brouard 8118: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 8119: fprintf(ficgp,"\n#\n");
8120: if(invalidvarcomb[k1]){
8121: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8122: continue;
8123: }
1.227 brouard 8124:
1.241 brouard 8125: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 8126: 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 8127: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
8128: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8129: k=3;
8130: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8131: if(j==1)
8132: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8133: else
8134: fprintf(ficgp,", '' ");
8135: l=(nlstate+ndeath)*(cpt-1) +j;
8136: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
8137: /* for (i=2; i<= nlstate+ndeath ; i ++) */
8138: /* fprintf(ficgp,"+$%d",k+l+i-1); */
8139: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
8140: } /* nlstate */
8141: fprintf(ficgp,", '' ");
8142: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
8143: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
8144: l=(nlstate+ndeath)*(cpt-1) +j;
8145: if(j < nlstate)
8146: fprintf(ficgp,"$%d +",k+l);
8147: else
8148: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
8149: }
1.264 brouard 8150: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 8151: } /* end cpt state*/
8152: } /* end covariate */
8153: } /* end nres */
1.227 brouard 8154:
1.220 brouard 8155: /* 6eme */
1.202 brouard 8156: /* CV preval stable (period) for each covariate */
1.237 brouard 8157: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8158: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8159: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8160: continue;
1.255 brouard 8161: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 8162: strcpy(gplotlabel,"(");
1.288 brouard 8163: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 8164: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8165: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8166: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8167: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8168: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8169: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8170: /* vlv= nbcode[Tvaraff[k]][lv]; */
8171: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8172: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8173: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 8174: }
1.237 brouard 8175: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8176: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8177: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8178: }
1.264 brouard 8179: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 8180: fprintf(ficgp,"\n#\n");
1.223 brouard 8181: if(invalidvarcomb[k1]){
1.227 brouard 8182: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8183: continue;
1.223 brouard 8184: }
1.227 brouard 8185:
1.241 brouard 8186: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 8187: 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 8188: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8189: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 8190: k=3; /* Offset */
1.255 brouard 8191: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 8192: if(i==1)
8193: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
8194: else
8195: fprintf(ficgp,", '' ");
1.255 brouard 8196: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 8197: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
8198: for (j=2; j<= nlstate ; j ++)
8199: fprintf(ficgp,"+$%d",k+l+j-1);
8200: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 8201: } /* nlstate */
1.264 brouard 8202: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 8203: } /* end cpt state*/
8204: } /* end covariate */
1.227 brouard 8205:
8206:
1.220 brouard 8207: /* 7eme */
1.296 brouard 8208: if(prevbcast == 1){
1.288 brouard 8209: /* CV backward prevalence for each covariate */
1.237 brouard 8210: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8211: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8212: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8213: continue;
1.268 brouard 8214: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 8215: strcpy(gplotlabel,"(");
1.288 brouard 8216: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8217: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.332 brouard 8218: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate number corresponding to k1 combination *\/ */
8219: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8220: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8221: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 8222: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8223: /* vlv= nbcode[Tvaraff[k]][lv]; */
8224: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8225: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8226: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8227: }
1.237 brouard 8228: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8229: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8230: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8231: }
1.264 brouard 8232: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8233: fprintf(ficgp,"\n#\n");
8234: if(invalidvarcomb[k1]){
8235: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8236: continue;
8237: }
8238:
1.241 brouard 8239: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 8240: 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 8241: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 8242: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 8243: k=3; /* Offset */
1.268 brouard 8244: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 8245: if(i==1)
8246: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
8247: else
8248: fprintf(ficgp,", '' ");
8249: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 8250: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.324 brouard 8251: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
8252: /* 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 8253: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 8254: /* for (j=2; j<= nlstate ; j ++) */
8255: /* fprintf(ficgp,"+$%d",k+l+j-1); */
8256: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 8257: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 8258: } /* nlstate */
1.264 brouard 8259: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 8260: } /* end cpt state*/
8261: } /* end covariate */
1.296 brouard 8262: } /* End if prevbcast */
1.218 brouard 8263:
1.223 brouard 8264: /* 8eme */
1.218 brouard 8265: if(prevfcast==1){
1.288 brouard 8266: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 8267:
1.237 brouard 8268: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8269: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 8270: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8271: continue;
1.211 brouard 8272: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 8273: strcpy(gplotlabel,"(");
1.288 brouard 8274: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 8275: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
1.332 brouard 8276: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8277: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8278: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8279: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8280: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8281: /* vlv= nbcode[Tvaraff[k]][lv]; */
8282: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8283: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 8284: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 8285: }
1.237 brouard 8286: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8287: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8288: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8289: }
1.264 brouard 8290: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 8291: fprintf(ficgp,"\n#\n");
8292: if(invalidvarcomb[k1]){
8293: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8294: continue;
8295: }
8296:
8297: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 8298: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 8299: 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 8300: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 8301: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 8302:
8303: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8304: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8305: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8306: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 8307: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8308: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8309: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8310: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 8311: if(i==istart){
1.227 brouard 8312: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
8313: }else{
8314: fprintf(ficgp,",\\\n '' ");
8315: }
8316: if(cptcoveff ==0){ /* No covariate */
8317: ioffset=2; /* Age is in 2 */
8318: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8319: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8320: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8321: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8322: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 8323: if(i==nlstate+1){
1.270 brouard 8324: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 8325: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8326: fprintf(ficgp,",\\\n '' ");
8327: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8328: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 8329: offyear, \
1.268 brouard 8330: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 8331: }else
1.227 brouard 8332: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
8333: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8334: }else{ /* more than 2 covariates */
1.270 brouard 8335: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8336: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8337: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8338: iyearc=ioffset-1;
8339: iagec=ioffset;
1.227 brouard 8340: fprintf(ficgp," u %d:(",ioffset);
8341: kl=0;
8342: strcpy(gplotcondition,"(");
8343: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8344: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8345: lv=codtabm(k1,TnsdVar[Tvaraff[k]]);
1.227 brouard 8346: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8347: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8348: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8349: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8350: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.227 brouard 8351: kl++;
8352: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8353: kl++;
8354: if(k <cptcoveff && cptcoveff>1)
8355: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8356: }
8357: strcpy(gplotcondition+strlen(gplotcondition),")");
8358: /* 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 *\/ */
8359: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8360: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8361: /* '' 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*/
8362: if(i==nlstate+1){
1.270 brouard 8363: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
8364: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 8365: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8366: fprintf(ficgp," u %d:(",iagec);
8367: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
8368: iyearc, iagec, offyear, \
8369: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 8370: /* '' 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 8371: }else{
8372: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
8373: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
8374: }
8375: } /* end if covariate */
8376: } /* nlstate */
1.264 brouard 8377: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 8378: } /* end cpt state*/
8379: } /* end covariate */
8380: } /* End if prevfcast */
1.227 brouard 8381:
1.296 brouard 8382: if(prevbcast==1){
1.268 brouard 8383: /* Back projection from cross-sectional to stable (mixed) for each covariate */
8384:
8385: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
8386: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
8387: if(m != 1 && TKresult[nres]!= k1)
8388: continue;
8389: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
8390: strcpy(gplotlabel,"(");
8391: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
8392: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
1.332 brouard 8393: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8394: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.268 brouard 8395: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8396: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8397: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8398: /* vlv= nbcode[Tvaraff[k]][lv]; */
8399: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.268 brouard 8400: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8401: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8402: }
8403: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8404: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8405: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.268 brouard 8406: }
8407: strcpy(gplotlabel+strlen(gplotlabel),")");
8408: fprintf(ficgp,"\n#\n");
8409: if(invalidvarcomb[k1]){
8410: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
8411: continue;
8412: }
8413:
8414: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
8415: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
8416: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
8417: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
8418: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
8419:
8420: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
8421: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
8422: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
8423: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
8424: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8425: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8426: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8427: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8428: if(i==istart){
8429: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
8430: }else{
8431: fprintf(ficgp,",\\\n '' ");
8432: }
8433: if(cptcoveff ==0){ /* No covariate */
8434: ioffset=2; /* Age is in 2 */
8435: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8436: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8437: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
8438: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
8439: fprintf(ficgp," u %d:(", ioffset);
8440: if(i==nlstate+1){
1.270 brouard 8441: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 8442: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
8443: fprintf(ficgp,",\\\n '' ");
8444: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 8445: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 8446: offbyear, \
8447: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
8448: }else
8449: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
8450: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
8451: }else{ /* more than 2 covariates */
1.270 brouard 8452: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
8453: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
8454: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
8455: iyearc=ioffset-1;
8456: iagec=ioffset;
1.268 brouard 8457: fprintf(ficgp," u %d:(",ioffset);
8458: kl=0;
8459: strcpy(gplotcondition,"(");
8460: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
1.332 brouard 8461: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to combination k1 and covariate k *\/ */
8462: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.268 brouard 8463: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8464: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8465: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8466: /* vlv= nbcode[Tvaraff[k]][lv]; /\* Value of the modality of Tvaraff[k] *\/ */
8467: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.268 brouard 8468: kl++;
8469: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
8470: kl++;
8471: if(k <cptcoveff && cptcoveff>1)
8472: sprintf(gplotcondition+strlen(gplotcondition)," && ");
8473: }
8474: strcpy(gplotcondition+strlen(gplotcondition),")");
8475: /* 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 *\/ */
8476: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
8477: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
8478: /* '' 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*/
8479: if(i==nlstate+1){
1.270 brouard 8480: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
8481: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 8482: fprintf(ficgp,",\\\n '' ");
1.270 brouard 8483: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 8484: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 8485: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
8486: iyearc,iagec,offbyear, \
8487: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 8488: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
8489: }else{
8490: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
8491: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
8492: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
8493: }
8494: } /* end if covariate */
8495: } /* nlstate */
8496: fprintf(ficgp,"\nset out; unset label;\n");
8497: } /* end cpt state*/
8498: } /* end covariate */
1.296 brouard 8499: } /* End if prevbcast */
1.268 brouard 8500:
1.227 brouard 8501:
1.238 brouard 8502: /* 9eme writing MLE parameters */
8503: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 8504: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 8505: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 8506: for(k=1; k <=(nlstate+ndeath); k++){
8507: if (k != i) {
1.227 brouard 8508: fprintf(ficgp,"# current state %d\n",k);
8509: for(j=1; j <=ncovmodel; j++){
8510: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
8511: jk++;
8512: }
8513: fprintf(ficgp,"\n");
1.126 brouard 8514: }
8515: }
1.223 brouard 8516: }
1.187 brouard 8517: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8518:
1.145 brouard 8519: /*goto avoid;*/
1.238 brouard 8520: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8521: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8522: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8523: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8524: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8525: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8526: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8527: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8528: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8529: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8530: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8531: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8532: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8533: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8534: fprintf(ficgp,"#\n");
1.223 brouard 8535: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8536: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8537: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8538: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8539: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8540: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8541: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8542: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8543: continue;
1.264 brouard 8544: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8545: strcpy(gplotlabel,"(");
1.276 brouard 8546: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8547: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
1.332 brouard 8548: /* lv= decodtabm(k1,k,cptcoveff); /\* Should be the covariate value corresponding to k1 combination and kth covariate *\/ */
8549: lv= codtabm(k1,TnsdVar[Tvaraff[k]]); /* Should be the covariate value corresponding to combination k1 and covariate k */
1.264 brouard 8550: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8551: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8552: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.332 brouard 8553: /* vlv= nbcode[Tvaraff[k]][lv]; */
8554: vlv= nbcode[Tvaraff[k]][codtabm(k1,TnsdVar[Tvaraff[k]])];
1.264 brouard 8555: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8556: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8557: }
1.237 brouard 8558: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.332 brouard 8559: fprintf(ficgp," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
8560: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][resultmodel[nres][k4]],Tqresult[nres][resultmodel[nres][k4]]);
1.237 brouard 8561: }
1.264 brouard 8562: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8563: fprintf(ficgp,"\n#\n");
1.264 brouard 8564: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8565: fprintf(ficgp,"\nset key outside ");
8566: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8567: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8568: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8569: if (ng==1){
8570: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8571: fprintf(ficgp,"\nunset log y");
8572: }else if (ng==2){
8573: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8574: fprintf(ficgp,"\nset log y");
8575: }else if (ng==3){
8576: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8577: fprintf(ficgp,"\nset log y");
8578: }else
8579: fprintf(ficgp,"\nunset title ");
8580: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8581: i=1;
8582: for(k2=1; k2<=nlstate; k2++) {
8583: k3=i;
8584: for(k=1; k<=(nlstate+ndeath); k++) {
8585: if (k != k2){
8586: switch( ng) {
8587: case 1:
8588: if(nagesqr==0)
8589: fprintf(ficgp," p%d+p%d*x",i,i+1);
8590: else /* nagesqr =1 */
8591: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8592: break;
8593: case 2: /* ng=2 */
8594: if(nagesqr==0)
8595: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8596: else /* nagesqr =1 */
8597: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8598: break;
8599: case 3:
8600: if(nagesqr==0)
8601: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8602: else /* nagesqr =1 */
8603: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8604: break;
8605: }
8606: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8607: ijp=1; /* product no age */
8608: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8609: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8610: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.329 brouard 8611: switch(Typevar[j]){
8612: case 1:
8613: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8614: if(j==Tage[ij]) { /* Product by age To be looked at!!*//* Bug valgrind */
8615: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8616: if(DummyV[j]==0){/* Bug valgrind */
8617: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8618: }else{ /* quantitative */
8619: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8620: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8621: }
8622: ij++;
1.268 brouard 8623: }
1.237 brouard 8624: }
1.329 brouard 8625: }
8626: break;
8627: case 2:
8628: if(cptcovprod >0){
8629: if(j==Tprod[ijp]) { /* */
8630: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8631: if(ijp <=cptcovprod) { /* Product */
8632: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8633: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8634: /* 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)]); */
8635: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8636: }else{ /* Vn is dummy and Vm is quanti */
8637: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8638: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8639: }
8640: }else{ /* Vn*Vm Vn is quanti */
8641: if(DummyV[Tvard[ijp][2]]==0){
8642: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8643: }else{ /* Both quanti */
8644: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8645: }
1.268 brouard 8646: }
1.329 brouard 8647: ijp++;
1.237 brouard 8648: }
1.329 brouard 8649: } /* end Tprod */
8650: }
8651: break;
8652: case 0:
8653: /* simple covariate */
1.264 brouard 8654: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8655: if(Dummy[j]==0){
8656: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8657: }else{ /* quantitative */
8658: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8659: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8660: }
1.329 brouard 8661: /* end simple */
8662: break;
8663: default:
8664: break;
8665: } /* end switch */
1.237 brouard 8666: } /* end j */
1.329 brouard 8667: }else{ /* k=k2 */
8668: if(ng !=1 ){ /* For logit formula of log p11 is more difficult to get */
8669: fprintf(ficgp," (1.");i=i-ncovmodel;
8670: }else
8671: i=i-ncovmodel;
1.223 brouard 8672: }
1.227 brouard 8673:
1.223 brouard 8674: if(ng != 1){
8675: fprintf(ficgp,")/(1");
1.227 brouard 8676:
1.264 brouard 8677: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8678: if(nagesqr==0)
1.264 brouard 8679: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8680: else /* nagesqr =1 */
1.264 brouard 8681: 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 8682:
1.223 brouard 8683: ij=1;
1.329 brouard 8684: ijp=1;
8685: /* for(j=3; j <=ncovmodel-nagesqr; j++){ */
8686: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
8687: switch(Typevar[j]){
8688: case 1:
8689: if(cptcovage >0){
8690: if(j==Tage[ij]) { /* Bug valgrind */
8691: if(ij <=cptcovage) { /* Bug valgrind */
8692: if(DummyV[j]==0){/* Bug valgrind */
8693: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]); */
8694: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,nbcode[Tvar[j]][codtabm(k1,j)]); */
8695: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]);
8696: /* fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);; */
8697: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8698: }else{ /* quantitative */
8699: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8700: fprintf(ficgp,"+p%d*%f*x",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8701: /* fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* Tqinvresult in decoderesult *\/ */
8702: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8703: }
8704: ij++;
8705: }
8706: }
8707: }
8708: break;
8709: case 2:
8710: if(cptcovprod >0){
8711: if(j==Tprod[ijp]) { /* */
8712: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8713: if(ijp <=cptcovprod) { /* Product */
8714: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8715: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8716: /* 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)]); */
8717: fprintf(ficgp,"+p%d*%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8718: /* fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]); */
8719: }else{ /* Vn is dummy and Vm is quanti */
8720: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8721: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8722: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8723: }
8724: }else{ /* Vn*Vm Vn is quanti */
8725: if(DummyV[Tvard[ijp][2]]==0){
8726: fprintf(ficgp,"+p%d*%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8727: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]); */
8728: }else{ /* Both quanti */
8729: fprintf(ficgp,"+p%d*%f*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8730: /* fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]); */
8731: }
8732: }
8733: ijp++;
8734: }
8735: } /* end Tprod */
8736: } /* end if */
8737: break;
8738: case 0:
8739: /* simple covariate */
8740: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
8741: if(Dummy[j]==0){
8742: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8743: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tinvresult[nres][Tvar[j]]); /* */
8744: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /\* *\/ */
8745: }else{ /* quantitative */
8746: fprintf(ficgp,"+p%d*%f",k3+(cpt-1)*ncovmodel+1+j+nagesqr,Tqinvresult[nres][Tvar[j]]); /* */
8747: /* fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /\* *\/ */
8748: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8749: }
8750: /* end simple */
8751: /* fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/\* Valgrind bug nbcode *\/ */
8752: break;
8753: default:
8754: break;
8755: } /* end switch */
1.223 brouard 8756: }
8757: fprintf(ficgp,")");
8758: }
8759: fprintf(ficgp,")");
8760: if(ng ==2)
1.276 brouard 8761: 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 8762: else /* ng= 3 */
1.276 brouard 8763: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"i%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.329 brouard 8764: }else{ /* end ng <> 1 */
1.223 brouard 8765: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8766: 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 8767: }
8768: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8769: fprintf(ficgp,",");
8770: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8771: fprintf(ficgp,",");
8772: i=i+ncovmodel;
8773: } /* end k */
8774: } /* end k2 */
1.276 brouard 8775: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8776: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8777: } /* end k1 */
1.223 brouard 8778: } /* end ng */
8779: /* avoid: */
8780: fflush(ficgp);
1.126 brouard 8781: } /* end gnuplot */
8782:
8783:
8784: /*************** Moving average **************/
1.219 brouard 8785: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8786: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8787:
1.222 brouard 8788: int i, cpt, cptcod;
8789: int modcovmax =1;
8790: int mobilavrange, mob;
8791: int iage=0;
1.288 brouard 8792: int firstA1=0, firstA2=0;
1.222 brouard 8793:
1.266 brouard 8794: double sum=0., sumr=0.;
1.222 brouard 8795: double age;
1.266 brouard 8796: double *sumnewp, *sumnewm, *sumnewmr;
8797: double *agemingood, *agemaxgood;
8798: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8799:
8800:
1.278 brouard 8801: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8802: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8803:
8804: sumnewp = vector(1,ncovcombmax);
8805: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8806: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8807: agemingood = vector(1,ncovcombmax);
1.266 brouard 8808: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8809: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8810: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8811:
8812: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8813: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8814: sumnewp[cptcod]=0.;
1.266 brouard 8815: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8816: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8817: }
8818: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8819:
1.266 brouard 8820: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8821: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8822: else mobilavrange=mobilav;
8823: for (age=bage; age<=fage; age++)
8824: for (i=1; i<=nlstate;i++)
8825: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8826: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8827: /* We keep the original values on the extreme ages bage, fage and for
8828: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8829: we use a 5 terms etc. until the borders are no more concerned.
8830: */
8831: for (mob=3;mob <=mobilavrange;mob=mob+2){
8832: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8833: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8834: sumnewm[cptcod]=0.;
8835: for (i=1; i<=nlstate;i++){
1.222 brouard 8836: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8837: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8838: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8839: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8840: }
8841: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8842: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8843: } /* end i */
8844: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8845: } /* end cptcod */
1.222 brouard 8846: }/* end age */
8847: }/* end mob */
1.266 brouard 8848: }else{
8849: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8850: return -1;
1.266 brouard 8851: }
8852:
8853: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8854: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8855: if(invalidvarcomb[cptcod]){
8856: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8857: continue;
8858: }
1.219 brouard 8859:
1.266 brouard 8860: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8861: sumnewm[cptcod]=0.;
8862: sumnewmr[cptcod]=0.;
8863: for (i=1; i<=nlstate;i++){
8864: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8865: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8866: }
8867: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8868: agemingoodr[cptcod]=age;
8869: }
8870: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8871: agemingood[cptcod]=age;
8872: }
8873: } /* age */
8874: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8875: sumnewm[cptcod]=0.;
1.266 brouard 8876: sumnewmr[cptcod]=0.;
1.222 brouard 8877: for (i=1; i<=nlstate;i++){
8878: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8879: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8880: }
8881: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8882: agemaxgoodr[cptcod]=age;
1.222 brouard 8883: }
8884: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8885: agemaxgood[cptcod]=age;
8886: }
8887: } /* age */
8888: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8889: /* but they will change */
1.288 brouard 8890: firstA1=0;firstA2=0;
1.266 brouard 8891: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8892: sumnewm[cptcod]=0.;
8893: sumnewmr[cptcod]=0.;
8894: for (i=1; i<=nlstate;i++){
8895: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8896: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8897: }
8898: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8899: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8900: agemaxgoodr[cptcod]=age; /* age min */
8901: for (i=1; i<=nlstate;i++)
8902: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8903: }else{ /* bad we change the value with the values of good ages */
8904: for (i=1; i<=nlstate;i++){
8905: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8906: } /* i */
8907: } /* end bad */
8908: }else{
8909: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8910: agemaxgood[cptcod]=age;
8911: }else{ /* bad we change the value with the values of good ages */
8912: for (i=1; i<=nlstate;i++){
8913: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8914: } /* i */
8915: } /* end bad */
8916: }/* end else */
8917: sum=0.;sumr=0.;
8918: for (i=1; i<=nlstate;i++){
8919: sum+=mobaverage[(int)age][i][cptcod];
8920: sumr+=probs[(int)age][i][cptcod];
8921: }
8922: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8923: if(!firstA1){
8924: firstA1=1;
8925: 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);
8926: }
8927: 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 8928: } /* end bad */
8929: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8930: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8931: if(!firstA2){
8932: firstA2=1;
8933: 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);
8934: }
8935: 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 8936: } /* end bad */
8937: }/* age */
1.266 brouard 8938:
8939: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8940: sumnewm[cptcod]=0.;
1.266 brouard 8941: sumnewmr[cptcod]=0.;
1.222 brouard 8942: for (i=1; i<=nlstate;i++){
8943: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8944: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8945: }
8946: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8947: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8948: agemingoodr[cptcod]=age;
8949: for (i=1; i<=nlstate;i++)
8950: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8951: }else{ /* bad we change the value with the values of good ages */
8952: for (i=1; i<=nlstate;i++){
8953: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8954: } /* i */
8955: } /* end bad */
8956: }else{
8957: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8958: agemingood[cptcod]=age;
8959: }else{ /* bad */
8960: for (i=1; i<=nlstate;i++){
8961: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8962: } /* i */
8963: } /* end bad */
8964: }/* end else */
8965: sum=0.;sumr=0.;
8966: for (i=1; i<=nlstate;i++){
8967: sum+=mobaverage[(int)age][i][cptcod];
8968: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8969: }
1.266 brouard 8970: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8971: 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 8972: } /* end bad */
8973: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8974: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8975: 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 8976: } /* end bad */
8977: }/* age */
1.266 brouard 8978:
1.222 brouard 8979:
8980: for (age=bage; age<=fage; age++){
1.235 brouard 8981: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8982: sumnewp[cptcod]=0.;
8983: sumnewm[cptcod]=0.;
8984: for (i=1; i<=nlstate;i++){
8985: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8986: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8987: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8988: }
8989: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8990: }
8991: /* printf("\n"); */
8992: /* } */
1.266 brouard 8993:
1.222 brouard 8994: /* brutal averaging */
1.266 brouard 8995: /* for (i=1; i<=nlstate;i++){ */
8996: /* for (age=1; age<=bage; age++){ */
8997: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8998: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8999: /* } */
9000: /* for (age=fage; age<=AGESUP; age++){ */
9001: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
9002: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
9003: /* } */
9004: /* } /\* end i status *\/ */
9005: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
9006: /* for (age=1; age<=AGESUP; age++){ */
9007: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
9008: /* mobaverage[(int)age][i][cptcod]=0.; */
9009: /* } */
9010: /* } */
1.222 brouard 9011: }/* end cptcod */
1.266 brouard 9012: free_vector(agemaxgoodr,1, ncovcombmax);
9013: free_vector(agemaxgood,1, ncovcombmax);
9014: free_vector(agemingood,1, ncovcombmax);
9015: free_vector(agemingoodr,1, ncovcombmax);
9016: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 9017: free_vector(sumnewm,1, ncovcombmax);
9018: free_vector(sumnewp,1, ncovcombmax);
9019: return 0;
9020: }/* End movingaverage */
1.218 brouard 9021:
1.126 brouard 9022:
1.296 brouard 9023:
1.126 brouard 9024: /************** Forecasting ******************/
1.296 brouard 9025: /* 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)*/
9026: 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){
9027: /* dateintemean, mean date of interviews
9028: dateprojd, year, month, day of starting projection
9029: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 9030: agemin, agemax range of age
9031: dateprev1 dateprev2 range of dates during which prevalence is computed
9032: */
1.296 brouard 9033: /* double anprojd, mprojd, jprojd; */
9034: /* double anprojf, mprojf, jprojf; */
1.267 brouard 9035: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 9036: double agec; /* generic age */
1.296 brouard 9037: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 9038: double *popeffectif,*popcount;
9039: double ***p3mat;
1.218 brouard 9040: /* double ***mobaverage; */
1.126 brouard 9041: char fileresf[FILENAMELENGTH];
9042:
9043: agelim=AGESUP;
1.211 brouard 9044: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9045: in each health status at the date of interview (if between dateprev1 and dateprev2).
9046: We still use firstpass and lastpass as another selection.
9047: */
1.214 brouard 9048: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9049: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 9050:
1.201 brouard 9051: strcpy(fileresf,"F_");
9052: strcat(fileresf,fileresu);
1.126 brouard 9053: if((ficresf=fopen(fileresf,"w"))==NULL) {
9054: printf("Problem with forecast resultfile: %s\n", fileresf);
9055: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
9056: }
1.235 brouard 9057: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
9058: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 9059:
1.225 brouard 9060: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 9061:
9062:
9063: stepsize=(int) (stepm+YEARM-1)/YEARM;
9064: if (stepm<=12) stepsize=1;
9065: if(estepm < stepm){
9066: printf ("Problem %d lower than %d\n",estepm, stepm);
9067: }
1.270 brouard 9068: else{
9069: hstepm=estepm;
9070: }
9071: if(estepm > stepm){ /* Yes every two year */
9072: stepsize=2;
9073: }
1.296 brouard 9074: hstepm=hstepm/stepm;
1.126 brouard 9075:
1.296 brouard 9076:
9077: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9078: /* fractional in yp1 *\/ */
9079: /* aintmean=yp; */
9080: /* yp2=modf((yp1*12),&yp); */
9081: /* mintmean=yp; */
9082: /* yp1=modf((yp2*30.5),&yp); */
9083: /* jintmean=yp; */
9084: /* if(jintmean==0) jintmean=1; */
9085: /* if(mintmean==0) mintmean=1; */
1.126 brouard 9086:
1.296 brouard 9087:
9088: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
9089: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
9090: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 9091: i1=pow(2,cptcoveff);
1.126 brouard 9092: if (cptcovn < 1){i1=1;}
9093:
1.296 brouard 9094: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 9095:
9096: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 9097:
1.126 brouard 9098: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 9099: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9100: for(k=1; k<=i1;k++){ /* We want to find the combination k corresponding to the values of the dummies given in this resut line (to be cleaned one day) */
1.253 brouard 9101: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 9102: continue;
1.227 brouard 9103: if(invalidvarcomb[k]){
9104: printf("\nCombination (%d) projection ignored because no cases \n",k);
9105: continue;
9106: }
9107: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
9108: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9109: /* fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); */
9110: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 9111: }
1.235 brouard 9112: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 9113: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 9114: }
1.227 brouard 9115: fprintf(ficresf," yearproj age");
9116: for(j=1; j<=nlstate+ndeath;j++){
9117: for(i=1; i<=nlstate;i++)
9118: fprintf(ficresf," p%d%d",i,j);
9119: fprintf(ficresf," wp.%d",j);
9120: }
1.296 brouard 9121: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 9122: fprintf(ficresf,"\n");
1.296 brouard 9123: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 9124: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
9125: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 9126: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
9127: nhstepm = nhstepm/hstepm;
9128: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9129: oldm=oldms;savm=savms;
1.268 brouard 9130: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 9131: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 9132: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 9133: for (h=0; h<=nhstepm; h++){
9134: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 9135: break;
9136: }
9137: }
9138: fprintf(ficresf,"\n");
9139: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9140: /* fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Tvaraff not correct *\/ */
9141: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* TnsdVar[Tvaraff] correct */
1.296 brouard 9142: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 9143:
9144: for(j=1; j<=nlstate+ndeath;j++) {
9145: ppij=0.;
9146: for(i=1; i<=nlstate;i++) {
1.278 brouard 9147: if (mobilav>=1)
9148: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
9149: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
9150: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
9151: }
1.268 brouard 9152: fprintf(ficresf," %.3f", p3mat[i][j][h]);
9153: } /* end i */
9154: fprintf(ficresf," %.3f", ppij);
9155: }/* end j */
1.227 brouard 9156: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9157: } /* end agec */
1.266 brouard 9158: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
9159: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 9160: } /* end yearp */
9161: } /* end k */
1.219 brouard 9162:
1.126 brouard 9163: fclose(ficresf);
1.215 brouard 9164: printf("End of Computing forecasting \n");
9165: fprintf(ficlog,"End of Computing forecasting\n");
9166:
1.126 brouard 9167: }
9168:
1.269 brouard 9169: /************** Back Forecasting ******************/
1.296 brouard 9170: /* 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){ */
9171: 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){
9172: /* back1, year, month, day of starting backprojection
1.267 brouard 9173: agemin, agemax range of age
9174: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 9175: anback2 year of end of backprojection (same day and month as back1).
9176: prevacurrent and prev are prevalences.
1.267 brouard 9177: */
9178: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
9179: double agec; /* generic age */
1.302 brouard 9180: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 9181: double *popeffectif,*popcount;
9182: double ***p3mat;
9183: /* double ***mobaverage; */
9184: char fileresfb[FILENAMELENGTH];
9185:
1.268 brouard 9186: agelim=AGEINF;
1.267 brouard 9187: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
9188: in each health status at the date of interview (if between dateprev1 and dateprev2).
9189: We still use firstpass and lastpass as another selection.
9190: */
9191: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
9192: /* firstpass, lastpass, stepm, weightopt, model); */
9193:
9194: /*Do we need to compute prevalence again?*/
9195:
9196: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
9197:
9198: strcpy(fileresfb,"FB_");
9199: strcat(fileresfb,fileresu);
9200: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
9201: printf("Problem with back forecast resultfile: %s\n", fileresfb);
9202: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
9203: }
9204: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9205: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
9206:
9207: if (cptcoveff==0) ncodemax[cptcoveff]=1;
9208:
9209:
9210: stepsize=(int) (stepm+YEARM-1)/YEARM;
9211: if (stepm<=12) stepsize=1;
9212: if(estepm < stepm){
9213: printf ("Problem %d lower than %d\n",estepm, stepm);
9214: }
1.270 brouard 9215: else{
9216: hstepm=estepm;
9217: }
9218: if(estepm >= stepm){ /* Yes every two year */
9219: stepsize=2;
9220: }
1.267 brouard 9221:
9222: hstepm=hstepm/stepm;
1.296 brouard 9223: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
9224: /* fractional in yp1 *\/ */
9225: /* aintmean=yp; */
9226: /* yp2=modf((yp1*12),&yp); */
9227: /* mintmean=yp; */
9228: /* yp1=modf((yp2*30.5),&yp); */
9229: /* jintmean=yp; */
9230: /* if(jintmean==0) jintmean=1; */
9231: /* if(mintmean==0) jintmean=1; */
1.267 brouard 9232:
9233: i1=pow(2,cptcoveff);
9234: if (cptcovn < 1){i1=1;}
9235:
1.296 brouard 9236: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
9237: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 9238:
9239: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
9240:
9241: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9242: for(k=1; k<=i1;k++){
9243: if(i1 != 1 && TKresult[nres]!= k)
9244: continue;
9245: if(invalidvarcomb[k]){
9246: printf("\nCombination (%d) projection ignored because no cases \n",k);
9247: continue;
9248: }
1.268 brouard 9249: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 9250: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9251: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.267 brouard 9252: }
9253: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
9254: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
9255: }
9256: fprintf(ficresfb," yearbproj age");
9257: for(j=1; j<=nlstate+ndeath;j++){
9258: for(i=1; i<=nlstate;i++)
1.268 brouard 9259: fprintf(ficresfb," b%d%d",i,j);
9260: fprintf(ficresfb," b.%d",j);
1.267 brouard 9261: }
1.296 brouard 9262: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 9263: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
9264: fprintf(ficresfb,"\n");
1.296 brouard 9265: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 9266: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 9267: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
9268: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 9269: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 9270: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 9271: nhstepm = nhstepm/hstepm;
9272: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9273: oldm=oldms;savm=savms;
1.268 brouard 9274: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 9275: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 9276: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 9277: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
9278: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
9279: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 9280: for (h=0; h<=nhstepm; h++){
1.268 brouard 9281: if (h*hstepm/YEARM*stepm ==-yearp) {
9282: break;
9283: }
9284: }
9285: fprintf(ficresfb,"\n");
9286: for(j=1;j<=cptcoveff;j++)
1.332 brouard 9287: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.296 brouard 9288: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 9289: for(i=1; i<=nlstate+ndeath;i++) {
9290: ppij=0.;ppi=0.;
9291: for(j=1; j<=nlstate;j++) {
9292: /* if (mobilav==1) */
1.269 brouard 9293: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
9294: ppi=ppi+prevacurrent[(int)agec][j][k];
9295: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
9296: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 9297: /* else { */
9298: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
9299: /* } */
1.268 brouard 9300: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
9301: } /* end j */
9302: if(ppi <0.99){
9303: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9304: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
9305: }
9306: fprintf(ficresfb," %.3f", ppij);
9307: }/* end j */
1.267 brouard 9308: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
9309: } /* end agec */
9310: } /* end yearp */
9311: } /* end k */
1.217 brouard 9312:
1.267 brouard 9313: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 9314:
1.267 brouard 9315: fclose(ficresfb);
9316: printf("End of Computing Back forecasting \n");
9317: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 9318:
1.267 brouard 9319: }
1.217 brouard 9320:
1.269 brouard 9321: /* Variance of prevalence limit: varprlim */
9322: 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 9323: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 9324:
9325: char fileresvpl[FILENAMELENGTH];
9326: FILE *ficresvpl;
9327: double **oldm, **savm;
9328: double **varpl; /* Variances of prevalence limits by age */
9329: int i1, k, nres, j ;
9330:
9331: strcpy(fileresvpl,"VPL_");
9332: strcat(fileresvpl,fileresu);
9333: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 9334: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 9335: exit(0);
9336: }
1.288 brouard 9337: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
9338: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 9339:
9340: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
9341: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
9342:
9343: i1=pow(2,cptcoveff);
9344: if (cptcovn < 1){i1=1;}
9345:
9346: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.332 brouard 9347: for(k=1; k<=i1;k++){ /* We find the combination equivalent to result line values of dummies */
1.269 brouard 9348: if(i1 != 1 && TKresult[nres]!= k)
9349: continue;
9350: fprintf(ficresvpl,"\n#****** ");
9351: printf("\n#****** ");
9352: fprintf(ficlog,"\n#****** ");
9353: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9354: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9355: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9356: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.269 brouard 9357: }
9358: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 9359: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9360: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9361: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.269 brouard 9362: }
9363: fprintf(ficresvpl,"******\n");
9364: printf("******\n");
9365: fprintf(ficlog,"******\n");
9366:
9367: varpl=matrix(1,nlstate,(int) bage, (int) fage);
9368: oldm=oldms;savm=savms;
9369: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
9370: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
9371: /*}*/
9372: }
9373:
9374: fclose(ficresvpl);
1.288 brouard 9375: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
9376: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 9377:
9378: }
9379: /* Variance of back prevalence: varbprlim */
9380: 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){
9381: /*------- Variance of back (stable) prevalence------*/
9382:
9383: char fileresvbl[FILENAMELENGTH];
9384: FILE *ficresvbl;
9385:
9386: double **oldm, **savm;
9387: double **varbpl; /* Variances of back prevalence limits by age */
9388: int i1, k, nres, j ;
9389:
9390: strcpy(fileresvbl,"VBL_");
9391: strcat(fileresvbl,fileresu);
9392: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
9393: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
9394: exit(0);
9395: }
9396: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
9397: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
9398:
9399:
9400: i1=pow(2,cptcoveff);
9401: if (cptcovn < 1){i1=1;}
9402:
9403: for(nres=1; nres <= nresult; nres++) /* For each resultline */
9404: for(k=1; k<=i1;k++){
9405: if(i1 != 1 && TKresult[nres]!= k)
9406: continue;
9407: fprintf(ficresvbl,"\n#****** ");
9408: printf("\n#****** ");
9409: fprintf(ficlog,"\n#****** ");
9410: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 9411: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9412: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
9413: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.269 brouard 9414: }
9415: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 9416: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9417: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
9418: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.269 brouard 9419: }
9420: fprintf(ficresvbl,"******\n");
9421: printf("******\n");
9422: fprintf(ficlog,"******\n");
9423:
9424: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
9425: oldm=oldms;savm=savms;
9426:
9427: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
9428: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
9429: /*}*/
9430: }
9431:
9432: fclose(ficresvbl);
9433: printf("done variance-covariance of back prevalence\n");fflush(stdout);
9434: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
9435:
9436: } /* End of varbprlim */
9437:
1.126 brouard 9438: /************** Forecasting *****not tested NB*************/
1.227 brouard 9439: /* 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 9440:
1.227 brouard 9441: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
9442: /* int *popage; */
9443: /* double calagedatem, agelim, kk1, kk2; */
9444: /* double *popeffectif,*popcount; */
9445: /* double ***p3mat,***tabpop,***tabpopprev; */
9446: /* /\* double ***mobaverage; *\/ */
9447: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 9448:
1.227 brouard 9449: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9450: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9451: /* agelim=AGESUP; */
9452: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 9453:
1.227 brouard 9454: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 9455:
9456:
1.227 brouard 9457: /* strcpy(filerespop,"POP_"); */
9458: /* strcat(filerespop,fileresu); */
9459: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
9460: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
9461: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
9462: /* } */
9463: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
9464: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 9465:
1.227 brouard 9466: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 9467:
1.227 brouard 9468: /* /\* if (mobilav!=0) { *\/ */
9469: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
9470: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
9471: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9472: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
9473: /* /\* } *\/ */
9474: /* /\* } *\/ */
1.126 brouard 9475:
1.227 brouard 9476: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
9477: /* if (stepm<=12) stepsize=1; */
1.126 brouard 9478:
1.227 brouard 9479: /* agelim=AGESUP; */
1.126 brouard 9480:
1.227 brouard 9481: /* hstepm=1; */
9482: /* hstepm=hstepm/stepm; */
1.218 brouard 9483:
1.227 brouard 9484: /* if (popforecast==1) { */
9485: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
9486: /* printf("Problem with population file : %s\n",popfile);exit(0); */
9487: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
9488: /* } */
9489: /* popage=ivector(0,AGESUP); */
9490: /* popeffectif=vector(0,AGESUP); */
9491: /* popcount=vector(0,AGESUP); */
1.126 brouard 9492:
1.227 brouard 9493: /* i=1; */
9494: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 9495:
1.227 brouard 9496: /* imx=i; */
9497: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
9498: /* } */
1.218 brouard 9499:
1.227 brouard 9500: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
9501: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
9502: /* k=k+1; */
9503: /* fprintf(ficrespop,"\n#******"); */
9504: /* for(j=1;j<=cptcoveff;j++) { */
9505: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
9506: /* } */
9507: /* fprintf(ficrespop,"******\n"); */
9508: /* fprintf(ficrespop,"# Age"); */
9509: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
9510: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 9511:
1.227 brouard 9512: /* for (cpt=0; cpt<=0;cpt++) { */
9513: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 9514:
1.227 brouard 9515: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9516: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9517: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9518:
1.227 brouard 9519: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9520: /* oldm=oldms;savm=savms; */
9521: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 9522:
1.227 brouard 9523: /* for (h=0; h<=nhstepm; h++){ */
9524: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9525: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9526: /* } */
9527: /* for(j=1; j<=nlstate+ndeath;j++) { */
9528: /* kk1=0.;kk2=0; */
9529: /* for(i=1; i<=nlstate;i++) { */
9530: /* if (mobilav==1) */
9531: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
9532: /* else { */
9533: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
9534: /* } */
9535: /* } */
9536: /* if (h==(int)(calagedatem+12*cpt)){ */
9537: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
9538: /* /\*fprintf(ficrespop," %.3f", kk1); */
9539: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
9540: /* } */
9541: /* } */
9542: /* for(i=1; i<=nlstate;i++){ */
9543: /* kk1=0.; */
9544: /* for(j=1; j<=nlstate;j++){ */
9545: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
9546: /* } */
9547: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
9548: /* } */
1.218 brouard 9549:
1.227 brouard 9550: /* if (h==(int)(calagedatem+12*cpt)) */
9551: /* for(j=1; j<=nlstate;j++) */
9552: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
9553: /* } */
9554: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9555: /* } */
9556: /* } */
1.218 brouard 9557:
1.227 brouard 9558: /* /\******\/ */
1.218 brouard 9559:
1.227 brouard 9560: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
9561: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
9562: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
9563: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
9564: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 9565:
1.227 brouard 9566: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9567: /* oldm=oldms;savm=savms; */
9568: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
9569: /* for (h=0; h<=nhstepm; h++){ */
9570: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
9571: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
9572: /* } */
9573: /* for(j=1; j<=nlstate+ndeath;j++) { */
9574: /* kk1=0.;kk2=0; */
9575: /* for(i=1; i<=nlstate;i++) { */
9576: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
9577: /* } */
9578: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
9579: /* } */
9580: /* } */
9581: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
9582: /* } */
9583: /* } */
9584: /* } */
9585: /* } */
1.218 brouard 9586:
1.227 brouard 9587: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 9588:
1.227 brouard 9589: /* if (popforecast==1) { */
9590: /* free_ivector(popage,0,AGESUP); */
9591: /* free_vector(popeffectif,0,AGESUP); */
9592: /* free_vector(popcount,0,AGESUP); */
9593: /* } */
9594: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9595: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9596: /* fclose(ficrespop); */
9597: /* } /\* End of popforecast *\/ */
1.218 brouard 9598:
1.126 brouard 9599: int fileappend(FILE *fichier, char *optionfich)
9600: {
9601: if((fichier=fopen(optionfich,"a"))==NULL) {
9602: printf("Problem with file: %s\n", optionfich);
9603: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9604: return (0);
9605: }
9606: fflush(fichier);
9607: return (1);
9608: }
9609:
9610:
9611: /**************** function prwizard **********************/
9612: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9613: {
9614:
9615: /* Wizard to print covariance matrix template */
9616:
1.164 brouard 9617: char ca[32], cb[32];
9618: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9619: int numlinepar;
9620:
9621: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9622: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9623: for(i=1; i <=nlstate; i++){
9624: jj=0;
9625: for(j=1; j <=nlstate+ndeath; j++){
9626: if(j==i) continue;
9627: jj++;
9628: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9629: printf("%1d%1d",i,j);
9630: fprintf(ficparo,"%1d%1d",i,j);
9631: for(k=1; k<=ncovmodel;k++){
9632: /* printf(" %lf",param[i][j][k]); */
9633: /* fprintf(ficparo," %lf",param[i][j][k]); */
9634: printf(" 0.");
9635: fprintf(ficparo," 0.");
9636: }
9637: printf("\n");
9638: fprintf(ficparo,"\n");
9639: }
9640: }
9641: printf("# Scales (for hessian or gradient estimation)\n");
9642: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9643: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9644: for(i=1; i <=nlstate; i++){
9645: jj=0;
9646: for(j=1; j <=nlstate+ndeath; j++){
9647: if(j==i) continue;
9648: jj++;
9649: fprintf(ficparo,"%1d%1d",i,j);
9650: printf("%1d%1d",i,j);
9651: fflush(stdout);
9652: for(k=1; k<=ncovmodel;k++){
9653: /* printf(" %le",delti3[i][j][k]); */
9654: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9655: printf(" 0.");
9656: fprintf(ficparo," 0.");
9657: }
9658: numlinepar++;
9659: printf("\n");
9660: fprintf(ficparo,"\n");
9661: }
9662: }
9663: printf("# Covariance matrix\n");
9664: /* # 121 Var(a12)\n\ */
9665: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9666: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9667: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9668: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9669: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9670: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9671: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9672: fflush(stdout);
9673: fprintf(ficparo,"# Covariance matrix\n");
9674: /* # 121 Var(a12)\n\ */
9675: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9676: /* # ...\n\ */
9677: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9678:
9679: for(itimes=1;itimes<=2;itimes++){
9680: jj=0;
9681: for(i=1; i <=nlstate; i++){
9682: for(j=1; j <=nlstate+ndeath; j++){
9683: if(j==i) continue;
9684: for(k=1; k<=ncovmodel;k++){
9685: jj++;
9686: ca[0]= k+'a'-1;ca[1]='\0';
9687: if(itimes==1){
9688: printf("#%1d%1d%d",i,j,k);
9689: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9690: }else{
9691: printf("%1d%1d%d",i,j,k);
9692: fprintf(ficparo,"%1d%1d%d",i,j,k);
9693: /* printf(" %.5le",matcov[i][j]); */
9694: }
9695: ll=0;
9696: for(li=1;li <=nlstate; li++){
9697: for(lj=1;lj <=nlstate+ndeath; lj++){
9698: if(lj==li) continue;
9699: for(lk=1;lk<=ncovmodel;lk++){
9700: ll++;
9701: if(ll<=jj){
9702: cb[0]= lk +'a'-1;cb[1]='\0';
9703: if(ll<jj){
9704: if(itimes==1){
9705: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9706: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9707: }else{
9708: printf(" 0.");
9709: fprintf(ficparo," 0.");
9710: }
9711: }else{
9712: if(itimes==1){
9713: printf(" Var(%s%1d%1d)",ca,i,j);
9714: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9715: }else{
9716: printf(" 0.");
9717: fprintf(ficparo," 0.");
9718: }
9719: }
9720: }
9721: } /* end lk */
9722: } /* end lj */
9723: } /* end li */
9724: printf("\n");
9725: fprintf(ficparo,"\n");
9726: numlinepar++;
9727: } /* end k*/
9728: } /*end j */
9729: } /* end i */
9730: } /* end itimes */
9731:
9732: } /* end of prwizard */
9733: /******************* Gompertz Likelihood ******************************/
9734: double gompertz(double x[])
9735: {
1.302 brouard 9736: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9737: int i,n=0; /* n is the size of the sample */
9738:
1.220 brouard 9739: for (i=1;i<=imx ; i++) {
1.126 brouard 9740: sump=sump+weight[i];
9741: /* sump=sump+1;*/
9742: num=num+1;
9743: }
1.302 brouard 9744: L=0.0;
9745: /* agegomp=AGEGOMP; */
1.126 brouard 9746: /* for (i=0; i<=imx; i++)
9747: 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]);*/
9748:
1.302 brouard 9749: for (i=1;i<=imx ; i++) {
9750: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9751: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9752: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9753: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9754: * +
9755: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9756: */
9757: if (wav[i] > 1 || agedc[i] < AGESUP) {
9758: if (cens[i] == 1){
9759: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9760: } else if (cens[i] == 0){
1.126 brouard 9761: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9762: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9763: } else
9764: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9765: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9766: L=L+A*weight[i];
1.126 brouard 9767: /* 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 9768: }
9769: }
1.126 brouard 9770:
1.302 brouard 9771: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9772:
9773: return -2*L*num/sump;
9774: }
9775:
1.136 brouard 9776: #ifdef GSL
9777: /******************* Gompertz_f Likelihood ******************************/
9778: double gompertz_f(const gsl_vector *v, void *params)
9779: {
1.302 brouard 9780: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9781: double *x= (double *) v->data;
9782: int i,n=0; /* n is the size of the sample */
9783:
9784: for (i=0;i<=imx-1 ; i++) {
9785: sump=sump+weight[i];
9786: /* sump=sump+1;*/
9787: num=num+1;
9788: }
9789:
9790:
9791: /* for (i=0; i<=imx; i++)
9792: 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]);*/
9793: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9794: for (i=1;i<=imx ; i++)
9795: {
9796: if (cens[i] == 1 && wav[i]>1)
9797: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9798:
9799: if (cens[i] == 0 && wav[i]>1)
9800: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9801: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9802:
9803: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9804: if (wav[i] > 1 ) { /* ??? */
9805: LL=LL+A*weight[i];
9806: /* 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]);*/
9807: }
9808: }
9809:
9810: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9811: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9812:
9813: return -2*LL*num/sump;
9814: }
9815: #endif
9816:
1.126 brouard 9817: /******************* Printing html file ***********/
1.201 brouard 9818: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9819: int lastpass, int stepm, int weightopt, char model[],\
9820: int imx, double p[],double **matcov,double agemortsup){
9821: int i,k;
9822:
9823: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9824: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9825: for (i=1;i<=2;i++)
9826: 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 9827: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9828: fprintf(fichtm,"</ul>");
9829:
9830: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9831:
9832: 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>");
9833:
9834: for (k=agegomp;k<(agemortsup-2);k++)
9835: 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]);
9836:
9837:
9838: fflush(fichtm);
9839: }
9840:
9841: /******************* Gnuplot file **************/
1.201 brouard 9842: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9843:
9844: char dirfileres[132],optfileres[132];
1.164 brouard 9845:
1.126 brouard 9846: int ng;
9847:
9848:
9849: /*#ifdef windows */
9850: fprintf(ficgp,"cd \"%s\" \n",pathc);
9851: /*#endif */
9852:
9853:
9854: strcpy(dirfileres,optionfilefiname);
9855: strcpy(optfileres,"vpl");
1.199 brouard 9856: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9857: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9858: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9859: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9860: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9861:
9862: }
9863:
1.136 brouard 9864: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9865: {
1.126 brouard 9866:
1.136 brouard 9867: /*-------- data file ----------*/
9868: FILE *fic;
9869: char dummy[]=" ";
1.240 brouard 9870: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9871: int lstra;
1.136 brouard 9872: int linei, month, year,iout;
1.302 brouard 9873: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9874: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9875: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9876: char *stratrunc;
1.223 brouard 9877:
1.240 brouard 9878: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9879: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.328 brouard 9880: for(v=1;v<NCOVMAX;v++){
9881: DummyV[v]=0;
9882: FixedV[v]=0;
9883: }
1.126 brouard 9884:
1.240 brouard 9885: for(v=1; v <=ncovcol;v++){
9886: DummyV[v]=0;
9887: FixedV[v]=0;
9888: }
9889: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9890: DummyV[v]=1;
9891: FixedV[v]=0;
9892: }
9893: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9894: DummyV[v]=0;
9895: FixedV[v]=1;
9896: }
9897: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9898: DummyV[v]=1;
9899: FixedV[v]=1;
9900: }
9901: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9902: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9903: 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]);
9904: }
1.126 brouard 9905:
1.136 brouard 9906: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9907: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9908: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9909: }
1.126 brouard 9910:
1.302 brouard 9911: /* Is it a BOM UTF-8 Windows file? */
9912: /* First data line */
9913: linei=0;
9914: while(fgets(line, MAXLINE, fic)) {
9915: noffset=0;
9916: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9917: {
9918: noffset=noffset+3;
9919: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9920: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9921: fflush(ficlog); return 1;
9922: }
9923: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9924: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9925: {
9926: noffset=noffset+2;
1.304 brouard 9927: 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);
9928: 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 9929: fflush(ficlog); return 1;
9930: }
9931: else if( line[0] == 0 && line[1] == 0)
9932: {
9933: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9934: noffset=noffset+4;
1.304 brouard 9935: 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);
9936: 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 9937: fflush(ficlog); return 1;
9938: }
9939: } else{
9940: ;/*printf(" Not a BOM file\n");*/
9941: }
9942: /* If line starts with a # it is a comment */
9943: if (line[noffset] == '#') {
9944: linei=linei+1;
9945: break;
9946: }else{
9947: break;
9948: }
9949: }
9950: fclose(fic);
9951: if((fic=fopen(datafile,"r"))==NULL) {
9952: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9953: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9954: }
9955: /* Not a Bom file */
9956:
1.136 brouard 9957: i=1;
9958: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9959: linei=linei+1;
9960: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9961: if(line[j] == '\t')
9962: line[j] = ' ';
9963: }
9964: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9965: ;
9966: };
9967: line[j+1]=0; /* Trims blanks at end of line */
9968: if(line[0]=='#'){
9969: fprintf(ficlog,"Comment line\n%s\n",line);
9970: printf("Comment line\n%s\n",line);
9971: continue;
9972: }
9973: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9974: strcpy(line, linetmp);
1.223 brouard 9975:
9976: /* Loops on waves */
9977: for (j=maxwav;j>=1;j--){
9978: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9979: cutv(stra, strb, line, ' ');
9980: if(strb[0]=='.') { /* Missing value */
9981: lval=-1;
9982: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9983: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9984: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9985: 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);
9986: 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);
9987: return 1;
9988: }
9989: }else{
9990: errno=0;
9991: /* what_kind_of_number(strb); */
9992: dval=strtod(strb,&endptr);
9993: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9994: /* if(strb != endptr && *endptr == '\0') */
9995: /* dval=dlval; */
9996: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9997: if( strb[0]=='\0' || (*endptr != '\0')){
9998: 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);
9999: 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);
10000: return 1;
10001: }
10002: cotqvar[j][iv][i]=dval;
10003: cotvar[j][ntv+iv][i]=dval;
10004: }
10005: strcpy(line,stra);
1.223 brouard 10006: }/* end loop ntqv */
1.225 brouard 10007:
1.223 brouard 10008: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 10009: cutv(stra, strb, line, ' ');
10010: if(strb[0]=='.') { /* Missing value */
10011: lval=-1;
10012: }else{
10013: errno=0;
10014: lval=strtol(strb,&endptr,10);
10015: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10016: if( strb[0]=='\0' || (*endptr != '\0')){
10017: printf("Error reading data around '%s' at line number %d for individual %d, '%s'\nShould be the %d th 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);
10018: 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);
10019: return 1;
10020: }
10021: }
10022: if(lval <-1 || lval >1){
10023: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10024: 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 10025: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10026: For example, for multinomial values like 1, 2 and 3,\n \
10027: build V1=0 V2=0 for the reference value (1),\n \
10028: V1=1 V2=0 for (2) \n \
1.223 brouard 10029: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10030: output of IMaCh is often meaningless.\n \
1.319 brouard 10031: Exiting.\n",lval,linei, i,line,iv,j);
1.238 brouard 10032: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.319 brouard 10033: 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 10034: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 10035: For example, for multinomial values like 1, 2 and 3,\n \
10036: build V1=0 V2=0 for the reference value (1),\n \
10037: V1=1 V2=0 for (2) \n \
1.223 brouard 10038: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 10039: output of IMaCh is often meaningless.\n \
1.319 brouard 10040: Exiting.\n",lval,linei, i,line,iv,j);fflush(ficlog);
1.238 brouard 10041: return 1;
10042: }
10043: cotvar[j][iv][i]=(double)(lval);
10044: strcpy(line,stra);
1.223 brouard 10045: }/* end loop ntv */
1.225 brouard 10046:
1.223 brouard 10047: /* Statuses at wave */
1.137 brouard 10048: cutv(stra, strb, line, ' ');
1.223 brouard 10049: if(strb[0]=='.') { /* Missing value */
1.238 brouard 10050: lval=-1;
1.136 brouard 10051: }else{
1.238 brouard 10052: errno=0;
10053: lval=strtol(strb,&endptr,10);
10054: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
10055: if( strb[0]=='\0' || (*endptr != '\0')){
10056: 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);
10057: 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);
10058: return 1;
10059: }
1.136 brouard 10060: }
1.225 brouard 10061:
1.136 brouard 10062: s[j][i]=lval;
1.225 brouard 10063:
1.223 brouard 10064: /* Date of Interview */
1.136 brouard 10065: strcpy(line,stra);
10066: cutv(stra, strb,line,' ');
1.169 brouard 10067: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10068: }
1.169 brouard 10069: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 10070: month=99;
10071: year=9999;
1.136 brouard 10072: }else{
1.225 brouard 10073: 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);
10074: 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);
10075: return 1;
1.136 brouard 10076: }
10077: anint[j][i]= (double) year;
1.302 brouard 10078: mint[j][i]= (double)month;
10079: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
10080: /* 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]); */
10081: /* 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]); */
10082: /* } */
1.136 brouard 10083: strcpy(line,stra);
1.223 brouard 10084: } /* End loop on waves */
1.225 brouard 10085:
1.223 brouard 10086: /* Date of death */
1.136 brouard 10087: cutv(stra, strb,line,' ');
1.169 brouard 10088: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10089: }
1.169 brouard 10090: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 10091: month=99;
10092: year=9999;
10093: }else{
1.141 brouard 10094: 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 10095: 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);
10096: return 1;
1.136 brouard 10097: }
10098: andc[i]=(double) year;
10099: moisdc[i]=(double) month;
10100: strcpy(line,stra);
10101:
1.223 brouard 10102: /* Date of birth */
1.136 brouard 10103: cutv(stra, strb,line,' ');
1.169 brouard 10104: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 10105: }
1.169 brouard 10106: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 10107: month=99;
10108: year=9999;
10109: }else{
1.141 brouard 10110: 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);
10111: 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 10112: return 1;
1.136 brouard 10113: }
10114: if (year==9999) {
1.141 brouard 10115: 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);
10116: 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 10117: return 1;
10118:
1.136 brouard 10119: }
10120: annais[i]=(double)(year);
1.302 brouard 10121: moisnais[i]=(double)(month);
10122: for (j=1;j<=maxwav;j++){
10123: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
10124: 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]);
10125: 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]);
10126: }
10127: }
10128:
1.136 brouard 10129: strcpy(line,stra);
1.225 brouard 10130:
1.223 brouard 10131: /* Sample weight */
1.136 brouard 10132: cutv(stra, strb,line,' ');
10133: errno=0;
10134: dval=strtod(strb,&endptr);
10135: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 10136: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
10137: 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 10138: fflush(ficlog);
10139: return 1;
10140: }
10141: weight[i]=dval;
10142: strcpy(line,stra);
1.225 brouard 10143:
1.223 brouard 10144: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
10145: cutv(stra, strb, line, ' ');
10146: if(strb[0]=='.') { /* Missing value */
1.225 brouard 10147: lval=-1;
1.311 brouard 10148: coqvar[iv][i]=NAN;
10149: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 10150: }else{
1.225 brouard 10151: errno=0;
10152: /* what_kind_of_number(strb); */
10153: dval=strtod(strb,&endptr);
10154: /* if(strb != endptr && *endptr == '\0') */
10155: /* dval=dlval; */
10156: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
10157: if( strb[0]=='\0' || (*endptr != '\0')){
10158: 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);
10159: 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);
10160: return 1;
10161: }
10162: coqvar[iv][i]=dval;
1.226 brouard 10163: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 10164: }
10165: strcpy(line,stra);
10166: }/* end loop nqv */
1.136 brouard 10167:
1.223 brouard 10168: /* Covariate values */
1.136 brouard 10169: for (j=ncovcol;j>=1;j--){
10170: cutv(stra, strb,line,' ');
1.223 brouard 10171: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 10172: lval=-1;
1.136 brouard 10173: }else{
1.225 brouard 10174: errno=0;
10175: lval=strtol(strb,&endptr,10);
10176: if( strb[0]=='\0' || (*endptr != '\0')){
10177: 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);
10178: 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);
10179: return 1;
10180: }
1.136 brouard 10181: }
10182: if(lval <-1 || lval >1){
1.225 brouard 10183: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10184: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10185: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10186: For example, for multinomial values like 1, 2 and 3,\n \
10187: build V1=0 V2=0 for the reference value (1),\n \
10188: V1=1 V2=0 for (2) \n \
1.136 brouard 10189: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10190: output of IMaCh is often meaningless.\n \
1.136 brouard 10191: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 10192: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 10193: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
10194: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 10195: For example, for multinomial values like 1, 2 and 3,\n \
10196: build V1=0 V2=0 for the reference value (1),\n \
10197: V1=1 V2=0 for (2) \n \
1.136 brouard 10198: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 10199: output of IMaCh is often meaningless.\n \
1.136 brouard 10200: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 10201: return 1;
1.136 brouard 10202: }
10203: covar[j][i]=(double)(lval);
10204: strcpy(line,stra);
10205: }
10206: lstra=strlen(stra);
1.225 brouard 10207:
1.136 brouard 10208: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
10209: stratrunc = &(stra[lstra-9]);
10210: num[i]=atol(stratrunc);
10211: }
10212: else
10213: num[i]=atol(stra);
10214: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
10215: 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;}*/
10216:
10217: i=i+1;
10218: } /* End loop reading data */
1.225 brouard 10219:
1.136 brouard 10220: *imax=i-1; /* Number of individuals */
10221: fclose(fic);
1.225 brouard 10222:
1.136 brouard 10223: return (0);
1.164 brouard 10224: /* endread: */
1.225 brouard 10225: printf("Exiting readdata: ");
10226: fclose(fic);
10227: return (1);
1.223 brouard 10228: }
1.126 brouard 10229:
1.234 brouard 10230: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 10231: char *p1 = *stri, *p2 = *stri;
1.235 brouard 10232: while (*p2 == ' ')
1.234 brouard 10233: p2++;
10234: /* while ((*p1++ = *p2++) !=0) */
10235: /* ; */
10236: /* do */
10237: /* while (*p2 == ' ') */
10238: /* p2++; */
10239: /* while (*p1++ == *p2++); */
10240: *stri=p2;
1.145 brouard 10241: }
10242:
1.330 brouard 10243: int decoderesult( char resultline[], int nres)
1.230 brouard 10244: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
10245: {
1.235 brouard 10246: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 10247: char resultsav[MAXLINE];
1.330 brouard 10248: /* int resultmodel[MAXLINE]; */
1.234 brouard 10249: int modelresult[MAXLINE];
1.230 brouard 10250: char stra[80], strb[80], strc[80], strd[80],stre[80];
10251:
1.234 brouard 10252: removefirstspace(&resultline);
1.332 brouard 10253: printf("decoderesult:%s\n",resultline);
1.230 brouard 10254:
1.332 brouard 10255: strcpy(resultsav,resultline);
10256: printf("Decoderesult resultsav=\"%s\" resultline=\"%s\"\n", resultsav, resultline);
1.230 brouard 10257: if (strlen(resultsav) >1){
10258: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
10259: }
1.253 brouard 10260: if(j == 0){ /* Resultline but no = */
10261: TKresult[nres]=0; /* Combination for the nresult and the model */
10262: return (0);
10263: }
1.234 brouard 10264: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.332 brouard 10265: printf("ERROR: the number of variables in the resultline which is %d, differs from the number %d of variables used in the model line, %s.\n",j, cptcovs, model);
10266: fprintf(ficlog,"ERROR: the number of variables in the resultline which is %d, differs from the number %d of variables used in the model line, %s.\n",j, cptcovs, model);
10267: /* return 1;*/
1.234 brouard 10268: }
10269: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
10270: if(nbocc(resultsav,'=') >1){
1.318 brouard 10271: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' ' (stra is the rest of the resultline to be analyzed in the next loop *//* resultsav= "V4=1 V5=25.1 V3=0" stra= "V5=25.1 V3=0" strb= "V4=1" */
1.332 brouard 10272: /* If resultsav= "V4= 1 V5=25.1 V3=0" with a blank then strb="V4=" and stra="1 V5=25.1 V3=0" */
1.318 brouard 10273: cutl(strc,strd,strb,'='); /* strb:"V4=1" strc="1" strd="V4" */
1.332 brouard 10274: /* If a blank, then strc="V4=" and strd='\0' */
10275: if(strc[0]=='\0'){
10276: printf("Error in resultline, probably a blank after the \"%s\", \"result:%s\", stra=\"%s\" resultsav=\"%s\"\n",strb,resultline, stra, resultsav);
10277: fprintf(ficlog,"Error in resultline, probably a blank after the \"V%s=\", resultline=%s\n",strb,resultline);
10278: return 1;
10279: }
1.234 brouard 10280: }else
10281: cutl(strc,strd,resultsav,'=');
1.318 brouard 10282: Tvalsel[k]=atof(strc); /* 1 */ /* Tvalsel of k is the float value of the kth covariate appearing in this result line */
1.234 brouard 10283:
1.230 brouard 10284: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
1.318 brouard 10285: 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 10286: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
10287: /* cptcovsel++; */
10288: if (nbocc(stra,'=') >0)
10289: strcpy(resultsav,stra); /* and analyzes it */
10290: }
1.235 brouard 10291: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10292: /* Feeds resultmodel[nres][k1]=k2 for k1th product covariate with age in the model equation fed by the index k2 of the resutline*/
1.318 brouard 10293: for(k1=1; k1<= cptcovt ;k1++){ /* Loop on model. model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.332 brouard 10294: if(Typevar[k1]==0){ /* Single covariate in model */
10295: /* 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.234 brouard 10296: match=0;
1.318 brouard 10297: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10298: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 10299: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.318 brouard 10300: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
1.234 brouard 10301: break;
10302: }
10303: }
10304: if(match == 0){
1.332 brouard 10305: printf("Error in result line (Dummy single): V%d is missing in result: %s according to model=%s. Tvar[k1=%d]=%d is different from Tvarsel[k2=%d]=%d.\n",Tvar[k1], resultline, model,k1, Tvar[k1], k2, Tvarsel[k2]);
10306: fprintf(ficlog,"Error in result line (Dummy single): V%d is missing in result: %s according to model=%s\n",Tvar[k1], resultline, model);
1.310 brouard 10307: return 1;
1.234 brouard 10308: }
1.332 brouard 10309: }else if(Typevar[k1]==1){ /* Product with age We want to get the position k2 in the resultline of the product k1 in the model line*/
10310: /* We feed resultmodel[k1]=k2; */
10311: match=0;
10312: for(k2=1; k2 <=j;k2++){/* Loop on resultline. jth occurence of = signs in the result line. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10313: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10314: modelresult[k2]=k1;/* we found a Vn=1 corrresponding to Vn*age in the model modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
10315: resultmodel[nres][k1]=k2; /* Added here */
10316: printf("Decoderesult first modelresult[k2=%d]=%d (k1) V%d*AGE\n",k2,k1,Tvar[k1]);
10317: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10318: break;
10319: }
10320: }
10321: if(match == 0){
10322: printf("Error in result line (Product with age): V%d is missing in result: %s according to model=%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
1.333 ! brouard 10323: fprintf(ficlog,"Error in result line (Product with age): V%d is missing in result: %s according to model=%s (Tvarsel[k2=%d]=%d)\n",Tvar[k1], resultline, model, k2, Tvarsel[k2]);
1.332 brouard 10324: return 1;
10325: }
10326: }else if(Typevar[k1]==2){ /* Product No age We want to get the position in the resultline of the product in the model line*/
10327: /* resultmodel[nres][of such a Vn * Vm product k1] is not unique, so can't exist, we feed Tvard[k1][1] and [2] */
10328: match=0;
10329: printf("Decoderesult very first Product Tvardk[k1=%d][1]=%d Tvardk[k1=%d][2]=%d V%d * V%d\n",k1,Tvardk[k1][1],k1,Tvardk[k1][2],Tvardk[k1][1],Tvardk[k1][2]);
10330: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10331: if(Tvardk[k1][1]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10332: /* modelresult[k2]=k1; */
10333: printf("Decoderesult first Product modelresult[k2=%d]=%d (k1) V%d * \n",k2,k1,Tvarsel[k2]);
10334: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10335: }
10336: }
10337: if(match == 0){
10338: printf("Error in result line (Product without age first variable): V%d is missing in result: %s according to model=%s\n",Tvardk[k1][1], resultline, model);
1.333 ! brouard 10339: fprintf(ficlog,"Error in result line (Product without age first variable): V%d is missing in result: %s according to model=%s\n",Tvardk[k1][1], resultline, model);
1.332 brouard 10340: return 1;
10341: }
10342: match=0;
10343: for(k2=1; k2 <=j;k2++){/* Loop on resultline. In result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
10344: if(Tvardk[k1][2]==Tvarsel[k2]) {/* Tvardk is coming from the model, Tvarsel from the result. Tvar[1]=5 == Tvarsel[2]=5 */
10345: /* modelresult[k2]=k1;*/
10346: printf("Decoderesult second Product modelresult[k2=%d]=%d (k1) * V%d \n ",k2,k1,Tvarsel[k2]);
10347: match=1; /* modelresult of k2 variable of resultline is identical to k1 variable of the model good */
10348: break;
10349: }
10350: }
10351: if(match == 0){
10352: printf("Error in result line (Product without age second variable): V%d is missing in result: %s according to model=%s\n",Tvardk[k1][2], resultline, model);
1.333 ! brouard 10353: fprintf(ficlog,"Error in result line (Product without age second variable): V%d is missing in result : %s according to model=%s\n",Tvardk[k1][2], resultline, model);
1.332 brouard 10354: return 1;
10355: }
10356: }/* End of testing */
1.333 ! brouard 10357: }/* End loop cptcovt */
1.235 brouard 10358: /* Checking for missing or useless values in comparison of current model needs */
1.332 brouard 10359: /* Feeds resultmodel[nres][k1]=k2 for single covariate (k1) in the model equation */
1.318 brouard 10360: 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 10361: match=0;
1.318 brouard 10362: for(k1=1; k1<= cptcovt ;k1++){ /* loop on model: model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.332 brouard 10363: if(Typevar[k1]==0){ /* Single only */
1.237 brouard 10364: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.330 brouard 10365: resultmodel[nres][k1]=k2; /* k1th position in the model equation corresponds to k2th position in the result line. resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 10366: ++match;
10367: }
10368: }
10369: }
10370: if(match == 0){
1.332 brouard 10371: printf("Error in result line: variable V%d is missing in model; result: %s, model=%s\n",Tvarsel[k2], resultline, model);
10372: fprintf(ficlog,"Error in result line: variable V%d is missing in model; result: %s, model=%s\n",Tvarsel[k2], resultline, model);
1.310 brouard 10373: return 1;
1.234 brouard 10374: }else if(match > 1){
10375: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 10376: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
10377: return 1;
1.234 brouard 10378: }
10379: }
1.235 brouard 10380:
1.234 brouard 10381: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 10382: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.330 brouard 10383: /* nres=1st result line: V4=1 V5=25.1 V3=0 V2=8 V1=1 */
10384: /* should correspond to the combination 6 of dummy: V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 1*1 + 0*2 + 1*4 = 5 + (1offset) = 6*/
10385: /* nres=2nd result line: V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.235 brouard 10386: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
10387: /* 1 0 0 0 */
10388: /* 2 1 0 0 */
10389: /* 3 0 1 0 */
1.330 brouard 10390: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 (nres=2)*/
1.235 brouard 10391: /* 5 0 0 1 */
1.330 brouard 10392: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 (nres=1)*/
1.235 brouard 10393: /* 7 0 1 1 */
10394: /* 8 1 1 1 */
1.237 brouard 10395: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
10396: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
10397: /* V5*age V5 known which value for nres? */
10398: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.330 brouard 10399: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* loop k1 on position in the model line (excluding product) */
1.331 brouard 10400: /* k counting number of combination of single dummies in the equation model */
10401: /* k4 counting single dummies in the equation model */
10402: /* k4q counting single quantitatives in the equation model */
10403: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Dummy and Single */
10404: /* k4+1= position in the resultline V(Tvarsel)=Tvalsel=Tresult[nres][pos](value); V(Tvresult[nres][pos] (variable): V(variable)=value) */
1.332 brouard 10405: /* modelresult[k3]=k1: k3th position in the result line corresponds to the k1 position in the model line (doesn't work with products)*/
1.330 brouard 10406: /* Value in the (current nres) resultline of the variable at the k1th position in the model equation resultmodel[nres][k1]= k3 */
1.332 brouard 10407: /* resultmodel[nres][k1]=k3: k1th position in the model correspond to the k3 position in the resultline */
10408: /* k3 is the position in the nres result line of the k1th variable of the model equation */
10409: /* Tvarsel[k3]: Name of the variable at the k3th position in the result line. */
10410: /* Tvalsel[k3]: Value of the variable at the k3th position in the result line. */
10411: /* Tresult[nres][result_position]= value of the dummy variable at the result_position in the nres resultline */
10412: /* Tvresult[nres][result_position]= id of the dummy variable at the result_position in the nres resultline */
10413: /* Tinvresult[nres][Name of a dummy variable]= value of the variable in the result line */
1.330 brouard 10414: /* TinvDoQresult[nres][Name of a Dummy or Q variable]= value of the variable in the result line */
1.332 brouard 10415: k3= resultmodel[nres][k1]; /* From position k1 in model get position k3 in result line */
10416: /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
10417: k2=(int)Tvarsel[k3]; /* from position k3 in resultline get name k2: nres=1 k1=2=>k3=1 Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 (V4); k1=3=>k3=2 Tvarsel[2]=3 (V3)*/
1.330 brouard 10418: k+=Tvalsel[k3]*pow(2,k4); /* nres=1 k1=2 Tvalsel[1]=1 (V4=1); k1=3 k3=2 Tvalsel[2]=0 (V3=0) */
1.332 brouard 10419: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Stores the value into the name of the variable. */
10420: /* Tinvresult[nres][4]=1 */
1.330 brouard 10421: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres=2][1]=1(V4=1) Tresult[nres=2][2]=0(V3=0) */
1.237 brouard 10422: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
10423: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.332 brouard 10424: precov[nres][k1]=Tvalsel[k3];
10425: printf("Decoderesult Dummy k=%d, k1=%d precov[nres=%d][k1=%d]=%.f V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k1, nres, k1,precov[nres][k1], k2, k3, (int)Tvalsel[k3], k4);
1.235 brouard 10426: k4++;;
1.331 brouard 10427: }else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Quantitative and single */
1.330 brouard 10428: /* Tqresult[nres][result_position]= value of the variable at the result_position in the nres resultline */
1.332 brouard 10429: /* Tvqresult[nres][result_position]= id of the variable at the result_position in the nres resultline */
1.330 brouard 10430: /* Tqinvresult[nres][Name of a quantitative variable]= value of the variable in the result line */
1.332 brouard 10431: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 5 =k3q */
10432: k2q=(int)Tvarsel[k3q]; /* Name of variable at k3q th position in the resultline */
10433: /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
1.237 brouard 10434: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
10435: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
10436: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.330 brouard 10437: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 10438: precov[nres][k1]=Tvalsel[k3q];
10439: printf("Decoderesult Quantitative nres=%d,precov[nres=%d][k1=%d]=%.f V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, nres, k1,precov[nres][k1], k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
1.235 brouard 10440: k4q++;;
1.331 brouard 10441: }else if( Dummy[k1]==2 ){ /* For dummy with age product */
10442: /* Tvar[k1]; */ /* Age variable */
1.332 brouard 10443: /* Wrong we want the value of variable name Tvar[k1] */
10444:
10445: k3= resultmodel[nres][k1]; /* nres=1 k1=2 resultmodel[2(V4)] = 1=k3 ; k1=3 resultmodel[3(V3)] = 2=k3*/
1.331 brouard 10446: k2=(int)Tvarsel[k3]; /* nres=1 k1=2=>k3=1 Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 (V4); k1=3=>k3=2 Tvarsel[2]=3 (V3)*/
10447: TinvDoQresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.332 brouard 10448: precov[nres][k1]=Tvalsel[k3];
10449: printf("Decoderesult Dummy with age k=%d, k1=%d precov[nres=%d][k1=%d]=%.f Tvar[%d]=V%d k2=Tvarsel[%d]=%d Tvalsel[%d]=%d\n",k, k1, nres, k1,precov[nres][k1], k1, Tvar[k1], k3,(int)Tvarsel[k3], k3, (int)Tvalsel[k3]);
1.331 brouard 10450: }else if( Dummy[k1]==3 ){ /* For quant with age product */
1.332 brouard 10451: k3q= resultmodel[nres][k1]; /* resultmodel[1(V5)] = 25.1=k3q */
1.331 brouard 10452: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[1]]= Tvarsel[1] = 4=k2 */
10453: TinvDoQresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.332 brouard 10454: precov[nres][k1]=Tvalsel[k3q];
10455: printf("Decoderesult Quantitative with age nres=%d, k1=%d, precov[nres=%d][k1=%d]=%.f Tvar[%d]=V%d V(k2q=%d)= Tvarsel[%d]=%d, Tvalsel[%d]=%f\n",nres, k1, nres, k1,precov[nres][k1], k1, Tvar[k1], k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
1.331 brouard 10456: }else if(Typevar[k1]==2 ){ /* For product quant or dummy (not with age) */
1.332 brouard 10457: precov[nres][k1]=TinvDoQresult[nres][Tvardk[k1][1]] * TinvDoQresult[nres][Tvardk[k1][2]];
10458: printf("Decoderesult Quantitative or Dummy (not with age) nres=%d k1=%d precov[nres=%d][k1=%d]=%.f V%d(=%.f) * V%d(=%.f) \n",nres, k1, nres, k1,precov[nres][k1], Tvardk[k1][1], TinvDoQresult[nres][Tvardk[k1][1]], Tvardk[k1][2], TinvDoQresult[nres][Tvardk[k1][2]]);
1.330 brouard 10459: }else{
1.332 brouard 10460: printf("Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
10461: fprintf(ficlog,"Error Decoderesult probably a product Dummy[%d]==%d && Typevar[%d]==%d\n", k1, Dummy[k1], k1, Typevar[k1]);
1.235 brouard 10462: }
10463: }
1.234 brouard 10464:
1.235 brouard 10465: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 10466: return (0);
10467: }
1.235 brouard 10468:
1.230 brouard 10469: int decodemodel( char model[], int lastobs)
10470: /**< This routine decodes the model and returns:
1.224 brouard 10471: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
10472: * - nagesqr = 1 if age*age in the model, otherwise 0.
10473: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
10474: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
10475: * - cptcovage number of covariates with age*products =2
10476: * - cptcovs number of simple covariates
10477: * - 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
10478: * which is a new column after the 9 (ncovcol) variables.
1.319 brouard 10479: * - if k is a product Vn*Vm, covar[k][i] is filled with correct values for each individual
1.224 brouard 10480: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
10481: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
10482: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
10483: */
1.319 brouard 10484: /* 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 10485: {
1.238 brouard 10486: int i, j, k, ks, v;
1.227 brouard 10487: int j1, k1, k2, k3, k4;
1.136 brouard 10488: char modelsav[80];
1.145 brouard 10489: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 10490: char *strpt;
1.136 brouard 10491:
1.145 brouard 10492: /*removespace(model);*/
1.136 brouard 10493: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 10494: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 10495: if (strstr(model,"AGE") !=0){
1.192 brouard 10496: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
10497: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 10498: return 1;
10499: }
1.141 brouard 10500: if (strstr(model,"v") !=0){
10501: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
10502: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
10503: return 1;
10504: }
1.187 brouard 10505: strcpy(modelsav,model);
10506: if ((strpt=strstr(model,"age*age")) !=0){
10507: printf(" strpt=%s, model=%s\n",strpt, model);
10508: if(strpt != model){
1.234 brouard 10509: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10510: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10511: corresponding column of parameters.\n",model);
1.234 brouard 10512: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 10513: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 10514: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 10515: return 1;
1.225 brouard 10516: }
1.187 brouard 10517: nagesqr=1;
10518: if (strstr(model,"+age*age") !=0)
1.234 brouard 10519: substrchaine(modelsav, model, "+age*age");
1.187 brouard 10520: else if (strstr(model,"age*age+") !=0)
1.234 brouard 10521: substrchaine(modelsav, model, "age*age+");
1.187 brouard 10522: else
1.234 brouard 10523: substrchaine(modelsav, model, "age*age");
1.187 brouard 10524: }else
10525: nagesqr=0;
10526: if (strlen(modelsav) >1){
10527: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
10528: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 10529: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 10530: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 10531: * cst, age and age*age
10532: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
10533: /* including age products which are counted in cptcovage.
10534: * but the covariates which are products must be treated
10535: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 10536: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
10537: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 10538:
10539:
1.187 brouard 10540: /* Design
10541: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
10542: * < ncovcol=8 >
10543: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
10544: * k= 1 2 3 4 5 6 7 8
10545: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
10546: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 10547: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
10548: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 10549: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
10550: * Tage[++cptcovage]=k
10551: * if products, new covar are created after ncovcol with k1
10552: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
10553: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
10554: * 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
10555: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
10556: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
10557: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
10558: * < ncovcol=8 >
10559: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
10560: * k= 1 2 3 4 5 6 7 8 9 10 11 12
10561: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
1.319 brouard 10562: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
1.187 brouard 10563: * p Tprod[1]@2={ 6, 5}
10564: *p Tvard[1][1]@4= {7, 8, 5, 6}
10565: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
10566: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
1.319 brouard 10567: *How to reorganize? Tvars(orted)
1.187 brouard 10568: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
10569: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
10570: * {2, 1, 4, 8, 5, 6, 3, 7}
10571: * Struct []
10572: */
1.225 brouard 10573:
1.187 brouard 10574: /* This loop fills the array Tvar from the string 'model'.*/
10575: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
10576: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
10577: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
10578: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
10579: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
10580: /* k=1 Tvar[1]=2 (from V2) */
10581: /* k=5 Tvar[5] */
10582: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 10583: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 10584: /* } */
1.198 brouard 10585: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 10586: /*
10587: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 10588: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
10589: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
10590: }
1.187 brouard 10591: cptcovage=0;
1.319 brouard 10592: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model line */
10593: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+' cutl from left to right
10594: 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" */
10595: if (nbocc(modelsav,'+')==0)
10596: strcpy(strb,modelsav); /* and analyzes it */
1.234 brouard 10597: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
10598: /*scanf("%d",i);*/
1.319 brouard 10599: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V5*age+ V4+V3*age strb=V3*age */
10600: 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 10601: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
10602: /* covar is not filled and then is empty */
10603: cptcovprod--;
10604: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
1.319 brouard 10605: 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 10606: Typevar[k]=1; /* 1 for age product */
1.319 brouard 10607: cptcovage++; /* Counts the number of covariates which include age as a product */
10608: 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 10609: /*printf("stre=%s ", stre);*/
10610: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
10611: cptcovprod--;
10612: cutl(stre,strb,strc,'V');
10613: Tvar[k]=atoi(stre);
10614: Typevar[k]=1; /* 1 for age product */
10615: cptcovage++;
10616: Tage[cptcovage]=k;
10617: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
10618: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
10619: cptcovn++;
10620: cptcovprodnoage++;k1++;
10621: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
10622: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
10623: because this model-covariate is a construction we invent a new column
10624: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
1.319 brouard 10625: If already ncovcol=4 and model=V2 + V1 +V1*V4 +age*V3 +V3*V2
10626: thus after V4 we invent V5 and V6 because age*V3 will be computed in 4
10627: Tvar[3=V1*V4]=4+1=5 Tvar[5=V3*V2]=4 + 2= 6, Tvar[4=age*V3]=4 etc */
1.234 brouard 10628: Typevar[k]=2; /* 2 for double fixed dummy covariates */
10629: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
10630: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
1.319 brouard 10631: Tposprod[k]=k1; /* Tposprod[3]=1, Tposprod[2]=5 */
1.234 brouard 10632: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
1.330 brouard 10633: Tvardk[k][1] =atoi(strc); /* m 1 for V1*/
1.234 brouard 10634: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
1.330 brouard 10635: Tvardk[k][2] =atoi(stre); /* n 4 for V4*/
1.234 brouard 10636: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
10637: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
10638: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 10639: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 10640: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
10641: for (i=1; i<=lastobs;i++){
10642: /* Computes the new covariate which is a product of
10643: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
10644: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
10645: }
10646: } /* End age is not in the model */
10647: } /* End if model includes a product */
1.319 brouard 10648: else { /* not a product */
1.234 brouard 10649: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
10650: /* scanf("%d",i);*/
10651: cutl(strd,strc,strb,'V');
10652: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
10653: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
10654: Tvar[k]=atoi(strd);
10655: Typevar[k]=0; /* 0 for simple covariates */
10656: }
10657: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 10658: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 10659: scanf("%d",i);*/
1.187 brouard 10660: } /* end of loop + on total covariates */
10661: } /* end if strlen(modelsave == 0) age*age might exist */
10662: } /* end if strlen(model == 0) */
1.136 brouard 10663:
10664: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
10665: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 10666:
1.136 brouard 10667: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 10668: printf("cptcovprod=%d ", cptcovprod);
10669: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
10670: scanf("%d ",i);*/
10671:
10672:
1.230 brouard 10673: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
10674: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 10675: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
10676: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
10677: k = 1 2 3 4 5 6 7 8 9
10678: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
1.319 brouard 10679: Typevar[k]= 0 0 0 2 1 0 2 1 0
1.227 brouard 10680: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
10681: Dummy[k] 1 0 0 0 3 1 1 2 3
10682: Tmodelind[combination of covar]=k;
1.225 brouard 10683: */
10684: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 10685: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 10686: /* 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 10687: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.318 brouard 10688: printf("Model=1+age+%s\n\
1.227 brouard 10689: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10690: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10691: 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 10692: fprintf(ficlog,"Model=1+age+%s\n\
1.227 brouard 10693: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
10694: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
10695: 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 10696: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 10697: 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 */
10698: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10699: Fixed[k]= 0;
10700: Dummy[k]= 0;
1.225 brouard 10701: ncoveff++;
1.232 brouard 10702: ncovf++;
1.234 brouard 10703: nsd++;
10704: modell[k].maintype= FTYPE;
10705: TvarsD[nsd]=Tvar[k];
10706: TvarsDind[nsd]=k;
1.330 brouard 10707: TnsdVar[Tvar[k]]=nsd;
1.234 brouard 10708: TvarF[ncovf]=Tvar[k];
10709: TvarFind[ncovf]=k;
10710: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10711: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10712: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10713: Fixed[k]= 0;
10714: Dummy[k]= 0;
10715: ncoveff++;
10716: ncovf++;
10717: modell[k].maintype= FTYPE;
10718: TvarF[ncovf]=Tvar[k];
1.330 brouard 10719: /* TnsdVar[Tvar[k]]=nsd; */ /* To be done */
1.234 brouard 10720: TvarFind[ncovf]=k;
1.230 brouard 10721: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10722: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10723: }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 10724: Fixed[k]= 0;
10725: Dummy[k]= 1;
1.230 brouard 10726: nqfveff++;
1.234 brouard 10727: modell[k].maintype= FTYPE;
10728: modell[k].subtype= FQ;
10729: nsq++;
10730: TvarsQ[nsq]=Tvar[k];
10731: TvarsQind[nsq]=k;
1.232 brouard 10732: ncovf++;
1.234 brouard 10733: TvarF[ncovf]=Tvar[k];
10734: TvarFind[ncovf]=k;
1.231 brouard 10735: 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 10736: 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 10737: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10738: Fixed[k]= 1;
10739: Dummy[k]= 0;
1.225 brouard 10740: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10741: modell[k].maintype= VTYPE;
10742: modell[k].subtype= VD;
10743: nsd++;
10744: TvarsD[nsd]=Tvar[k];
10745: TvarsDind[nsd]=k;
1.330 brouard 10746: TnsdVar[Tvar[k]]=nsd; /* To be verified */
1.234 brouard 10747: ncovv++; /* Only simple time varying variables */
10748: TvarV[ncovv]=Tvar[k];
1.242 brouard 10749: 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 10750: 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 */
10751: 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 10752: 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);
10753: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10754: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10755: Fixed[k]= 1;
10756: Dummy[k]= 1;
10757: nqtveff++;
10758: modell[k].maintype= VTYPE;
10759: modell[k].subtype= VQ;
10760: ncovv++; /* Only simple time varying variables */
10761: nsq++;
1.319 brouard 10762: TvarsQ[nsq]=Tvar[k]; /* k=1 Tvar=5 nsq=1 TvarsQ[1]=5 */
1.332 brouard 10763: TvarsQind[nsq]=k; /* For single quantitative covariate gives the model position of each single quantitative covariate */
1.234 brouard 10764: TvarV[ncovv]=Tvar[k];
1.242 brouard 10765: 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 10766: 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 */
10767: 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 10768: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10769: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10770: 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 10771: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10772: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10773: ncova++;
10774: TvarA[ncova]=Tvar[k];
10775: TvarAind[ncova]=k;
1.231 brouard 10776: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10777: Fixed[k]= 2;
10778: Dummy[k]= 2;
10779: modell[k].maintype= ATYPE;
10780: modell[k].subtype= APFD;
10781: /* ncoveff++; */
1.227 brouard 10782: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10783: Fixed[k]= 2;
10784: Dummy[k]= 3;
10785: modell[k].maintype= ATYPE;
10786: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10787: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10788: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10789: Fixed[k]= 3;
10790: Dummy[k]= 2;
10791: modell[k].maintype= ATYPE;
10792: modell[k].subtype= APVD; /* Product age * varying dummy */
10793: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10794: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10795: Fixed[k]= 3;
10796: Dummy[k]= 3;
10797: modell[k].maintype= ATYPE;
10798: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10799: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10800: }
10801: }else if (Typevar[k] == 2) { /* product without age */
10802: k1=Tposprod[k];
10803: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10804: if(Tvard[k1][2] <=ncovcol){
10805: Fixed[k]= 1;
10806: Dummy[k]= 0;
10807: modell[k].maintype= FTYPE;
10808: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10809: ncovf++; /* Fixed variables without age */
10810: TvarF[ncovf]=Tvar[k];
10811: TvarFind[ncovf]=k;
10812: }else if(Tvard[k1][2] <=ncovcol+nqv){
10813: Fixed[k]= 0; /* or 2 ?*/
10814: Dummy[k]= 1;
10815: modell[k].maintype= FTYPE;
10816: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10817: ncovf++; /* Varying variables without age */
10818: TvarF[ncovf]=Tvar[k];
10819: TvarFind[ncovf]=k;
10820: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10821: Fixed[k]= 1;
10822: Dummy[k]= 0;
10823: modell[k].maintype= VTYPE;
10824: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10825: ncovv++; /* Varying variables without age */
10826: TvarV[ncovv]=Tvar[k];
10827: TvarVind[ncovv]=k;
10828: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10829: Fixed[k]= 1;
10830: Dummy[k]= 1;
10831: modell[k].maintype= VTYPE;
10832: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10833: ncovv++; /* Varying variables without age */
10834: TvarV[ncovv]=Tvar[k];
10835: TvarVind[ncovv]=k;
10836: }
1.227 brouard 10837: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10838: if(Tvard[k1][2] <=ncovcol){
10839: Fixed[k]= 0; /* or 2 ?*/
10840: Dummy[k]= 1;
10841: modell[k].maintype= FTYPE;
10842: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10843: ncovf++; /* Fixed variables without age */
10844: TvarF[ncovf]=Tvar[k];
10845: TvarFind[ncovf]=k;
10846: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10847: Fixed[k]= 1;
10848: Dummy[k]= 1;
10849: modell[k].maintype= VTYPE;
10850: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10851: ncovv++; /* Varying variables without age */
10852: TvarV[ncovv]=Tvar[k];
10853: TvarVind[ncovv]=k;
10854: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10855: Fixed[k]= 1;
10856: Dummy[k]= 1;
10857: modell[k].maintype= VTYPE;
10858: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10859: ncovv++; /* Varying variables without age */
10860: TvarV[ncovv]=Tvar[k];
10861: TvarVind[ncovv]=k;
10862: ncovv++; /* Varying variables without age */
10863: TvarV[ncovv]=Tvar[k];
10864: TvarVind[ncovv]=k;
10865: }
1.227 brouard 10866: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10867: if(Tvard[k1][2] <=ncovcol){
10868: Fixed[k]= 1;
10869: Dummy[k]= 1;
10870: modell[k].maintype= VTYPE;
10871: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10872: ncovv++; /* Varying variables without age */
10873: TvarV[ncovv]=Tvar[k];
10874: TvarVind[ncovv]=k;
10875: }else if(Tvard[k1][2] <=ncovcol+nqv){
10876: Fixed[k]= 1;
10877: Dummy[k]= 1;
10878: modell[k].maintype= VTYPE;
10879: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10880: ncovv++; /* Varying variables without age */
10881: TvarV[ncovv]=Tvar[k];
10882: TvarVind[ncovv]=k;
10883: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10884: Fixed[k]= 1;
10885: Dummy[k]= 0;
10886: modell[k].maintype= VTYPE;
10887: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10888: ncovv++; /* Varying variables without age */
10889: TvarV[ncovv]=Tvar[k];
10890: TvarVind[ncovv]=k;
10891: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10892: Fixed[k]= 1;
10893: Dummy[k]= 1;
10894: modell[k].maintype= VTYPE;
10895: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10896: ncovv++; /* Varying variables without age */
10897: TvarV[ncovv]=Tvar[k];
10898: TvarVind[ncovv]=k;
10899: }
1.227 brouard 10900: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10901: if(Tvard[k1][2] <=ncovcol){
10902: Fixed[k]= 1;
10903: Dummy[k]= 1;
10904: modell[k].maintype= VTYPE;
10905: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10906: ncovv++; /* Varying variables without age */
10907: TvarV[ncovv]=Tvar[k];
10908: TvarVind[ncovv]=k;
10909: }else if(Tvard[k1][2] <=ncovcol+nqv){
10910: Fixed[k]= 1;
10911: Dummy[k]= 1;
10912: modell[k].maintype= VTYPE;
10913: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10914: ncovv++; /* Varying variables without age */
10915: TvarV[ncovv]=Tvar[k];
10916: TvarVind[ncovv]=k;
10917: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10918: Fixed[k]= 1;
10919: Dummy[k]= 1;
10920: modell[k].maintype= VTYPE;
10921: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10922: ncovv++; /* Varying variables without age */
10923: TvarV[ncovv]=Tvar[k];
10924: TvarVind[ncovv]=k;
10925: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10926: Fixed[k]= 1;
10927: Dummy[k]= 1;
10928: modell[k].maintype= VTYPE;
10929: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10930: ncovv++; /* Varying variables without age */
10931: TvarV[ncovv]=Tvar[k];
10932: TvarVind[ncovv]=k;
10933: }
1.227 brouard 10934: }else{
1.240 brouard 10935: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10936: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10937: } /*end k1*/
1.225 brouard 10938: }else{
1.226 brouard 10939: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10940: 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 10941: }
1.227 brouard 10942: 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 10943: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10944: 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]);
10945: }
10946: /* Searching for doublons in the model */
10947: for(k1=1; k1<= cptcovt;k1++){
10948: for(k2=1; k2 <k1;k2++){
1.285 brouard 10949: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10950: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10951: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10952: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10953: 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]);
10954: 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 10955: return(1);
10956: }
10957: }else if (Typevar[k1] ==2){
10958: k3=Tposprod[k1];
10959: k4=Tposprod[k2];
10960: 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])) ){
10961: 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]]);
10962: 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);
10963: return(1);
10964: }
10965: }
1.227 brouard 10966: }
10967: }
1.225 brouard 10968: }
10969: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10970: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10971: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10972: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10973: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10974: /*endread:*/
1.225 brouard 10975: printf("Exiting decodemodel: ");
10976: return (1);
1.136 brouard 10977: }
10978:
1.169 brouard 10979: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10980: {/* Check ages at death */
1.136 brouard 10981: int i, m;
1.218 brouard 10982: int firstone=0;
10983:
1.136 brouard 10984: for (i=1; i<=imx; i++) {
10985: for(m=2; (m<= maxwav); m++) {
10986: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10987: anint[m][i]=9999;
1.216 brouard 10988: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10989: s[m][i]=-1;
1.136 brouard 10990: }
10991: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10992: *nberr = *nberr + 1;
1.218 brouard 10993: if(firstone == 0){
10994: firstone=1;
1.260 brouard 10995: 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 10996: }
1.262 brouard 10997: 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 10998: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10999: }
11000: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 11001: (*nberr)++;
1.259 brouard 11002: 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 11003: 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 11004: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 11005: }
11006: }
11007: }
11008:
11009: for (i=1; i<=imx; i++) {
11010: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
11011: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 11012: 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 11013: if (s[m][i] >= nlstate+1) {
1.169 brouard 11014: if(agedc[i]>0){
11015: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 11016: agev[m][i]=agedc[i];
1.214 brouard 11017: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 11018: }else {
1.136 brouard 11019: if ((int)andc[i]!=9999){
11020: nbwarn++;
11021: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
11022: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
11023: agev[m][i]=-1;
11024: }
11025: }
1.169 brouard 11026: } /* agedc > 0 */
1.214 brouard 11027: } /* end if */
1.136 brouard 11028: else if(s[m][i] !=9){ /* Standard case, age in fractional
11029: years but with the precision of a month */
11030: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
11031: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
11032: agev[m][i]=1;
11033: else if(agev[m][i] < *agemin){
11034: *agemin=agev[m][i];
11035: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
11036: }
11037: else if(agev[m][i] >*agemax){
11038: *agemax=agev[m][i];
1.156 brouard 11039: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 11040: }
11041: /*agev[m][i]=anint[m][i]-annais[i];*/
11042: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 11043: } /* en if 9*/
1.136 brouard 11044: else { /* =9 */
1.214 brouard 11045: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 11046: agev[m][i]=1;
11047: s[m][i]=-1;
11048: }
11049: }
1.214 brouard 11050: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 11051: agev[m][i]=1;
1.214 brouard 11052: else{
11053: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11054: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
11055: agev[m][i]=0;
11056: }
11057: } /* End for lastpass */
11058: }
1.136 brouard 11059:
11060: for (i=1; i<=imx; i++) {
11061: for(m=firstpass; (m<=lastpass); m++){
11062: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 11063: (*nberr)++;
1.136 brouard 11064: 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);
11065: 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);
11066: return 1;
11067: }
11068: }
11069: }
11070:
11071: /*for (i=1; i<=imx; i++){
11072: for (m=firstpass; (m<lastpass); m++){
11073: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
11074: }
11075:
11076: }*/
11077:
11078:
1.139 brouard 11079: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
11080: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 11081:
11082: return (0);
1.164 brouard 11083: /* endread:*/
1.136 brouard 11084: printf("Exiting calandcheckages: ");
11085: return (1);
11086: }
11087:
1.172 brouard 11088: #if defined(_MSC_VER)
11089: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11090: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
11091: //#include "stdafx.h"
11092: //#include <stdio.h>
11093: //#include <tchar.h>
11094: //#include <windows.h>
11095: //#include <iostream>
11096: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
11097:
11098: LPFN_ISWOW64PROCESS fnIsWow64Process;
11099:
11100: BOOL IsWow64()
11101: {
11102: BOOL bIsWow64 = FALSE;
11103:
11104: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
11105: // (HANDLE, PBOOL);
11106:
11107: //LPFN_ISWOW64PROCESS fnIsWow64Process;
11108:
11109: HMODULE module = GetModuleHandle(_T("kernel32"));
11110: const char funcName[] = "IsWow64Process";
11111: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
11112: GetProcAddress(module, funcName);
11113:
11114: if (NULL != fnIsWow64Process)
11115: {
11116: if (!fnIsWow64Process(GetCurrentProcess(),
11117: &bIsWow64))
11118: //throw std::exception("Unknown error");
11119: printf("Unknown error\n");
11120: }
11121: return bIsWow64 != FALSE;
11122: }
11123: #endif
1.177 brouard 11124:
1.191 brouard 11125: void syscompilerinfo(int logged)
1.292 brouard 11126: {
11127: #include <stdint.h>
11128:
11129: /* #include "syscompilerinfo.h"*/
1.185 brouard 11130: /* command line Intel compiler 32bit windows, XP compatible:*/
11131: /* /GS /W3 /Gy
11132: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
11133: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
11134: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 11135: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
11136: */
11137: /* 64 bits */
1.185 brouard 11138: /*
11139: /GS /W3 /Gy
11140: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
11141: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
11142: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
11143: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
11144: /* Optimization are useless and O3 is slower than O2 */
11145: /*
11146: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
11147: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
11148: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
11149: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
11150: */
1.186 brouard 11151: /* Link is */ /* /OUT:"visual studio
1.185 brouard 11152: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
11153: /PDB:"visual studio
11154: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
11155: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
11156: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
11157: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
11158: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
11159: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
11160: uiAccess='false'"
11161: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
11162: /NOLOGO /TLBID:1
11163: */
1.292 brouard 11164:
11165:
1.177 brouard 11166: #if defined __INTEL_COMPILER
1.178 brouard 11167: #if defined(__GNUC__)
11168: struct utsname sysInfo; /* For Intel on Linux and OS/X */
11169: #endif
1.177 brouard 11170: #elif defined(__GNUC__)
1.179 brouard 11171: #ifndef __APPLE__
1.174 brouard 11172: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 11173: #endif
1.177 brouard 11174: struct utsname sysInfo;
1.178 brouard 11175: int cross = CROSS;
11176: if (cross){
11177: printf("Cross-");
1.191 brouard 11178: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 11179: }
1.174 brouard 11180: #endif
11181:
1.191 brouard 11182: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 11183: #if defined(__clang__)
1.191 brouard 11184: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 11185: #endif
11186: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 11187: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 11188: #endif
11189: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 11190: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 11191: #endif
11192: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 11193: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 11194: #endif
11195: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 11196: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 11197: #endif
11198: #if defined(_MSC_VER)
1.191 brouard 11199: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 11200: #endif
11201: #if defined(__PGI)
1.191 brouard 11202: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 11203: #endif
11204: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 11205: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 11206: #endif
1.191 brouard 11207: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 11208:
1.167 brouard 11209: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
11210: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
11211: // Windows (x64 and x86)
1.191 brouard 11212: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 11213: #elif __unix__ // all unices, not all compilers
11214: // Unix
1.191 brouard 11215: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 11216: #elif __linux__
11217: // linux
1.191 brouard 11218: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 11219: #elif __APPLE__
1.174 brouard 11220: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 11221: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 11222: #endif
11223:
11224: /* __MINGW32__ */
11225: /* __CYGWIN__ */
11226: /* __MINGW64__ */
11227: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
11228: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
11229: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
11230: /* _WIN64 // Defined for applications for Win64. */
11231: /* _M_X64 // Defined for compilations that target x64 processors. */
11232: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 11233:
1.167 brouard 11234: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 11235: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 11236: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 11237: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 11238: #else
1.191 brouard 11239: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 11240: #endif
11241:
1.169 brouard 11242: #if defined(__GNUC__)
11243: # if defined(__GNUC_PATCHLEVEL__)
11244: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11245: + __GNUC_MINOR__ * 100 \
11246: + __GNUC_PATCHLEVEL__)
11247: # else
11248: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
11249: + __GNUC_MINOR__ * 100)
11250: # endif
1.174 brouard 11251: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 11252: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 11253:
11254: if (uname(&sysInfo) != -1) {
11255: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 11256: 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 11257: }
11258: else
11259: perror("uname() error");
1.179 brouard 11260: //#ifndef __INTEL_COMPILER
11261: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 11262: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 11263: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 11264: #endif
1.169 brouard 11265: #endif
1.172 brouard 11266:
1.286 brouard 11267: // void main ()
1.172 brouard 11268: // {
1.169 brouard 11269: #if defined(_MSC_VER)
1.174 brouard 11270: if (IsWow64()){
1.191 brouard 11271: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
11272: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 11273: }
11274: else{
1.191 brouard 11275: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
11276: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 11277: }
1.172 brouard 11278: // printf("\nPress Enter to continue...");
11279: // getchar();
11280: // }
11281:
1.169 brouard 11282: #endif
11283:
1.167 brouard 11284:
1.219 brouard 11285: }
1.136 brouard 11286:
1.219 brouard 11287: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 11288: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.332 brouard 11289: /* Computes the prevalence limit for each combination of the dummy covariates */
1.235 brouard 11290: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 11291: /* double ftolpl = 1.e-10; */
1.180 brouard 11292: double age, agebase, agelim;
1.203 brouard 11293: double tot;
1.180 brouard 11294:
1.202 brouard 11295: strcpy(filerespl,"PL_");
11296: strcat(filerespl,fileresu);
11297: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 11298: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
11299: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 11300: }
1.288 brouard 11301: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
11302: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 11303: pstamp(ficrespl);
1.288 brouard 11304: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 11305: fprintf(ficrespl,"#Age ");
11306: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
11307: fprintf(ficrespl,"\n");
1.180 brouard 11308:
1.219 brouard 11309: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 11310:
1.219 brouard 11311: agebase=ageminpar;
11312: agelim=agemaxpar;
1.180 brouard 11313:
1.227 brouard 11314: /* i1=pow(2,ncoveff); */
1.234 brouard 11315: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 11316: if (cptcovn < 1){i1=1;}
1.180 brouard 11317:
1.238 brouard 11318: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
11319: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 11320: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11321: continue;
1.235 brouard 11322:
1.238 brouard 11323: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11324: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
11325: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
11326: /* k=k+1; */
11327: /* to clean */
1.332 brouard 11328: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11329: fprintf(ficrespl,"#******");
11330: printf("#******");
11331: fprintf(ficlog,"#******");
11332: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
1.332 brouard 11333: /* fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,Tvaraff[j])]); /\* Here problem for varying dummy*\/ */
11334: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]); /* Here problem for varying dummy*/
11335: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11336: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11337: }
11338: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11339: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11340: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11341: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11342: }
11343: fprintf(ficrespl,"******\n");
11344: printf("******\n");
11345: fprintf(ficlog,"******\n");
11346: if(invalidvarcomb[k]){
11347: printf("\nCombination (%d) ignored because no case \n",k);
11348: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
11349: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
11350: continue;
11351: }
1.219 brouard 11352:
1.238 brouard 11353: fprintf(ficrespl,"#Age ");
11354: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 11355: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11356: }
11357: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
11358: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 11359:
1.238 brouard 11360: for (age=agebase; age<=agelim; age++){
11361: /* for (age=agebase; age<=agebase; age++){ */
11362: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
11363: fprintf(ficrespl,"%.0f ",age );
11364: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11365: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11366: tot=0.;
11367: for(i=1; i<=nlstate;i++){
11368: tot += prlim[i][i];
11369: fprintf(ficrespl," %.5f", prlim[i][i]);
11370: }
11371: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
11372: } /* Age */
11373: /* was end of cptcod */
11374: } /* cptcov */
11375: } /* nres */
1.219 brouard 11376: return 0;
1.180 brouard 11377: }
11378:
1.218 brouard 11379: 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 11380: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 11381:
11382: /* Computes the back prevalence limit for any combination of covariate values
11383: * at any age between ageminpar and agemaxpar
11384: */
1.235 brouard 11385: int i, j, k, i1, nres=0 ;
1.217 brouard 11386: /* double ftolpl = 1.e-10; */
11387: double age, agebase, agelim;
11388: double tot;
1.218 brouard 11389: /* double ***mobaverage; */
11390: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 11391:
11392: strcpy(fileresplb,"PLB_");
11393: strcat(fileresplb,fileresu);
11394: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 11395: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
11396: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 11397: }
1.288 brouard 11398: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
11399: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 11400: pstamp(ficresplb);
1.288 brouard 11401: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 11402: fprintf(ficresplb,"#Age ");
11403: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
11404: fprintf(ficresplb,"\n");
11405:
1.218 brouard 11406:
11407: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
11408:
11409: agebase=ageminpar;
11410: agelim=agemaxpar;
11411:
11412:
1.227 brouard 11413: i1=pow(2,cptcoveff);
1.218 brouard 11414: if (cptcovn < 1){i1=1;}
1.227 brouard 11415:
1.238 brouard 11416: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11417: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11418: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11419: continue;
1.332 brouard 11420: /*printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));*/
1.238 brouard 11421: fprintf(ficresplb,"#******");
11422: printf("#******");
11423: fprintf(ficlog,"#******");
11424: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
1.332 brouard 11425: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11426: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
11427: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11428: }
11429: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 11430: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
11431: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
11432: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 11433: }
11434: fprintf(ficresplb,"******\n");
11435: printf("******\n");
11436: fprintf(ficlog,"******\n");
11437: if(invalidvarcomb[k]){
11438: printf("\nCombination (%d) ignored because no cases \n",k);
11439: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
11440: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
11441: continue;
11442: }
1.218 brouard 11443:
1.238 brouard 11444: fprintf(ficresplb,"#Age ");
11445: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 11446: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11447: }
11448: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
11449: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 11450:
11451:
1.238 brouard 11452: for (age=agebase; age<=agelim; age++){
11453: /* for (age=agebase; age<=agebase; age++){ */
11454: if(mobilavproj > 0){
11455: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
11456: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11457: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 11458: }else if (mobilavproj == 0){
11459: 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);
11460: 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);
11461: exit(1);
11462: }else{
11463: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 11464: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 11465: /* printf("TOTOT\n"); */
11466: /* exit(1); */
1.238 brouard 11467: }
11468: fprintf(ficresplb,"%.0f ",age );
11469: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11470: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11471: tot=0.;
11472: for(i=1; i<=nlstate;i++){
11473: tot += bprlim[i][i];
11474: fprintf(ficresplb," %.5f", bprlim[i][i]);
11475: }
11476: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
11477: } /* Age */
11478: /* was end of cptcod */
1.255 brouard 11479: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 11480: } /* end of any combination */
11481: } /* end of nres */
1.218 brouard 11482: /* hBijx(p, bage, fage); */
11483: /* fclose(ficrespijb); */
11484:
11485: return 0;
1.217 brouard 11486: }
1.218 brouard 11487:
1.180 brouard 11488: int hPijx(double *p, int bage, int fage){
11489: /*------------- h Pij x at various ages ------------*/
11490:
11491: int stepsize;
11492: int agelim;
11493: int hstepm;
11494: int nhstepm;
1.235 brouard 11495: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 11496:
11497: double agedeb;
11498: double ***p3mat;
11499:
1.201 brouard 11500: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 11501: if((ficrespij=fopen(filerespij,"w"))==NULL) {
11502: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
11503: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
11504: }
11505: printf("Computing pij: result on file '%s' \n", filerespij);
11506: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
11507:
11508: stepsize=(int) (stepm+YEARM-1)/YEARM;
11509: /*if (stepm<=24) stepsize=2;*/
11510:
11511: agelim=AGESUP;
11512: hstepm=stepsize*YEARM; /* Every year of age */
11513: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 11514:
1.180 brouard 11515: /* hstepm=1; aff par mois*/
11516: pstamp(ficrespij);
11517: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 11518: i1= pow(2,cptcoveff);
1.218 brouard 11519: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11520: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11521: /* k=k+1; */
1.235 brouard 11522: for(nres=1; nres <= nresult; nres++) /* For each resultline */
11523: for(k=1; k<=i1;k++){
1.253 brouard 11524: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 11525: continue;
1.183 brouard 11526: fprintf(ficrespij,"\n#****** ");
1.227 brouard 11527: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11528: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 11529: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
11530: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11531: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
11532: }
1.183 brouard 11533: fprintf(ficrespij,"******\n");
11534:
11535: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
11536: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
11537: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
11538:
11539: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 11540:
1.183 brouard 11541: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11542: oldm=oldms;savm=savms;
1.235 brouard 11543: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 11544: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
11545: for(i=1; i<=nlstate;i++)
11546: for(j=1; j<=nlstate+ndeath;j++)
11547: fprintf(ficrespij," %1d-%1d",i,j);
11548: fprintf(ficrespij,"\n");
11549: for (h=0; h<=nhstepm; h++){
11550: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11551: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 11552: for(i=1; i<=nlstate;i++)
11553: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 11554: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 11555: fprintf(ficrespij,"\n");
11556: }
1.183 brouard 11557: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11558: fprintf(ficrespij,"\n");
11559: }
1.180 brouard 11560: /*}*/
11561: }
1.218 brouard 11562: return 0;
1.180 brouard 11563: }
1.218 brouard 11564:
11565: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 11566: /*------------- h Bij x at various ages ------------*/
11567:
11568: int stepsize;
1.218 brouard 11569: /* int agelim; */
11570: int ageminl;
1.217 brouard 11571: int hstepm;
11572: int nhstepm;
1.238 brouard 11573: int h, i, i1, j, k, nres;
1.218 brouard 11574:
1.217 brouard 11575: double agedeb;
11576: double ***p3mat;
1.218 brouard 11577:
11578: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
11579: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
11580: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11581: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
11582: }
11583: printf("Computing pij back: result on file '%s' \n", filerespijb);
11584: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
11585:
11586: stepsize=(int) (stepm+YEARM-1)/YEARM;
11587: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 11588:
1.218 brouard 11589: /* agelim=AGESUP; */
1.289 brouard 11590: ageminl=AGEINF; /* was 30 */
1.218 brouard 11591: hstepm=stepsize*YEARM; /* Every year of age */
11592: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
11593:
11594: /* hstepm=1; aff par mois*/
11595: pstamp(ficrespijb);
1.255 brouard 11596: 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 11597: i1= pow(2,cptcoveff);
1.218 brouard 11598: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
11599: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
11600: /* k=k+1; */
1.238 brouard 11601: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
11602: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 11603: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 11604: continue;
11605: fprintf(ficrespijb,"\n#****** ");
11606: for(j=1;j<=cptcoveff;j++)
1.332 brouard 11607: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.238 brouard 11608: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 11609: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.238 brouard 11610: }
11611: fprintf(ficrespijb,"******\n");
1.264 brouard 11612: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 11613: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
11614: continue;
11615: }
11616:
11617: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
11618: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
11619: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 11620: 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 */
11621: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 11622:
11623: /* nhstepm=nhstepm*YEARM; aff par mois*/
11624:
1.266 brouard 11625: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
11626: /* and memory limitations if stepm is small */
11627:
1.238 brouard 11628: /* oldm=oldms;savm=savms; */
11629: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.325 brouard 11630: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);/* Bug valgrind */
1.238 brouard 11631: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 11632: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 11633: for(i=1; i<=nlstate;i++)
11634: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 11635: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 11636: fprintf(ficrespijb,"\n");
1.238 brouard 11637: for (h=0; h<=nhstepm; h++){
11638: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
11639: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
11640: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
11641: for(i=1; i<=nlstate;i++)
11642: for(j=1; j<=nlstate+ndeath;j++)
1.325 brouard 11643: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);/* Bug valgrind */
1.238 brouard 11644: fprintf(ficrespijb,"\n");
11645: }
11646: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
11647: fprintf(ficrespijb,"\n");
11648: } /* end age deb */
11649: } /* end combination */
11650: } /* end nres */
1.218 brouard 11651: return 0;
11652: } /* hBijx */
1.217 brouard 11653:
1.180 brouard 11654:
1.136 brouard 11655: /***********************************************/
11656: /**************** Main Program *****************/
11657: /***********************************************/
11658:
11659: int main(int argc, char *argv[])
11660: {
11661: #ifdef GSL
11662: const gsl_multimin_fminimizer_type *T;
11663: size_t iteri = 0, it;
11664: int rval = GSL_CONTINUE;
11665: int status = GSL_SUCCESS;
11666: double ssval;
11667: #endif
11668: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 11669: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
11670: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 11671: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 11672: int jj, ll, li, lj, lk;
1.136 brouard 11673: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 11674: int num_filled;
1.136 brouard 11675: int itimes;
11676: int NDIM=2;
11677: int vpopbased=0;
1.235 brouard 11678: int nres=0;
1.258 brouard 11679: int endishere=0;
1.277 brouard 11680: int noffset=0;
1.274 brouard 11681: int ncurrv=0; /* Temporary variable */
11682:
1.164 brouard 11683: char ca[32], cb[32];
1.136 brouard 11684: /* FILE *fichtm; *//* Html File */
11685: /* FILE *ficgp;*/ /*Gnuplot File */
11686: struct stat info;
1.191 brouard 11687: double agedeb=0.;
1.194 brouard 11688:
11689: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 11690: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 11691:
1.165 brouard 11692: double fret;
1.191 brouard 11693: double dum=0.; /* Dummy variable */
1.136 brouard 11694: double ***p3mat;
1.218 brouard 11695: /* double ***mobaverage; */
1.319 brouard 11696: double wald;
1.164 brouard 11697:
11698: char line[MAXLINE];
1.197 brouard 11699: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
11700:
1.234 brouard 11701: char modeltemp[MAXLINE];
1.332 brouard 11702: char resultline[MAXLINE], resultlineori[MAXLINE];
1.230 brouard 11703:
1.136 brouard 11704: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11705: char *tok, *val; /* pathtot */
1.290 brouard 11706: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 11707: int c, h , cpt, c2;
1.191 brouard 11708: int jl=0;
11709: int i1, j1, jk, stepsize=0;
1.194 brouard 11710: int count=0;
11711:
1.164 brouard 11712: int *tab;
1.136 brouard 11713: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11714: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11715: /* double anprojf, mprojf, jprojf; */
11716: /* double jintmean,mintmean,aintmean; */
11717: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11718: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11719: double yrfproj= 10.0; /* Number of years of forward projections */
11720: double yrbproj= 10.0; /* Number of years of backward projections */
11721: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11722: int mobilav=0,popforecast=0;
1.191 brouard 11723: int hstepm=0, nhstepm=0;
1.136 brouard 11724: int agemortsup;
11725: float sumlpop=0.;
11726: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11727: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11728:
1.191 brouard 11729: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11730: double ftolpl=FTOL;
11731: double **prlim;
1.217 brouard 11732: double **bprlim;
1.317 brouard 11733: double ***param; /* Matrix of parameters, param[i][j][k] param=ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel)
11734: state of origin, state of destination including death, for each covariate: constante, age, and V1 V2 etc. */
1.251 brouard 11735: double ***paramstart; /* Matrix of starting parameter values */
11736: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11737: double **matcov; /* Matrix of covariance */
1.203 brouard 11738: double **hess; /* Hessian matrix */
1.136 brouard 11739: double ***delti3; /* Scale */
11740: double *delti; /* Scale */
11741: double ***eij, ***vareij;
11742: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11743:
1.136 brouard 11744: double *epj, vepp;
1.164 brouard 11745:
1.273 brouard 11746: double dateprev1, dateprev2;
1.296 brouard 11747: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11748: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11749:
1.217 brouard 11750:
1.136 brouard 11751: double **ximort;
1.145 brouard 11752: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11753: int *dcwave;
11754:
1.164 brouard 11755: char z[1]="c";
1.136 brouard 11756:
11757: /*char *strt;*/
11758: char strtend[80];
1.126 brouard 11759:
1.164 brouard 11760:
1.126 brouard 11761: /* setlocale (LC_ALL, ""); */
11762: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11763: /* textdomain (PACKAGE); */
11764: /* setlocale (LC_CTYPE, ""); */
11765: /* setlocale (LC_MESSAGES, ""); */
11766:
11767: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11768: rstart_time = time(NULL);
11769: /* (void) gettimeofday(&start_time,&tzp);*/
11770: start_time = *localtime(&rstart_time);
1.126 brouard 11771: curr_time=start_time;
1.157 brouard 11772: /*tml = *localtime(&start_time.tm_sec);*/
11773: /* strcpy(strstart,asctime(&tml)); */
11774: strcpy(strstart,asctime(&start_time));
1.126 brouard 11775:
11776: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11777: /* tp.tm_sec = tp.tm_sec +86400; */
11778: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11779: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11780: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11781: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11782: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11783: /* strt=asctime(&tmg); */
11784: /* printf("Time(after) =%s",strstart); */
11785: /* (void) time (&time_value);
11786: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11787: * tm = *localtime(&time_value);
11788: * strstart=asctime(&tm);
11789: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11790: */
11791:
11792: nberr=0; /* Number of errors and warnings */
11793: nbwarn=0;
1.184 brouard 11794: #ifdef WIN32
11795: _getcwd(pathcd, size);
11796: #else
1.126 brouard 11797: getcwd(pathcd, size);
1.184 brouard 11798: #endif
1.191 brouard 11799: syscompilerinfo(0);
1.196 brouard 11800: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11801: if(argc <=1){
11802: printf("\nEnter the parameter file name: ");
1.205 brouard 11803: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11804: printf("ERROR Empty parameter file name\n");
11805: goto end;
11806: }
1.126 brouard 11807: i=strlen(pathr);
11808: if(pathr[i-1]=='\n')
11809: pathr[i-1]='\0';
1.156 brouard 11810: i=strlen(pathr);
1.205 brouard 11811: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11812: pathr[i-1]='\0';
1.205 brouard 11813: }
11814: i=strlen(pathr);
11815: if( i==0 ){
11816: printf("ERROR Empty parameter file name\n");
11817: goto end;
11818: }
11819: for (tok = pathr; tok != NULL; ){
1.126 brouard 11820: printf("Pathr |%s|\n",pathr);
11821: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11822: printf("val= |%s| pathr=%s\n",val,pathr);
11823: strcpy (pathtot, val);
11824: if(pathr[0] == '\0') break; /* Dirty */
11825: }
11826: }
1.281 brouard 11827: else if (argc<=2){
11828: strcpy(pathtot,argv[1]);
11829: }
1.126 brouard 11830: else{
11831: strcpy(pathtot,argv[1]);
1.281 brouard 11832: strcpy(z,argv[2]);
11833: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11834: }
11835: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11836: /*cygwin_split_path(pathtot,path,optionfile);
11837: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11838: /* cutv(path,optionfile,pathtot,'\\');*/
11839:
11840: /* Split argv[0], imach program to get pathimach */
11841: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11842: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11843: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11844: /* strcpy(pathimach,argv[0]); */
11845: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11846: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11847: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11848: #ifdef WIN32
11849: _chdir(path); /* Can be a relative path */
11850: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11851: #else
1.126 brouard 11852: chdir(path); /* Can be a relative path */
1.184 brouard 11853: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11854: #endif
11855: printf("Current directory %s!\n",pathcd);
1.126 brouard 11856: strcpy(command,"mkdir ");
11857: strcat(command,optionfilefiname);
11858: if((outcmd=system(command)) != 0){
1.169 brouard 11859: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11860: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11861: /* fclose(ficlog); */
11862: /* exit(1); */
11863: }
11864: /* if((imk=mkdir(optionfilefiname))<0){ */
11865: /* perror("mkdir"); */
11866: /* } */
11867:
11868: /*-------- arguments in the command line --------*/
11869:
1.186 brouard 11870: /* Main Log file */
1.126 brouard 11871: strcat(filelog, optionfilefiname);
11872: strcat(filelog,".log"); /* */
11873: if((ficlog=fopen(filelog,"w"))==NULL) {
11874: printf("Problem with logfile %s\n",filelog);
11875: goto end;
11876: }
11877: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11878: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11879: fprintf(ficlog,"\nEnter the parameter file name: \n");
11880: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11881: path=%s \n\
11882: optionfile=%s\n\
11883: optionfilext=%s\n\
1.156 brouard 11884: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11885:
1.197 brouard 11886: syscompilerinfo(1);
1.167 brouard 11887:
1.126 brouard 11888: printf("Local time (at start):%s",strstart);
11889: fprintf(ficlog,"Local time (at start): %s",strstart);
11890: fflush(ficlog);
11891: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11892: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11893:
11894: /* */
11895: strcpy(fileres,"r");
11896: strcat(fileres, optionfilefiname);
1.201 brouard 11897: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11898: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11899: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11900:
1.186 brouard 11901: /* Main ---------arguments file --------*/
1.126 brouard 11902:
11903: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11904: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11905: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11906: fflush(ficlog);
1.149 brouard 11907: /* goto end; */
11908: exit(70);
1.126 brouard 11909: }
11910:
11911: strcpy(filereso,"o");
1.201 brouard 11912: strcat(filereso,fileresu);
1.126 brouard 11913: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11914: printf("Problem with Output resultfile: %s\n", filereso);
11915: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11916: fflush(ficlog);
11917: goto end;
11918: }
1.278 brouard 11919: /*-------- Rewriting parameter file ----------*/
11920: strcpy(rfileres,"r"); /* "Rparameterfile */
11921: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11922: strcat(rfileres,"."); /* */
11923: strcat(rfileres,optionfilext); /* Other files have txt extension */
11924: if((ficres =fopen(rfileres,"w"))==NULL) {
11925: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11926: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11927: fflush(ficlog);
11928: goto end;
11929: }
11930: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11931:
1.278 brouard 11932:
1.126 brouard 11933: /* Reads comments: lines beginning with '#' */
11934: numlinepar=0;
1.277 brouard 11935: /* Is it a BOM UTF-8 Windows file? */
11936: /* First parameter line */
1.197 brouard 11937: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11938: noffset=0;
11939: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11940: {
11941: noffset=noffset+3;
11942: printf("# File is an UTF8 Bom.\n"); // 0xBF
11943: }
1.302 brouard 11944: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11945: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11946: {
11947: noffset=noffset+2;
11948: printf("# File is an UTF16BE BOM file\n");
11949: }
11950: else if( line[0] == 0 && line[1] == 0)
11951: {
11952: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11953: noffset=noffset+4;
11954: printf("# File is an UTF16BE BOM file\n");
11955: }
11956: } else{
11957: ;/*printf(" Not a BOM file\n");*/
11958: }
11959:
1.197 brouard 11960: /* If line starts with a # it is a comment */
1.277 brouard 11961: if (line[noffset] == '#') {
1.197 brouard 11962: numlinepar++;
11963: fputs(line,stdout);
11964: fputs(line,ficparo);
1.278 brouard 11965: fputs(line,ficres);
1.197 brouard 11966: fputs(line,ficlog);
11967: continue;
11968: }else
11969: break;
11970: }
11971: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11972: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11973: if (num_filled != 5) {
11974: printf("Should be 5 parameters\n");
1.283 brouard 11975: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11976: }
1.126 brouard 11977: numlinepar++;
1.197 brouard 11978: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11979: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11980: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11981: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11982: }
11983: /* Second parameter line */
11984: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11985: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11986: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11987: if (line[0] == '#') {
11988: numlinepar++;
1.283 brouard 11989: printf("%s",line);
11990: fprintf(ficres,"%s",line);
11991: fprintf(ficparo,"%s",line);
11992: fprintf(ficlog,"%s",line);
1.197 brouard 11993: continue;
11994: }else
11995: break;
11996: }
1.223 brouard 11997: 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", \
11998: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11999: if (num_filled != 11) {
12000: 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 12001: printf("but line=%s\n",line);
1.283 brouard 12002: 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");
12003: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 12004: }
1.286 brouard 12005: if( lastpass > maxwav){
12006: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12007: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
12008: fflush(ficlog);
12009: goto end;
12010: }
12011: 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 12012: 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 12013: 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 12014: 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 12015: }
1.203 brouard 12016: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 12017: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 12018: /* Third parameter line */
12019: while(fgets(line, MAXLINE, ficpar)) {
12020: /* If line starts with a # it is a comment */
12021: if (line[0] == '#') {
12022: numlinepar++;
1.283 brouard 12023: printf("%s",line);
12024: fprintf(ficres,"%s",line);
12025: fprintf(ficparo,"%s",line);
12026: fprintf(ficlog,"%s",line);
1.197 brouard 12027: continue;
12028: }else
12029: break;
12030: }
1.201 brouard 12031: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 12032: if (num_filled != 1){
1.302 brouard 12033: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
12034: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 12035: model[0]='\0';
12036: goto end;
12037: }
12038: else{
12039: if (model[0]=='+'){
12040: for(i=1; i<=strlen(model);i++)
12041: modeltemp[i-1]=model[i];
1.201 brouard 12042: strcpy(model,modeltemp);
1.197 brouard 12043: }
12044: }
1.199 brouard 12045: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 12046: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 12047: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
12048: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
12049: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 12050: }
12051: /* 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); */
12052: /* numlinepar=numlinepar+3; /\* In general *\/ */
12053: /* 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 12054: /* 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); */
12055: /* 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 12056: fflush(ficlog);
1.190 brouard 12057: /* if(model[0]=='#'|| model[0]== '\0'){ */
12058: if(model[0]=='#'){
1.279 brouard 12059: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
12060: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
12061: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 12062: if(mle != -1){
1.279 brouard 12063: 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 12064: exit(1);
12065: }
12066: }
1.126 brouard 12067: while((c=getc(ficpar))=='#' && c!= EOF){
12068: ungetc(c,ficpar);
12069: fgets(line, MAXLINE, ficpar);
12070: numlinepar++;
1.195 brouard 12071: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
12072: z[0]=line[1];
12073: }
12074: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 12075: fputs(line, stdout);
12076: //puts(line);
1.126 brouard 12077: fputs(line,ficparo);
12078: fputs(line,ficlog);
12079: }
12080: ungetc(c,ficpar);
12081:
12082:
1.290 brouard 12083: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
12084: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
12085: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
12086: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 12087: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
12088: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
12089: v1+v2*age+v2*v3 makes cptcovn = 3
12090: */
12091: if (strlen(model)>1)
1.187 brouard 12092: 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 12093: else
1.187 brouard 12094: ncovmodel=2; /* Constant and age */
1.133 brouard 12095: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
12096: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 12097: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
12098: 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);
12099: 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);
12100: fflush(stdout);
12101: fclose (ficlog);
12102: goto end;
12103: }
1.126 brouard 12104: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12105: delti=delti3[1][1];
12106: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
12107: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 12108: /* We could also provide initial parameters values giving by simple logistic regression
12109: * only one way, that is without matrix product. We will have nlstate maximizations */
12110: /* for(i=1;i<nlstate;i++){ */
12111: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12112: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12113: /* } */
1.126 brouard 12114: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 12115: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
12116: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12117: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
12118: fclose (ficparo);
12119: fclose (ficlog);
12120: goto end;
12121: exit(0);
1.220 brouard 12122: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 12123: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 12124: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
12125: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 12126: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
12127: matcov=matrix(1,npar,1,npar);
1.203 brouard 12128: hess=matrix(1,npar,1,npar);
1.220 brouard 12129: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 12130: /* Read guessed parameters */
1.126 brouard 12131: /* Reads comments: lines beginning with '#' */
12132: while((c=getc(ficpar))=='#' && c!= EOF){
12133: ungetc(c,ficpar);
12134: fgets(line, MAXLINE, ficpar);
12135: numlinepar++;
1.141 brouard 12136: fputs(line,stdout);
1.126 brouard 12137: fputs(line,ficparo);
12138: fputs(line,ficlog);
12139: }
12140: ungetc(c,ficpar);
12141:
12142: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 12143: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 12144: for(i=1; i <=nlstate; i++){
1.234 brouard 12145: j=0;
1.126 brouard 12146: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 12147: if(jj==i) continue;
12148: j++;
1.292 brouard 12149: while((c=getc(ficpar))=='#' && c!= EOF){
12150: ungetc(c,ficpar);
12151: fgets(line, MAXLINE, ficpar);
12152: numlinepar++;
12153: fputs(line,stdout);
12154: fputs(line,ficparo);
12155: fputs(line,ficlog);
12156: }
12157: ungetc(c,ficpar);
1.234 brouard 12158: fscanf(ficpar,"%1d%1d",&i1,&j1);
12159: if ((i1 != i) || (j1 != jj)){
12160: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 12161: It might be a problem of design; if ncovcol and the model are correct\n \
12162: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 12163: exit(1);
12164: }
12165: fprintf(ficparo,"%1d%1d",i1,j1);
12166: if(mle==1)
12167: printf("%1d%1d",i,jj);
12168: fprintf(ficlog,"%1d%1d",i,jj);
12169: for(k=1; k<=ncovmodel;k++){
12170: fscanf(ficpar," %lf",¶m[i][j][k]);
12171: if(mle==1){
12172: printf(" %lf",param[i][j][k]);
12173: fprintf(ficlog," %lf",param[i][j][k]);
12174: }
12175: else
12176: fprintf(ficlog," %lf",param[i][j][k]);
12177: fprintf(ficparo," %lf",param[i][j][k]);
12178: }
12179: fscanf(ficpar,"\n");
12180: numlinepar++;
12181: if(mle==1)
12182: printf("\n");
12183: fprintf(ficlog,"\n");
12184: fprintf(ficparo,"\n");
1.126 brouard 12185: }
12186: }
12187: fflush(ficlog);
1.234 brouard 12188:
1.251 brouard 12189: /* Reads parameters values */
1.126 brouard 12190: p=param[1][1];
1.251 brouard 12191: pstart=paramstart[1][1];
1.126 brouard 12192:
12193: /* Reads comments: lines beginning with '#' */
12194: while((c=getc(ficpar))=='#' && c!= EOF){
12195: ungetc(c,ficpar);
12196: fgets(line, MAXLINE, ficpar);
12197: numlinepar++;
1.141 brouard 12198: fputs(line,stdout);
1.126 brouard 12199: fputs(line,ficparo);
12200: fputs(line,ficlog);
12201: }
12202: ungetc(c,ficpar);
12203:
12204: for(i=1; i <=nlstate; i++){
12205: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 12206: fscanf(ficpar,"%1d%1d",&i1,&j1);
12207: if ( (i1-i) * (j1-j) != 0){
12208: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
12209: exit(1);
12210: }
12211: printf("%1d%1d",i,j);
12212: fprintf(ficparo,"%1d%1d",i1,j1);
12213: fprintf(ficlog,"%1d%1d",i1,j1);
12214: for(k=1; k<=ncovmodel;k++){
12215: fscanf(ficpar,"%le",&delti3[i][j][k]);
12216: printf(" %le",delti3[i][j][k]);
12217: fprintf(ficparo," %le",delti3[i][j][k]);
12218: fprintf(ficlog," %le",delti3[i][j][k]);
12219: }
12220: fscanf(ficpar,"\n");
12221: numlinepar++;
12222: printf("\n");
12223: fprintf(ficparo,"\n");
12224: fprintf(ficlog,"\n");
1.126 brouard 12225: }
12226: }
12227: fflush(ficlog);
1.234 brouard 12228:
1.145 brouard 12229: /* Reads covariance matrix */
1.126 brouard 12230: delti=delti3[1][1];
1.220 brouard 12231:
12232:
1.126 brouard 12233: /* 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 12234:
1.126 brouard 12235: /* Reads comments: lines beginning with '#' */
12236: while((c=getc(ficpar))=='#' && c!= EOF){
12237: ungetc(c,ficpar);
12238: fgets(line, MAXLINE, ficpar);
12239: numlinepar++;
1.141 brouard 12240: fputs(line,stdout);
1.126 brouard 12241: fputs(line,ficparo);
12242: fputs(line,ficlog);
12243: }
12244: ungetc(c,ficpar);
1.220 brouard 12245:
1.126 brouard 12246: matcov=matrix(1,npar,1,npar);
1.203 brouard 12247: hess=matrix(1,npar,1,npar);
1.131 brouard 12248: for(i=1; i <=npar; i++)
12249: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 12250:
1.194 brouard 12251: /* Scans npar lines */
1.126 brouard 12252: for(i=1; i <=npar; i++){
1.226 brouard 12253: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 12254: if(count != 3){
1.226 brouard 12255: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12256: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12257: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12258: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 12259: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
12260: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 12261: exit(1);
1.220 brouard 12262: }else{
1.226 brouard 12263: if(mle==1)
12264: printf("%1d%1d%d",i1,j1,jk);
12265: }
12266: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
12267: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 12268: for(j=1; j <=i; j++){
1.226 brouard 12269: fscanf(ficpar," %le",&matcov[i][j]);
12270: if(mle==1){
12271: printf(" %.5le",matcov[i][j]);
12272: }
12273: fprintf(ficlog," %.5le",matcov[i][j]);
12274: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 12275: }
12276: fscanf(ficpar,"\n");
12277: numlinepar++;
12278: if(mle==1)
1.220 brouard 12279: printf("\n");
1.126 brouard 12280: fprintf(ficlog,"\n");
12281: fprintf(ficparo,"\n");
12282: }
1.194 brouard 12283: /* End of read covariance matrix npar lines */
1.126 brouard 12284: for(i=1; i <=npar; i++)
12285: for(j=i+1;j<=npar;j++)
1.226 brouard 12286: matcov[i][j]=matcov[j][i];
1.126 brouard 12287:
12288: if(mle==1)
12289: printf("\n");
12290: fprintf(ficlog,"\n");
12291:
12292: fflush(ficlog);
12293:
12294: } /* End of mle != -3 */
1.218 brouard 12295:
1.186 brouard 12296: /* Main data
12297: */
1.290 brouard 12298: nobs=lastobs-firstobs+1; /* was = lastobs;*/
12299: /* num=lvector(1,n); */
12300: /* moisnais=vector(1,n); */
12301: /* annais=vector(1,n); */
12302: /* moisdc=vector(1,n); */
12303: /* andc=vector(1,n); */
12304: /* weight=vector(1,n); */
12305: /* agedc=vector(1,n); */
12306: /* cod=ivector(1,n); */
12307: /* for(i=1;i<=n;i++){ */
12308: num=lvector(firstobs,lastobs);
12309: moisnais=vector(firstobs,lastobs);
12310: annais=vector(firstobs,lastobs);
12311: moisdc=vector(firstobs,lastobs);
12312: andc=vector(firstobs,lastobs);
12313: weight=vector(firstobs,lastobs);
12314: agedc=vector(firstobs,lastobs);
12315: cod=ivector(firstobs,lastobs);
12316: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 12317: num[i]=0;
12318: moisnais[i]=0;
12319: annais[i]=0;
12320: moisdc[i]=0;
12321: andc[i]=0;
12322: agedc[i]=0;
12323: cod[i]=0;
12324: weight[i]=1.0; /* Equal weights, 1 by default */
12325: }
1.290 brouard 12326: mint=matrix(1,maxwav,firstobs,lastobs);
12327: anint=matrix(1,maxwav,firstobs,lastobs);
1.325 brouard 12328: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
12329: printf("BUG ncovmodel=%d NCOVMAX=%d 2**ncovmodel=%f BUG\n",ncovmodel,NCOVMAX,pow(2,ncovmodel));
1.126 brouard 12330: tab=ivector(1,NCOVMAX);
1.144 brouard 12331: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 12332: 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 12333:
1.136 brouard 12334: /* Reads data from file datafile */
12335: if (readdata(datafile, firstobs, lastobs, &imx)==1)
12336: goto end;
12337:
12338: /* Calculation of the number of parameters from char model */
1.234 brouard 12339: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 12340: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
12341: k=3 V4 Tvar[k=3]= 4 (from V4)
12342: k=2 V1 Tvar[k=2]= 1 (from V1)
12343: k=1 Tvar[1]=2 (from V2)
1.234 brouard 12344: */
12345:
12346: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
12347: TvarsDind=ivector(1,NCOVMAX); /* */
1.330 brouard 12348: TnsdVar=ivector(1,NCOVMAX); /* */
1.234 brouard 12349: TvarsD=ivector(1,NCOVMAX); /* */
12350: TvarsQind=ivector(1,NCOVMAX); /* */
12351: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 12352: TvarF=ivector(1,NCOVMAX); /* */
12353: TvarFind=ivector(1,NCOVMAX); /* */
12354: TvarV=ivector(1,NCOVMAX); /* */
12355: TvarVind=ivector(1,NCOVMAX); /* */
12356: TvarA=ivector(1,NCOVMAX); /* */
12357: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 12358: TvarFD=ivector(1,NCOVMAX); /* */
12359: TvarFDind=ivector(1,NCOVMAX); /* */
12360: TvarFQ=ivector(1,NCOVMAX); /* */
12361: TvarFQind=ivector(1,NCOVMAX); /* */
12362: TvarVD=ivector(1,NCOVMAX); /* */
12363: TvarVDind=ivector(1,NCOVMAX); /* */
12364: TvarVQ=ivector(1,NCOVMAX); /* */
12365: TvarVQind=ivector(1,NCOVMAX); /* */
12366:
1.230 brouard 12367: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 12368: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 12369: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
12370: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
12371: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 12372: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
12373: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
12374: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
12375: */
12376: /* For model-covariate k tells which data-covariate to use but
12377: because this model-covariate is a construction we invent a new column
12378: ncovcol + k1
12379: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
12380: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 12381: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
12382: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 12383: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
12384: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 12385: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 12386: */
1.145 brouard 12387: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
12388: 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 12389: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
12390: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.330 brouard 12391: Tvardk=imatrix(1,NCOVMAX,1,2);
1.145 brouard 12392: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 12393: 4 covariates (3 plus signs)
12394: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
1.328 brouard 12395: */
12396: for(i=1;i<NCOVMAX;i++)
12397: Tage[i]=0;
1.230 brouard 12398: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 12399: * individual dummy, fixed or varying:
12400: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
12401: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 12402: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
12403: * V1 df, V2 qf, V3 & V4 dv, V5 qv
12404: * Tmodelind[1]@9={9,0,3,2,}*/
12405: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
12406: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 12407: * individual quantitative, fixed or varying:
12408: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
12409: * 3, 1, 0, 0, 0, 0, 0, 0},
12410: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 12411: /* Main decodemodel */
12412:
1.187 brouard 12413:
1.223 brouard 12414: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 12415: goto end;
12416:
1.137 brouard 12417: if((double)(lastobs-imx)/(double)imx > 1.10){
12418: nbwarn++;
12419: 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);
12420: 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);
12421: }
1.136 brouard 12422: /* if(mle==1){*/
1.137 brouard 12423: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
12424: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 12425: }
12426:
12427: /*-calculation of age at interview from date of interview and age at death -*/
12428: agev=matrix(1,maxwav,1,imx);
12429:
12430: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
12431: goto end;
12432:
1.126 brouard 12433:
1.136 brouard 12434: agegomp=(int)agemin;
1.290 brouard 12435: free_vector(moisnais,firstobs,lastobs);
12436: free_vector(annais,firstobs,lastobs);
1.126 brouard 12437: /* free_matrix(mint,1,maxwav,1,n);
12438: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 12439: /* free_vector(moisdc,1,n); */
12440: /* free_vector(andc,1,n); */
1.145 brouard 12441: /* */
12442:
1.126 brouard 12443: wav=ivector(1,imx);
1.214 brouard 12444: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
12445: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
12446: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
12447: 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.*/
12448: bh=imatrix(1,lastpass-firstpass+2,1,imx);
12449: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 12450:
12451: /* Concatenates waves */
1.214 brouard 12452: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
12453: Death is a valid wave (if date is known).
12454: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
12455: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
12456: and mw[mi+1][i]. dh depends on stepm.
12457: */
12458:
1.126 brouard 12459: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 12460: /* Concatenates waves */
1.145 brouard 12461:
1.290 brouard 12462: free_vector(moisdc,firstobs,lastobs);
12463: free_vector(andc,firstobs,lastobs);
1.215 brouard 12464:
1.126 brouard 12465: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
12466: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
12467: ncodemax[1]=1;
1.145 brouard 12468: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 12469: cptcoveff=0;
1.220 brouard 12470: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
12471: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 12472: }
12473:
12474: ncovcombmax=pow(2,cptcoveff);
12475: invalidvarcomb=ivector(1, ncovcombmax);
12476: for(i=1;i<ncovcombmax;i++)
12477: invalidvarcomb[i]=0;
12478:
1.211 brouard 12479: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 12480: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 12481: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 12482:
1.200 brouard 12483: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 12484: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 12485: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 12486: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
12487: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
12488: * (currently 0 or 1) in the data.
12489: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
12490: * corresponding modality (h,j).
12491: */
12492:
1.145 brouard 12493: h=0;
12494: /*if (cptcovn > 0) */
1.126 brouard 12495: m=pow(2,cptcoveff);
12496:
1.144 brouard 12497: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 12498: * For k=4 covariates, h goes from 1 to m=2**k
12499: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
12500: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.329 brouard 12501: * h\k 1 2 3 4 * h-1\k-1 4 3 2 1
12502: *______________________________ *______________________
12503: * 1 i=1 1 i=1 1 i=1 1 i=1 1 * 0 0 0 0 0
12504: * 2 2 1 1 1 * 1 0 0 0 1
12505: * 3 i=2 1 2 1 1 * 2 0 0 1 0
12506: * 4 2 2 1 1 * 3 0 0 1 1
12507: * 5 i=3 1 i=2 1 2 1 * 4 0 1 0 0
12508: * 6 2 1 2 1 * 5 0 1 0 1
12509: * 7 i=4 1 2 2 1 * 6 0 1 1 0
12510: * 8 2 2 2 1 * 7 0 1 1 1
12511: * 9 i=5 1 i=3 1 i=2 1 2 * 8 1 0 0 0
12512: * 10 2 1 1 2 * 9 1 0 0 1
12513: * 11 i=6 1 2 1 2 * 10 1 0 1 0
12514: * 12 2 2 1 2 * 11 1 0 1 1
12515: * 13 i=7 1 i=4 1 2 2 * 12 1 1 0 0
12516: * 14 2 1 2 2 * 13 1 1 0 1
12517: * 15 i=8 1 2 2 2 * 14 1 1 1 0
12518: * 16 2 2 2 2 * 15 1 1 1 1
12519: */
1.212 brouard 12520: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 12521: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
12522: * and the value of each covariate?
12523: * V1=1, V2=1, V3=2, V4=1 ?
12524: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
12525: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
12526: * In order to get the real value in the data, we use nbcode
12527: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
12528: * We are keeping this crazy system in order to be able (in the future?)
12529: * to have more than 2 values (0 or 1) for a covariate.
12530: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
12531: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
12532: * bbbbbbbb
12533: * 76543210
12534: * h-1 00000101 (6-1=5)
1.219 brouard 12535: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 12536: * &
12537: * 1 00000001 (1)
1.219 brouard 12538: * 00000000 = 1 & ((h-1) >> (k-1))
12539: * +1= 00000001 =1
1.211 brouard 12540: *
12541: * h=14, k=3 => h'=h-1=13, k'=k-1=2
12542: * h' 1101 =2^3+2^2+0x2^1+2^0
12543: * >>k' 11
12544: * & 00000001
12545: * = 00000001
12546: * +1 = 00000010=2 = codtabm(14,3)
12547: * Reverse h=6 and m=16?
12548: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
12549: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
12550: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
12551: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
12552: * V3=decodtabm(14,3,2**4)=2
12553: * h'=13 1101 =2^3+2^2+0x2^1+2^0
12554: *(h-1) >> (j-1) 0011 =13 >> 2
12555: * &1 000000001
12556: * = 000000001
12557: * +1= 000000010 =2
12558: * 2211
12559: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
12560: * V3=2
1.220 brouard 12561: * codtabm and decodtabm are identical
1.211 brouard 12562: */
12563:
1.145 brouard 12564:
12565: free_ivector(Ndum,-1,NCOVMAX);
12566:
12567:
1.126 brouard 12568:
1.186 brouard 12569: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 12570: strcpy(optionfilegnuplot,optionfilefiname);
12571: if(mle==-3)
1.201 brouard 12572: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 12573: strcat(optionfilegnuplot,".gp");
12574:
12575: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
12576: printf("Problem with file %s",optionfilegnuplot);
12577: }
12578: else{
1.204 brouard 12579: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 12580: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 12581: //fprintf(ficgp,"set missing 'NaNq'\n");
12582: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 12583: }
12584: /* fclose(ficgp);*/
1.186 brouard 12585:
12586:
12587: /* Initialisation of --------- index.htm --------*/
1.126 brouard 12588:
12589: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
12590: if(mle==-3)
1.201 brouard 12591: strcat(optionfilehtm,"-MORT_");
1.126 brouard 12592: strcat(optionfilehtm,".htm");
12593: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 12594: printf("Problem with %s \n",optionfilehtm);
12595: exit(0);
1.126 brouard 12596: }
12597:
12598: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
12599: strcat(optionfilehtmcov,"-cov.htm");
12600: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
12601: printf("Problem with %s \n",optionfilehtmcov), exit(0);
12602: }
12603: else{
12604: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
12605: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12606: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 12607: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
12608: }
12609:
1.332 brouard 12610: 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-2013-2022-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 12611: <hr size=\"2\" color=\"#EC5E5E\"> \n\
12612: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 12613: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 12614: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 12615: \n\
12616: <hr size=\"2\" color=\"#EC5E5E\">\
12617: <ul><li><h4>Parameter files</h4>\n\
12618: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
12619: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
12620: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
12621: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
12622: - Date and time at start: %s</ul>\n",\
12623: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
12624: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
12625: fileres,fileres,\
12626: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
12627: fflush(fichtm);
12628:
12629: strcpy(pathr,path);
12630: strcat(pathr,optionfilefiname);
1.184 brouard 12631: #ifdef WIN32
12632: _chdir(optionfilefiname); /* Move to directory named optionfile */
12633: #else
1.126 brouard 12634: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 12635: #endif
12636:
1.126 brouard 12637:
1.220 brouard 12638: /* Calculates basic frequencies. Computes observed prevalence at single age
12639: and for any valid combination of covariates
1.126 brouard 12640: and prints on file fileres'p'. */
1.251 brouard 12641: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 12642: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 12643:
12644: fprintf(fichtm,"\n");
1.286 brouard 12645: 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 12646: ftol, stepm);
12647: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
12648: ncurrv=1;
12649: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
12650: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
12651: ncurrv=i;
12652: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12653: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 12654: ncurrv=i;
12655: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 12656: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 12657: ncurrv=i;
12658: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
12659: 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", \
12660: nlstate, ndeath, maxwav, mle, weightopt);
12661:
12662: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
12663: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
12664:
12665:
1.317 brouard 12666: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Number of (used) observations=%d <br>\n\
1.126 brouard 12667: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
12668: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 12669: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 12670: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 12671: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12672: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12673: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
12674: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 12675:
1.126 brouard 12676: /* For Powell, parameters are in a vector p[] starting at p[1]
12677: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
12678: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
12679:
12680: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 12681: /* For mortality only */
1.126 brouard 12682: if (mle==-3){
1.136 brouard 12683: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 12684: for(i=1;i<=NDIM;i++)
12685: for(j=1;j<=NDIM;j++)
12686: ximort[i][j]=0.;
1.186 brouard 12687: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 12688: cens=ivector(firstobs,lastobs);
12689: ageexmed=vector(firstobs,lastobs);
12690: agecens=vector(firstobs,lastobs);
12691: dcwave=ivector(firstobs,lastobs);
1.223 brouard 12692:
1.126 brouard 12693: for (i=1; i<=imx; i++){
12694: dcwave[i]=-1;
12695: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 12696: if (s[m][i]>nlstate) {
12697: dcwave[i]=m;
12698: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
12699: break;
12700: }
1.126 brouard 12701: }
1.226 brouard 12702:
1.126 brouard 12703: for (i=1; i<=imx; i++) {
12704: if (wav[i]>0){
1.226 brouard 12705: ageexmed[i]=agev[mw[1][i]][i];
12706: j=wav[i];
12707: agecens[i]=1.;
12708:
12709: if (ageexmed[i]> 1 && wav[i] > 0){
12710: agecens[i]=agev[mw[j][i]][i];
12711: cens[i]= 1;
12712: }else if (ageexmed[i]< 1)
12713: cens[i]= -1;
12714: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12715: cens[i]=0 ;
1.126 brouard 12716: }
12717: else cens[i]=-1;
12718: }
12719:
12720: for (i=1;i<=NDIM;i++) {
12721: for (j=1;j<=NDIM;j++)
1.226 brouard 12722: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12723: }
12724:
1.302 brouard 12725: p[1]=0.0268; p[NDIM]=0.083;
12726: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12727:
12728:
1.136 brouard 12729: #ifdef GSL
12730: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12731: #else
1.126 brouard 12732: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12733: #endif
1.201 brouard 12734: strcpy(filerespow,"POW-MORT_");
12735: strcat(filerespow,fileresu);
1.126 brouard 12736: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12737: printf("Problem with resultfile: %s\n", filerespow);
12738: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12739: }
1.136 brouard 12740: #ifdef GSL
12741: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12742: #else
1.126 brouard 12743: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12744: #endif
1.126 brouard 12745: /* for (i=1;i<=nlstate;i++)
12746: for(j=1;j<=nlstate+ndeath;j++)
12747: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12748: */
12749: fprintf(ficrespow,"\n");
1.136 brouard 12750: #ifdef GSL
12751: /* gsl starts here */
12752: T = gsl_multimin_fminimizer_nmsimplex;
12753: gsl_multimin_fminimizer *sfm = NULL;
12754: gsl_vector *ss, *x;
12755: gsl_multimin_function minex_func;
12756:
12757: /* Initial vertex size vector */
12758: ss = gsl_vector_alloc (NDIM);
12759:
12760: if (ss == NULL){
12761: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12762: }
12763: /* Set all step sizes to 1 */
12764: gsl_vector_set_all (ss, 0.001);
12765:
12766: /* Starting point */
1.126 brouard 12767:
1.136 brouard 12768: x = gsl_vector_alloc (NDIM);
12769:
12770: if (x == NULL){
12771: gsl_vector_free(ss);
12772: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12773: }
12774:
12775: /* Initialize method and iterate */
12776: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12777: /* gsl_vector_set(x, 0, 0.0268); */
12778: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12779: gsl_vector_set(x, 0, p[1]);
12780: gsl_vector_set(x, 1, p[2]);
12781:
12782: minex_func.f = &gompertz_f;
12783: minex_func.n = NDIM;
12784: minex_func.params = (void *)&p; /* ??? */
12785:
12786: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12787: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12788:
12789: printf("Iterations beginning .....\n\n");
12790: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12791:
12792: iteri=0;
12793: while (rval == GSL_CONTINUE){
12794: iteri++;
12795: status = gsl_multimin_fminimizer_iterate(sfm);
12796:
12797: if (status) printf("error: %s\n", gsl_strerror (status));
12798: fflush(0);
12799:
12800: if (status)
12801: break;
12802:
12803: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12804: ssval = gsl_multimin_fminimizer_size (sfm);
12805:
12806: if (rval == GSL_SUCCESS)
12807: printf ("converged to a local maximum at\n");
12808:
12809: printf("%5d ", iteri);
12810: for (it = 0; it < NDIM; it++){
12811: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12812: }
12813: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12814: }
12815:
12816: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12817:
12818: gsl_vector_free(x); /* initial values */
12819: gsl_vector_free(ss); /* inital step size */
12820: for (it=0; it<NDIM; it++){
12821: p[it+1]=gsl_vector_get(sfm->x,it);
12822: fprintf(ficrespow," %.12lf", p[it]);
12823: }
12824: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12825: #endif
12826: #ifdef POWELL
12827: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12828: #endif
1.126 brouard 12829: fclose(ficrespow);
12830:
1.203 brouard 12831: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12832:
12833: for(i=1; i <=NDIM; i++)
12834: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12835: matcov[i][j]=matcov[j][i];
1.126 brouard 12836:
12837: printf("\nCovariance matrix\n ");
1.203 brouard 12838: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12839: for(i=1; i <=NDIM; i++) {
12840: for(j=1;j<=NDIM;j++){
1.220 brouard 12841: printf("%f ",matcov[i][j]);
12842: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12843: }
1.203 brouard 12844: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12845: }
12846:
12847: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12848: for (i=1;i<=NDIM;i++) {
1.126 brouard 12849: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12850: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12851: }
1.302 brouard 12852: lsurv=vector(agegomp,AGESUP);
12853: lpop=vector(agegomp,AGESUP);
12854: tpop=vector(agegomp,AGESUP);
1.126 brouard 12855: lsurv[agegomp]=100000;
12856:
12857: for (k=agegomp;k<=AGESUP;k++) {
12858: agemortsup=k;
12859: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12860: }
12861:
12862: for (k=agegomp;k<agemortsup;k++)
12863: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12864:
12865: for (k=agegomp;k<agemortsup;k++){
12866: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12867: sumlpop=sumlpop+lpop[k];
12868: }
12869:
12870: tpop[agegomp]=sumlpop;
12871: for (k=agegomp;k<(agemortsup-3);k++){
12872: /* tpop[k+1]=2;*/
12873: tpop[k+1]=tpop[k]-lpop[k];
12874: }
12875:
12876:
12877: printf("\nAge lx qx dx Lx Tx e(x)\n");
12878: for (k=agegomp;k<(agemortsup-2);k++)
12879: 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]);
12880:
12881:
12882: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12883: ageminpar=50;
12884: agemaxpar=100;
1.194 brouard 12885: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12886: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12887: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12888: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12889: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12890: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12891: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12892: }else{
12893: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12894: 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 12895: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12896: }
1.201 brouard 12897: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12898: stepm, weightopt,\
12899: model,imx,p,matcov,agemortsup);
12900:
1.302 brouard 12901: free_vector(lsurv,agegomp,AGESUP);
12902: free_vector(lpop,agegomp,AGESUP);
12903: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12904: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12905: free_ivector(dcwave,firstobs,lastobs);
12906: free_vector(agecens,firstobs,lastobs);
12907: free_vector(ageexmed,firstobs,lastobs);
12908: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12909: #ifdef GSL
1.136 brouard 12910: #endif
1.186 brouard 12911: } /* Endof if mle==-3 mortality only */
1.205 brouard 12912: /* Standard */
12913: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12914: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12915: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12916: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12917: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12918: for (k=1; k<=npar;k++)
12919: printf(" %d %8.5f",k,p[k]);
12920: printf("\n");
1.205 brouard 12921: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12922: /* mlikeli uses func not funcone */
1.247 brouard 12923: /* for(i=1;i<nlstate;i++){ */
12924: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12925: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12926: /* } */
1.205 brouard 12927: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12928: }
12929: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12930: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12931: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12932: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12933: }
12934: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12935: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12936: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12937: for (k=1; k<=npar;k++)
12938: printf(" %d %8.5f",k,p[k]);
12939: printf("\n");
12940:
12941: /*--------- results files --------------*/
1.283 brouard 12942: /* 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 12943:
12944:
12945: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12946: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n"); /* Printing model equation */
1.126 brouard 12947: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
1.319 brouard 12948:
12949: printf("#model= 1 + age ");
12950: fprintf(ficres,"#model= 1 + age ");
12951: fprintf(ficlog,"#model= 1 + age ");
12952: fprintf(fichtm,"\n<ul><li> model=1+age+%s\n \
12953: </ul>", model);
12954:
12955: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">\n");
12956: fprintf(fichtm, "<tr><th>Model=</th><th>1</th><th>+ age</th>");
12957: if(nagesqr==1){
12958: printf(" + age*age ");
12959: fprintf(ficres," + age*age ");
12960: fprintf(ficlog," + age*age ");
12961: fprintf(fichtm, "<th>+ age*age</th>");
12962: }
12963: for(j=1;j <=ncovmodel-2;j++){
12964: if(Typevar[j]==0) {
12965: printf(" + V%d ",Tvar[j]);
12966: fprintf(ficres," + V%d ",Tvar[j]);
12967: fprintf(ficlog," + V%d ",Tvar[j]);
12968: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
12969: }else if(Typevar[j]==1) {
12970: printf(" + V%d*age ",Tvar[j]);
12971: fprintf(ficres," + V%d*age ",Tvar[j]);
12972: fprintf(ficlog," + V%d*age ",Tvar[j]);
12973: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
12974: }else if(Typevar[j]==2) {
12975: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12976: fprintf(ficres," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12977: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12978: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
12979: }
12980: }
12981: printf("\n");
12982: fprintf(ficres,"\n");
12983: fprintf(ficlog,"\n");
12984: fprintf(fichtm, "</tr>");
12985: fprintf(fichtm, "\n");
12986:
12987:
1.126 brouard 12988: for(i=1,jk=1; i <=nlstate; i++){
12989: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12990: if (k != i) {
1.319 brouard 12991: fprintf(fichtm, "<tr>");
1.225 brouard 12992: printf("%d%d ",i,k);
12993: fprintf(ficlog,"%d%d ",i,k);
12994: fprintf(ficres,"%1d%1d ",i,k);
1.319 brouard 12995: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 12996: for(j=1; j <=ncovmodel; j++){
12997: printf("%12.7f ",p[jk]);
12998: fprintf(ficlog,"%12.7f ",p[jk]);
12999: fprintf(ficres,"%12.7f ",p[jk]);
1.319 brouard 13000: fprintf(fichtm, "<td>%12.7f</td>",p[jk]);
1.225 brouard 13001: jk++;
13002: }
13003: printf("\n");
13004: fprintf(ficlog,"\n");
13005: fprintf(ficres,"\n");
1.319 brouard 13006: fprintf(fichtm, "</tr>\n");
1.225 brouard 13007: }
1.126 brouard 13008: }
13009: }
1.319 brouard 13010: /* fprintf(fichtm,"</tr>\n"); */
13011: fprintf(fichtm,"</table>\n");
13012: fprintf(fichtm, "\n");
13013:
1.203 brouard 13014: if(mle != 0){
13015: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 13016: ftolhess=ftol; /* Usually correct */
1.203 brouard 13017: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
13018: 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");
13019: 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 13020: 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 13021: fprintf(fichtm,"\n<table style=\"text-align:center; border: 1px solid\">");
13022: fprintf(fichtm, "\n<tr><th>Model=</th><th>1</th><th>+ age</th>");
13023: if(nagesqr==1){
13024: printf(" + age*age ");
13025: fprintf(ficres," + age*age ");
13026: fprintf(ficlog," + age*age ");
13027: fprintf(fichtm, "<th>+ age*age</th>");
13028: }
13029: for(j=1;j <=ncovmodel-2;j++){
13030: if(Typevar[j]==0) {
13031: printf(" + V%d ",Tvar[j]);
13032: fprintf(fichtm, "<th>+ V%d</th>",Tvar[j]);
13033: }else if(Typevar[j]==1) {
13034: printf(" + V%d*age ",Tvar[j]);
13035: fprintf(fichtm, "<th>+ V%d*age</th>",Tvar[j]);
13036: }else if(Typevar[j]==2) {
13037: fprintf(fichtm, "<th>+ V%d*V%d</th>",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
13038: }
13039: }
13040: fprintf(fichtm, "</tr>\n");
13041:
1.203 brouard 13042: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 13043: for(k=1; k <=(nlstate+ndeath); k++){
13044: if (k != i) {
1.319 brouard 13045: fprintf(fichtm, "<tr valign=top>");
1.225 brouard 13046: printf("%d%d ",i,k);
13047: fprintf(ficlog,"%d%d ",i,k);
1.319 brouard 13048: fprintf(fichtm, "<td>%1d%1d</td>",i,k);
1.225 brouard 13049: for(j=1; j <=ncovmodel; j++){
1.319 brouard 13050: wald=p[jk]/sqrt(matcov[jk][jk]);
1.324 brouard 13051: 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]));
13052: 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 13053: if(fabs(wald) > 1.96){
1.321 brouard 13054: fprintf(fichtm, "<td><b>%12.7f</b></br> (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
1.319 brouard 13055: }else{
13056: fprintf(fichtm, "<td>%12.7f (%12.7f)</br>",p[jk],sqrt(matcov[jk][jk]));
13057: }
1.324 brouard 13058: fprintf(fichtm,"W=%8.3f</br>",wald);
1.319 brouard 13059: 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 13060: jk++;
13061: }
13062: printf("\n");
13063: fprintf(ficlog,"\n");
1.319 brouard 13064: fprintf(fichtm, "</tr>\n");
1.225 brouard 13065: }
13066: }
1.193 brouard 13067: }
1.203 brouard 13068: } /* end of hesscov and Wald tests */
1.319 brouard 13069: fprintf(fichtm,"</table>\n");
1.225 brouard 13070:
1.203 brouard 13071: /* */
1.126 brouard 13072: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
13073: printf("# Scales (for hessian or gradient estimation)\n");
13074: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
13075: for(i=1,jk=1; i <=nlstate; i++){
13076: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 13077: if (j!=i) {
13078: fprintf(ficres,"%1d%1d",i,j);
13079: printf("%1d%1d",i,j);
13080: fprintf(ficlog,"%1d%1d",i,j);
13081: for(k=1; k<=ncovmodel;k++){
13082: printf(" %.5e",delti[jk]);
13083: fprintf(ficlog," %.5e",delti[jk]);
13084: fprintf(ficres," %.5e",delti[jk]);
13085: jk++;
13086: }
13087: printf("\n");
13088: fprintf(ficlog,"\n");
13089: fprintf(ficres,"\n");
13090: }
1.126 brouard 13091: }
13092: }
13093:
13094: 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 13095: if(mle >= 1) /* To big for the screen */
1.126 brouard 13096: 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");
13097: 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");
13098: /* # 121 Var(a12)\n\ */
13099: /* # 122 Cov(b12,a12) Var(b12)\n\ */
13100: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
13101: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
13102: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
13103: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
13104: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
13105: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
13106:
13107:
13108: /* Just to have a covariance matrix which will be more understandable
13109: even is we still don't want to manage dictionary of variables
13110: */
13111: for(itimes=1;itimes<=2;itimes++){
13112: jj=0;
13113: for(i=1; i <=nlstate; i++){
1.225 brouard 13114: for(j=1; j <=nlstate+ndeath; j++){
13115: if(j==i) continue;
13116: for(k=1; k<=ncovmodel;k++){
13117: jj++;
13118: ca[0]= k+'a'-1;ca[1]='\0';
13119: if(itimes==1){
13120: if(mle>=1)
13121: printf("#%1d%1d%d",i,j,k);
13122: fprintf(ficlog,"#%1d%1d%d",i,j,k);
13123: fprintf(ficres,"#%1d%1d%d",i,j,k);
13124: }else{
13125: if(mle>=1)
13126: printf("%1d%1d%d",i,j,k);
13127: fprintf(ficlog,"%1d%1d%d",i,j,k);
13128: fprintf(ficres,"%1d%1d%d",i,j,k);
13129: }
13130: ll=0;
13131: for(li=1;li <=nlstate; li++){
13132: for(lj=1;lj <=nlstate+ndeath; lj++){
13133: if(lj==li) continue;
13134: for(lk=1;lk<=ncovmodel;lk++){
13135: ll++;
13136: if(ll<=jj){
13137: cb[0]= lk +'a'-1;cb[1]='\0';
13138: if(ll<jj){
13139: if(itimes==1){
13140: if(mle>=1)
13141: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13142: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13143: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
13144: }else{
13145: if(mle>=1)
13146: printf(" %.5e",matcov[jj][ll]);
13147: fprintf(ficlog," %.5e",matcov[jj][ll]);
13148: fprintf(ficres," %.5e",matcov[jj][ll]);
13149: }
13150: }else{
13151: if(itimes==1){
13152: if(mle>=1)
13153: printf(" Var(%s%1d%1d)",ca,i,j);
13154: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
13155: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
13156: }else{
13157: if(mle>=1)
13158: printf(" %.7e",matcov[jj][ll]);
13159: fprintf(ficlog," %.7e",matcov[jj][ll]);
13160: fprintf(ficres," %.7e",matcov[jj][ll]);
13161: }
13162: }
13163: }
13164: } /* end lk */
13165: } /* end lj */
13166: } /* end li */
13167: if(mle>=1)
13168: printf("\n");
13169: fprintf(ficlog,"\n");
13170: fprintf(ficres,"\n");
13171: numlinepar++;
13172: } /* end k*/
13173: } /*end j */
1.126 brouard 13174: } /* end i */
13175: } /* end itimes */
13176:
13177: fflush(ficlog);
13178: fflush(ficres);
1.225 brouard 13179: while(fgets(line, MAXLINE, ficpar)) {
13180: /* If line starts with a # it is a comment */
13181: if (line[0] == '#') {
13182: numlinepar++;
13183: fputs(line,stdout);
13184: fputs(line,ficparo);
13185: fputs(line,ficlog);
1.299 brouard 13186: fputs(line,ficres);
1.225 brouard 13187: continue;
13188: }else
13189: break;
13190: }
13191:
1.209 brouard 13192: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
13193: /* ungetc(c,ficpar); */
13194: /* fgets(line, MAXLINE, ficpar); */
13195: /* fputs(line,stdout); */
13196: /* fputs(line,ficparo); */
13197: /* } */
13198: /* ungetc(c,ficpar); */
1.126 brouard 13199:
13200: estepm=0;
1.209 brouard 13201: 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 13202:
13203: if (num_filled != 6) {
13204: 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);
13205: 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);
13206: goto end;
13207: }
13208: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
13209: }
13210: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
13211: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
13212:
1.209 brouard 13213: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 13214: if (estepm==0 || estepm < stepm) estepm=stepm;
13215: if (fage <= 2) {
13216: bage = ageminpar;
13217: fage = agemaxpar;
13218: }
13219:
13220: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 13221: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
13222: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 13223:
1.186 brouard 13224: /* Other stuffs, more or less useful */
1.254 brouard 13225: while(fgets(line, MAXLINE, ficpar)) {
13226: /* If line starts with a # it is a comment */
13227: if (line[0] == '#') {
13228: numlinepar++;
13229: fputs(line,stdout);
13230: fputs(line,ficparo);
13231: fputs(line,ficlog);
1.299 brouard 13232: fputs(line,ficres);
1.254 brouard 13233: continue;
13234: }else
13235: break;
13236: }
13237:
13238: 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){
13239:
13240: if (num_filled != 7) {
13241: 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);
13242: 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);
13243: goto end;
13244: }
13245: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
13246: 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);
13247: 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);
13248: 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 13249: }
1.254 brouard 13250:
13251: while(fgets(line, MAXLINE, ficpar)) {
13252: /* If line starts with a # it is a comment */
13253: if (line[0] == '#') {
13254: numlinepar++;
13255: fputs(line,stdout);
13256: fputs(line,ficparo);
13257: fputs(line,ficlog);
1.299 brouard 13258: fputs(line,ficres);
1.254 brouard 13259: continue;
13260: }else
13261: break;
1.126 brouard 13262: }
13263:
13264:
13265: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
13266: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
13267:
1.254 brouard 13268: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
13269: if (num_filled != 1) {
13270: 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);
13271: 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);
13272: goto end;
13273: }
13274: printf("pop_based=%d\n",popbased);
13275: fprintf(ficlog,"pop_based=%d\n",popbased);
13276: fprintf(ficparo,"pop_based=%d\n",popbased);
13277: fprintf(ficres,"pop_based=%d\n",popbased);
13278: }
13279:
1.258 brouard 13280: /* Results */
1.332 brouard 13281: /* Value of covariate in each resultine will be compututed (if product) and sorted according to model rank */
13282: /* It is precov[] because we need the varying age in order to compute the real cov[] of the model equation */
13283: precov=matrix(1,MAXRESULTLINESPONE,1,NCOVMAX+1);
1.307 brouard 13284: endishere=0;
1.258 brouard 13285: nresult=0;
1.308 brouard 13286: parameterline=0;
1.258 brouard 13287: do{
13288: if(!fgets(line, MAXLINE, ficpar)){
13289: endishere=1;
1.308 brouard 13290: parameterline=15;
1.258 brouard 13291: }else if (line[0] == '#') {
13292: /* If line starts with a # it is a comment */
1.254 brouard 13293: numlinepar++;
13294: fputs(line,stdout);
13295: fputs(line,ficparo);
13296: fputs(line,ficlog);
1.299 brouard 13297: fputs(line,ficres);
1.254 brouard 13298: continue;
1.258 brouard 13299: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
13300: parameterline=11;
1.296 brouard 13301: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 13302: parameterline=12;
1.307 brouard 13303: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 13304: parameterline=13;
1.307 brouard 13305: }
1.258 brouard 13306: else{
13307: parameterline=14;
1.254 brouard 13308: }
1.308 brouard 13309: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 13310: case 11:
1.296 brouard 13311: 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)){
13312: 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 13313: 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);
13314: 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);
13315: 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);
13316: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 13317: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
13318: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 13319: prvforecast = 1;
13320: }
13321: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 13322: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13323: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
13324: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 13325: prvforecast = 2;
13326: }
13327: else {
13328: 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);
13329: 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);
13330: goto end;
1.258 brouard 13331: }
1.254 brouard 13332: break;
1.258 brouard 13333: case 12:
1.296 brouard 13334: 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)){
13335: 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);
13336: 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);
13337: 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);
13338: 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);
13339: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 13340: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
13341: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 13342: prvbackcast = 1;
13343: }
13344: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 13345: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13346: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
13347: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 13348: prvbackcast = 2;
13349: }
13350: else {
13351: 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);
13352: 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);
13353: goto end;
1.258 brouard 13354: }
1.230 brouard 13355: break;
1.258 brouard 13356: case 13:
1.332 brouard 13357: num_filled=sscanf(line,"result:%[^\n]\n",resultlineori);
1.307 brouard 13358: nresult++; /* Sum of resultlines */
1.332 brouard 13359: printf("Result %d: result:%s\n",nresult, resultlineori);
13360: /* removefirstspace(&resultlineori); */
13361:
13362: if(strstr(resultlineori,"v") !=0){
13363: printf("Error. 'v' must be in upper case 'V' result: %s ",resultlineori);
13364: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultlineori);fflush(ficlog);
13365: return 1;
13366: }
13367: trimbb(resultline, resultlineori); /* Suppressing double blank in the resultline */
13368: printf("Decoderesult resultline=\"%s\" resultlineori=\"%s\"\n", resultline, resultlineori);
1.318 brouard 13369: if(nresult > MAXRESULTLINESPONE-1){
13370: 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);
13371: 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 13372: goto end;
13373: }
1.332 brouard 13374:
1.310 brouard 13375: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 brouard 13376: fprintf(ficparo,"result: %s\n",resultline);
13377: fprintf(ficres,"result: %s\n",resultline);
13378: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 13379: } else
13380: goto end;
1.307 brouard 13381: break;
13382: case 14:
13383: printf("Error: Unknown command '%s'\n",line);
13384: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 brouard 13385: if(line[0] == ' ' || line[0] == '\n'){
13386: printf("It should not be an empty line '%s'\n",line);
13387: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
13388: }
1.307 brouard 13389: if(ncovmodel >=2 && nresult==0 ){
13390: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
13391: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 13392: }
1.307 brouard 13393: /* goto end; */
13394: break;
1.308 brouard 13395: case 15:
13396: printf("End of resultlines.\n");
13397: fprintf(ficlog,"End of resultlines.\n");
13398: break;
13399: default: /* parameterline =0 */
1.307 brouard 13400: nresult=1;
13401: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 13402: } /* End switch parameterline */
13403: }while(endishere==0); /* End do */
1.126 brouard 13404:
1.230 brouard 13405: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 13406: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 13407:
13408: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 13409: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 13410: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13411: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13412: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 13413: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 13414: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
13415: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 13416: }else{
1.270 brouard 13417: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 13418: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
13419: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
13420: if(prvforecast==1){
13421: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
13422: jprojd=jproj1;
13423: mprojd=mproj1;
13424: anprojd=anproj1;
13425: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
13426: jprojf=jproj2;
13427: mprojf=mproj2;
13428: anprojf=anproj2;
13429: } else if(prvforecast == 2){
13430: dateprojd=dateintmean;
13431: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
13432: dateprojf=dateintmean+yrfproj;
13433: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
13434: }
13435: if(prvbackcast==1){
13436: datebackd=(jback1+12*mback1+365*anback1)/365;
13437: jbackd=jback1;
13438: mbackd=mback1;
13439: anbackd=anback1;
13440: datebackf=(jback2+12*mback2+365*anback2)/365;
13441: jbackf=jback2;
13442: mbackf=mback2;
13443: anbackf=anback2;
13444: } else if(prvbackcast == 2){
13445: datebackd=dateintmean;
13446: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
13447: datebackf=dateintmean-yrbproj;
13448: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
13449: }
13450:
13451: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 13452: }
13453: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 13454: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
13455: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 13456:
1.225 brouard 13457: /*------------ free_vector -------------*/
13458: /* chdir(path); */
1.220 brouard 13459:
1.215 brouard 13460: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
13461: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
13462: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
13463: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 13464: free_lvector(num,firstobs,lastobs);
13465: free_vector(agedc,firstobs,lastobs);
1.126 brouard 13466: /*free_matrix(covar,0,NCOVMAX,1,n);*/
13467: /*free_matrix(covar,1,NCOVMAX,1,n);*/
13468: fclose(ficparo);
13469: fclose(ficres);
1.220 brouard 13470:
13471:
1.186 brouard 13472: /* Other results (useful)*/
1.220 brouard 13473:
13474:
1.126 brouard 13475: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 13476: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
13477: prlim=matrix(1,nlstate,1,nlstate);
1.332 brouard 13478: /* Computes the prevalence limit for each combination k of the dummy covariates by calling prevalim(k) */
1.209 brouard 13479: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 13480: fclose(ficrespl);
13481:
13482: /*------------- h Pij x at various ages ------------*/
1.180 brouard 13483: /*#include "hpijx.h"*/
1.332 brouard 13484: /** h Pij x Probability to be in state j at age x+h being in i at x, for each combination k of dummies in the model line or to nres?*/
13485: /* calls hpxij with combination k */
1.180 brouard 13486: hPijx(p, bage, fage);
1.145 brouard 13487: fclose(ficrespij);
1.227 brouard 13488:
1.220 brouard 13489: /* ncovcombmax= pow(2,cptcoveff); */
1.332 brouard 13490: /*-------------- Variance of one-step probabilities for a combination ij or for nres ?---*/
1.145 brouard 13491: k=1;
1.126 brouard 13492: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 13493:
1.269 brouard 13494: /* Prevalence for each covariate combination in probs[age][status][cov] */
13495: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13496: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 13497: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 13498: for(k=1;k<=ncovcombmax;k++)
13499: probs[i][j][k]=0.;
1.269 brouard 13500: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
13501: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 13502: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 13503: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
13504: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 13505: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 13506: for(k=1;k<=ncovcombmax;k++)
13507: mobaverages[i][j][k]=0.;
1.219 brouard 13508: mobaverage=mobaverages;
13509: if (mobilav!=0) {
1.235 brouard 13510: printf("Movingaveraging observed prevalence\n");
1.258 brouard 13511: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 13512: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
13513: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
13514: printf(" Error in movingaverage mobilav=%d\n",mobilav);
13515: }
1.269 brouard 13516: } else if (mobilavproj !=0) {
1.235 brouard 13517: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 13518: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 13519: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
13520: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
13521: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
13522: }
1.269 brouard 13523: }else{
13524: printf("Internal error moving average\n");
13525: fflush(stdout);
13526: exit(1);
1.219 brouard 13527: }
13528: }/* end if moving average */
1.227 brouard 13529:
1.126 brouard 13530: /*---------- Forecasting ------------------*/
1.296 brouard 13531: if(prevfcast==1){
13532: /* /\* if(stepm ==1){*\/ */
13533: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13534: /*This done previously after freqsummary.*/
13535: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
13536: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
13537:
13538: /* } else if (prvforecast==2){ */
13539: /* /\* if(stepm ==1){*\/ */
13540: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
13541: /* } */
13542: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
13543: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 13544: }
1.269 brouard 13545:
1.296 brouard 13546: /* Prevbcasting */
13547: if(prevbcast==1){
1.219 brouard 13548: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13549: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13550: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
13551:
13552: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
13553:
13554: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 13555:
1.219 brouard 13556: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
13557: fclose(ficresplb);
13558:
1.222 brouard 13559: hBijx(p, bage, fage, mobaverage);
13560: fclose(ficrespijb);
1.219 brouard 13561:
1.296 brouard 13562: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
13563: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
13564: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
13565: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
13566: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
13567: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
13568:
13569:
1.269 brouard 13570: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13571:
13572:
1.269 brouard 13573: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 13574: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13575: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
13576: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 13577: } /* end Prevbcasting */
1.268 brouard 13578:
1.186 brouard 13579:
13580: /* ------ Other prevalence ratios------------ */
1.126 brouard 13581:
1.215 brouard 13582: free_ivector(wav,1,imx);
13583: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
13584: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
13585: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 13586:
13587:
1.127 brouard 13588: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 13589:
1.201 brouard 13590: strcpy(filerese,"E_");
13591: strcat(filerese,fileresu);
1.126 brouard 13592: if((ficreseij=fopen(filerese,"w"))==NULL) {
13593: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13594: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
13595: }
1.208 brouard 13596: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
13597: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 13598:
13599: pstamp(ficreseij);
1.219 brouard 13600:
1.235 brouard 13601: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13602: if (cptcovn < 1){i1=1;}
13603:
13604: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13605: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13606: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13607: continue;
1.219 brouard 13608: fprintf(ficreseij,"\n#****** ");
1.235 brouard 13609: printf("\n#****** ");
1.225 brouard 13610: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 13611: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
13612: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.235 brouard 13613: }
13614: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 13615: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
13616: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.219 brouard 13617: }
13618: fprintf(ficreseij,"******\n");
1.235 brouard 13619: printf("******\n");
1.219 brouard 13620:
13621: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13622: oldm=oldms;savm=savms;
1.330 brouard 13623: /* printf("HELLO Entering evsij bage=%d fage=%d k=%d estepm=%d nres=%d\n",(int) bage, (int)fage, k, estepm, nres); */
1.235 brouard 13624: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 13625:
1.219 brouard 13626: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 13627: }
13628: fclose(ficreseij);
1.208 brouard 13629: printf("done evsij\n");fflush(stdout);
13630: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 13631:
1.218 brouard 13632:
1.227 brouard 13633: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 13634:
1.201 brouard 13635: strcpy(filerest,"T_");
13636: strcat(filerest,fileresu);
1.127 brouard 13637: if((ficrest=fopen(filerest,"w"))==NULL) {
13638: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
13639: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
13640: }
1.208 brouard 13641: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
13642: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 13643: strcpy(fileresstde,"STDE_");
13644: strcat(fileresstde,fileresu);
1.126 brouard 13645: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 13646: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
13647: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 13648: }
1.227 brouard 13649: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
13650: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 13651:
1.201 brouard 13652: strcpy(filerescve,"CVE_");
13653: strcat(filerescve,fileresu);
1.126 brouard 13654: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 13655: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
13656: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 13657: }
1.227 brouard 13658: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
13659: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 13660:
1.201 brouard 13661: strcpy(fileresv,"V_");
13662: strcat(fileresv,fileresu);
1.126 brouard 13663: if((ficresvij=fopen(fileresv,"w"))==NULL) {
13664: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
13665: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
13666: }
1.227 brouard 13667: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
13668: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 13669:
1.235 brouard 13670: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
13671: if (cptcovn < 1){i1=1;}
13672:
13673: for(nres=1; nres <= nresult; nres++) /* For each resultline */
13674: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 13675: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 13676: continue;
1.321 brouard 13677: printf("\n# model %s \n#****** Result for:", model);
13678: fprintf(ficrest,"\n# model %s \n#****** Result for:", model);
13679: fprintf(ficlog,"\n# model %s \n#****** Result for:", model);
1.227 brouard 13680: for(j=1;j<=cptcoveff;j++){
1.332 brouard 13681: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
13682: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
13683: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.227 brouard 13684: }
1.235 brouard 13685: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 13686: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
13687: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
13688: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.235 brouard 13689: }
1.208 brouard 13690: fprintf(ficrest,"******\n");
1.227 brouard 13691: fprintf(ficlog,"******\n");
13692: printf("******\n");
1.208 brouard 13693:
13694: fprintf(ficresstdeij,"\n#****** ");
13695: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 13696: for(j=1;j<=cptcoveff;j++) {
1.332 brouard 13697: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
13698: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[Tvaraff[j]])]);
1.208 brouard 13699: }
1.235 brouard 13700: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 13701: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
13702: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]);
1.235 brouard 13703: }
1.208 brouard 13704: fprintf(ficresstdeij,"******\n");
13705: fprintf(ficrescveij,"******\n");
13706:
13707: fprintf(ficresvij,"\n#****** ");
1.238 brouard 13708: /* pstamp(ficresvij); */
1.225 brouard 13709: for(j=1;j<=cptcoveff;j++)
1.332 brouard 13710: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,TnsdVar[TnsdVar[Tvaraff[j]]])]);
1.235 brouard 13711: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
1.332 brouard 13712: /* fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]); /\* To solve *\/ */
13713: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][resultmodel[nres][j]]); /* Solved */
1.235 brouard 13714: }
1.208 brouard 13715: fprintf(ficresvij,"******\n");
13716:
13717: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13718: oldm=oldms;savm=savms;
1.235 brouard 13719: printf(" cvevsij ");
13720: fprintf(ficlog, " cvevsij ");
13721: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 13722: printf(" end cvevsij \n ");
13723: fprintf(ficlog, " end cvevsij \n ");
13724:
13725: /*
13726: */
13727: /* goto endfree; */
13728:
13729: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
13730: pstamp(ficrest);
13731:
1.269 brouard 13732: epj=vector(1,nlstate+1);
1.208 brouard 13733: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 13734: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
13735: cptcod= 0; /* To be deleted */
13736: printf("varevsij vpopbased=%d \n",vpopbased);
13737: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 13738: 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 13739: 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 ");
13740: if(vpopbased==1)
13741: 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);
13742: else
1.288 brouard 13743: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13744: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
13745: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
13746: fprintf(ficrest,"\n");
13747: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 13748: printf("Computing age specific forward period (stable) prevalences in each health state \n");
13749: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 13750: for(age=bage; age <=fage ;age++){
1.235 brouard 13751: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 13752: if (vpopbased==1) {
13753: if(mobilav ==0){
13754: for(i=1; i<=nlstate;i++)
13755: prlim[i][i]=probs[(int)age][i][k];
13756: }else{ /* mobilav */
13757: for(i=1; i<=nlstate;i++)
13758: prlim[i][i]=mobaverage[(int)age][i][k];
13759: }
13760: }
1.219 brouard 13761:
1.227 brouard 13762: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
13763: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
13764: /* printf(" age %4.0f ",age); */
13765: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
13766: for(i=1, epj[j]=0.;i <=nlstate;i++) {
13767: epj[j] += prlim[i][i]*eij[i][j][(int)age];
13768: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
13769: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
13770: }
13771: epj[nlstate+1] +=epj[j];
13772: }
13773: /* printf(" age %4.0f \n",age); */
1.219 brouard 13774:
1.227 brouard 13775: for(i=1, vepp=0.;i <=nlstate;i++)
13776: for(j=1;j <=nlstate;j++)
13777: vepp += vareij[i][j][(int)age];
13778: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
13779: for(j=1;j <=nlstate;j++){
13780: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
13781: }
13782: fprintf(ficrest,"\n");
13783: }
1.208 brouard 13784: } /* End vpopbased */
1.269 brouard 13785: free_vector(epj,1,nlstate+1);
1.208 brouard 13786: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
13787: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 13788: printf("done selection\n");fflush(stdout);
13789: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 13790:
1.235 brouard 13791: } /* End k selection */
1.227 brouard 13792:
13793: printf("done State-specific expectancies\n");fflush(stdout);
13794: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
13795:
1.288 brouard 13796: /* variance-covariance of forward period prevalence*/
1.269 brouard 13797: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 13798:
1.227 brouard 13799:
1.290 brouard 13800: free_vector(weight,firstobs,lastobs);
1.330 brouard 13801: free_imatrix(Tvardk,1,NCOVMAX,1,2);
1.227 brouard 13802: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 13803: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
13804: free_matrix(anint,1,maxwav,firstobs,lastobs);
13805: free_matrix(mint,1,maxwav,firstobs,lastobs);
13806: free_ivector(cod,firstobs,lastobs);
1.227 brouard 13807: free_ivector(tab,1,NCOVMAX);
13808: fclose(ficresstdeij);
13809: fclose(ficrescveij);
13810: fclose(ficresvij);
13811: fclose(ficrest);
13812: fclose(ficpar);
13813:
13814:
1.126 brouard 13815: /*---------- End : free ----------------*/
1.219 brouard 13816: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 13817: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
13818: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 13819: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
13820: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 13821: } /* mle==-3 arrives here for freeing */
1.227 brouard 13822: /* endfree:*/
13823: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
13824: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
13825: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 13826: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
13827: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
13828: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
13829: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 13830: free_matrix(matcov,1,npar,1,npar);
13831: free_matrix(hess,1,npar,1,npar);
13832: /*free_vector(delti,1,npar);*/
13833: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13834: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 13835: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 13836: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13837:
13838: free_ivector(ncodemax,1,NCOVMAX);
13839: free_ivector(ncodemaxwundef,1,NCOVMAX);
13840: free_ivector(Dummy,-1,NCOVMAX);
13841: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 13842: free_ivector(DummyV,1,NCOVMAX);
13843: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 13844: free_ivector(Typevar,-1,NCOVMAX);
13845: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 13846: free_ivector(TvarsQ,1,NCOVMAX);
13847: free_ivector(TvarsQind,1,NCOVMAX);
13848: free_ivector(TvarsD,1,NCOVMAX);
1.330 brouard 13849: free_ivector(TnsdVar,1,NCOVMAX);
1.234 brouard 13850: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 13851: free_ivector(TvarFD,1,NCOVMAX);
13852: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 13853: free_ivector(TvarF,1,NCOVMAX);
13854: free_ivector(TvarFind,1,NCOVMAX);
13855: free_ivector(TvarV,1,NCOVMAX);
13856: free_ivector(TvarVind,1,NCOVMAX);
13857: free_ivector(TvarA,1,NCOVMAX);
13858: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 13859: free_ivector(TvarFQ,1,NCOVMAX);
13860: free_ivector(TvarFQind,1,NCOVMAX);
13861: free_ivector(TvarVD,1,NCOVMAX);
13862: free_ivector(TvarVDind,1,NCOVMAX);
13863: free_ivector(TvarVQ,1,NCOVMAX);
13864: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 13865: free_ivector(Tvarsel,1,NCOVMAX);
13866: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 13867: free_ivector(Tposprod,1,NCOVMAX);
13868: free_ivector(Tprod,1,NCOVMAX);
13869: free_ivector(Tvaraff,1,NCOVMAX);
13870: free_ivector(invalidvarcomb,1,ncovcombmax);
13871: free_ivector(Tage,1,NCOVMAX);
13872: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 13873: free_ivector(TmodelInvind,1,NCOVMAX);
13874: free_ivector(TmodelInvQind,1,NCOVMAX);
1.332 brouard 13875:
13876: free_matrix(precov, 1,MAXRESULTLINESPONE,1,NCOVMAX+1); /* Could be elsewhere ?*/
13877:
1.227 brouard 13878: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
13879: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 13880: fflush(fichtm);
13881: fflush(ficgp);
13882:
1.227 brouard 13883:
1.126 brouard 13884: if((nberr >0) || (nbwarn>0)){
1.216 brouard 13885: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
13886: 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 13887: }else{
13888: printf("End of Imach\n");
13889: fprintf(ficlog,"End of Imach\n");
13890: }
13891: printf("See log file on %s\n",filelog);
13892: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 13893: /*(void) gettimeofday(&end_time,&tzp);*/
13894: rend_time = time(NULL);
13895: end_time = *localtime(&rend_time);
13896: /* tml = *localtime(&end_time.tm_sec); */
13897: strcpy(strtend,asctime(&end_time));
1.126 brouard 13898: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
13899: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 13900: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13901:
1.157 brouard 13902: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13903: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13904: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13905: /* printf("Total time was %d uSec.\n", total_usecs);*/
13906: /* if(fileappend(fichtm,optionfilehtm)){ */
13907: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13908: fclose(fichtm);
13909: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13910: fclose(fichtmcov);
13911: fclose(ficgp);
13912: fclose(ficlog);
13913: /*------ End -----------*/
1.227 brouard 13914:
1.281 brouard 13915:
13916: /* Executes gnuplot */
1.227 brouard 13917:
13918: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13919: #ifdef WIN32
1.227 brouard 13920: if (_chdir(pathcd) != 0)
13921: printf("Can't move to directory %s!\n",path);
13922: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13923: #else
1.227 brouard 13924: if(chdir(pathcd) != 0)
13925: printf("Can't move to directory %s!\n", path);
13926: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13927: #endif
1.126 brouard 13928: printf("Current directory %s!\n",pathcd);
13929: /*strcat(plotcmd,CHARSEPARATOR);*/
13930: sprintf(plotcmd,"gnuplot");
1.157 brouard 13931: #ifdef _WIN32
1.126 brouard 13932: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13933: #endif
13934: if(!stat(plotcmd,&info)){
1.158 brouard 13935: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13936: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13937: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13938: }else
13939: strcpy(pplotcmd,plotcmd);
1.157 brouard 13940: #ifdef __unix
1.126 brouard 13941: strcpy(plotcmd,GNUPLOTPROGRAM);
13942: if(!stat(plotcmd,&info)){
1.158 brouard 13943: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13944: }else
13945: strcpy(pplotcmd,plotcmd);
13946: #endif
13947: }else
13948: strcpy(pplotcmd,plotcmd);
13949:
13950: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13951: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13952: strcpy(pplotcmd,plotcmd);
1.227 brouard 13953:
1.126 brouard 13954: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13955: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13956: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13957: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13958: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13959: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13960: strcpy(plotcmd,pplotcmd);
13961: }
1.126 brouard 13962: }
1.158 brouard 13963: printf(" Successful, please wait...");
1.126 brouard 13964: while (z[0] != 'q') {
13965: /* chdir(path); */
1.154 brouard 13966: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13967: scanf("%s",z);
13968: /* if (z[0] == 'c') system("./imach"); */
13969: if (z[0] == 'e') {
1.158 brouard 13970: #ifdef __APPLE__
1.152 brouard 13971: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13972: #elif __linux
13973: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13974: #else
1.152 brouard 13975: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13976: #endif
13977: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13978: system(pplotcmd);
1.126 brouard 13979: }
13980: else if (z[0] == 'g') system(plotcmd);
13981: else if (z[0] == 'q') exit(0);
13982: }
1.227 brouard 13983: end:
1.126 brouard 13984: while (z[0] != 'q') {
1.195 brouard 13985: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13986: scanf("%s",z);
13987: }
1.283 brouard 13988: printf("End\n");
1.282 brouard 13989: exit(0);
1.126 brouard 13990: }
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