Annotation of imach/src/imach.c, revision 1.314
1.314 ! brouard 1: /* $Id: imach.c,v 1.313 2022/04/11 15:57:42 brouard Exp $
1.126 brouard 2: $State: Exp $
1.163 brouard 3: $Log: imach.c,v $
1.314 ! brouard 4: Revision 1.313 2022/04/11 15:57:42 brouard
! 5: * imach.c (Module): Error in rewriting the 'r' file with yearsfproj or yearsbproj fixed
! 6:
1.313 brouard 7: Revision 1.312 2022/04/05 21:24:39 brouard
8: *** empty log message ***
9:
1.312 brouard 10: Revision 1.311 2022/04/05 21:03:51 brouard
11: Summary: Fixed quantitative covariates
12:
13: Fixed covariates (dummy or quantitative)
14: with missing values have never been allowed but are ERRORS and
15: program quits. Standard deviations of fixed covariates were
16: wrongly computed. Mean and standard deviations of time varying
17: covariates are still not computed.
18:
1.311 brouard 19: Revision 1.310 2022/03/17 08:45:53 brouard
20: Summary: 99r25
21:
22: Improving detection of errors: result lines should be compatible with
23: the model.
24:
1.310 brouard 25: Revision 1.309 2021/05/20 12:39:14 brouard
26: Summary: Version 0.99r24
27:
1.309 brouard 28: Revision 1.308 2021/03/31 13:11:57 brouard
29: Summary: Version 0.99r23
30:
31:
32: * imach.c (Module): Still bugs in the result loop. Thank to Holly Benett
33:
1.308 brouard 34: Revision 1.307 2021/03/08 18:11:32 brouard
35: Summary: 0.99r22 fixed bug on result:
36:
1.307 brouard 37: Revision 1.306 2021/02/20 15:44:02 brouard
38: Summary: Version 0.99r21
39:
40: * imach.c (Module): Fix bug on quitting after result lines!
41: (Module): Version 0.99r21
42:
1.306 brouard 43: Revision 1.305 2021/02/20 15:28:30 brouard
44: * imach.c (Module): Fix bug on quitting after result lines!
45:
1.305 brouard 46: Revision 1.304 2021/02/12 11:34:20 brouard
47: * imach.c (Module): The use of a Windows BOM (huge) file is now an error
48:
1.304 brouard 49: Revision 1.303 2021/02/11 19:50:15 brouard
50: * (Module): imach.c Someone entered 'results:' instead of 'result:'. Now it is an error which is printed.
51:
1.303 brouard 52: Revision 1.302 2020/02/22 21:00:05 brouard
53: * (Module): imach.c Update mle=-3 (for computing Life expectancy
54: and life table from the data without any state)
55:
1.302 brouard 56: Revision 1.301 2019/06/04 13:51:20 brouard
57: Summary: Error in 'r'parameter file backcast yearsbproj instead of yearsfproj
58:
1.301 brouard 59: Revision 1.300 2019/05/22 19:09:45 brouard
60: Summary: version 0.99r19 of May 2019
61:
1.300 brouard 62: Revision 1.299 2019/05/22 18:37:08 brouard
63: Summary: Cleaned 0.99r19
64:
1.299 brouard 65: Revision 1.298 2019/05/22 18:19:56 brouard
66: *** empty log message ***
67:
1.298 brouard 68: Revision 1.297 2019/05/22 17:56:10 brouard
69: Summary: Fix bug by moving date2dmy and nhstepm which gaefin=-1
70:
1.297 brouard 71: Revision 1.296 2019/05/20 13:03:18 brouard
72: Summary: Projection syntax simplified
73:
74:
75: We can now start projections, forward or backward, from the mean date
76: of inteviews up to or down to a number of years of projection:
77: prevforecast=1 yearsfproj=15.3 mobil_average=0
78: or
79: prevforecast=1 starting-proj-date=1/1/2007 final-proj-date=12/31/2017 mobil_average=0
80: or
81: prevbackcast=1 yearsbproj=12.3 mobil_average=1
82: or
83: prevbackcast=1 starting-back-date=1/10/1999 final-back-date=1/1/1985 mobil_average=1
84:
1.296 brouard 85: Revision 1.295 2019/05/18 09:52:50 brouard
86: Summary: doxygen tex bug
87:
1.295 brouard 88: Revision 1.294 2019/05/16 14:54:33 brouard
89: Summary: There was some wrong lines added
90:
1.294 brouard 91: Revision 1.293 2019/05/09 15:17:34 brouard
92: *** empty log message ***
93:
1.293 brouard 94: Revision 1.292 2019/05/09 14:17:20 brouard
95: Summary: Some updates
96:
1.292 brouard 97: Revision 1.291 2019/05/09 13:44:18 brouard
98: Summary: Before ncovmax
99:
1.291 brouard 100: Revision 1.290 2019/05/09 13:39:37 brouard
101: Summary: 0.99r18 unlimited number of individuals
102:
103: 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.
104:
1.290 brouard 105: Revision 1.289 2018/12/13 09:16:26 brouard
106: Summary: Bug for young ages (<-30) will be in r17
107:
1.289 brouard 108: Revision 1.288 2018/05/02 20:58:27 brouard
109: Summary: Some bugs fixed
110:
1.288 brouard 111: Revision 1.287 2018/05/01 17:57:25 brouard
112: Summary: Bug fixed by providing frequencies only for non missing covariates
113:
1.287 brouard 114: Revision 1.286 2018/04/27 14:27:04 brouard
115: Summary: some minor bugs
116:
1.286 brouard 117: Revision 1.285 2018/04/21 21:02:16 brouard
118: Summary: Some bugs fixed, valgrind tested
119:
1.285 brouard 120: Revision 1.284 2018/04/20 05:22:13 brouard
121: Summary: Computing mean and stdeviation of fixed quantitative variables
122:
1.284 brouard 123: Revision 1.283 2018/04/19 14:49:16 brouard
124: Summary: Some minor bugs fixed
125:
1.283 brouard 126: Revision 1.282 2018/02/27 22:50:02 brouard
127: *** empty log message ***
128:
1.282 brouard 129: Revision 1.281 2018/02/27 19:25:23 brouard
130: Summary: Adding second argument for quitting
131:
1.281 brouard 132: Revision 1.280 2018/02/21 07:58:13 brouard
133: Summary: 0.99r15
134:
135: New Makefile with recent VirtualBox 5.26. Bug in sqrt negatve in imach.c
136:
1.280 brouard 137: Revision 1.279 2017/07/20 13:35:01 brouard
138: Summary: temporary working
139:
1.279 brouard 140: Revision 1.278 2017/07/19 14:09:02 brouard
141: Summary: Bug for mobil_average=0 and prevforecast fixed(?)
142:
1.278 brouard 143: Revision 1.277 2017/07/17 08:53:49 brouard
144: Summary: BOM files can be read now
145:
1.277 brouard 146: Revision 1.276 2017/06/30 15:48:31 brouard
147: Summary: Graphs improvements
148:
1.276 brouard 149: Revision 1.275 2017/06/30 13:39:33 brouard
150: Summary: Saito's color
151:
1.275 brouard 152: Revision 1.274 2017/06/29 09:47:08 brouard
153: Summary: Version 0.99r14
154:
1.274 brouard 155: Revision 1.273 2017/06/27 11:06:02 brouard
156: Summary: More documentation on projections
157:
1.273 brouard 158: Revision 1.272 2017/06/27 10:22:40 brouard
159: Summary: Color of backprojection changed from 6 to 5(yellow)
160:
1.272 brouard 161: Revision 1.271 2017/06/27 10:17:50 brouard
162: Summary: Some bug with rint
163:
1.271 brouard 164: Revision 1.270 2017/05/24 05:45:29 brouard
165: *** empty log message ***
166:
1.270 brouard 167: Revision 1.269 2017/05/23 08:39:25 brouard
168: Summary: Code into subroutine, cleanings
169:
1.269 brouard 170: Revision 1.268 2017/05/18 20:09:32 brouard
171: Summary: backprojection and confidence intervals of backprevalence
172:
1.268 brouard 173: Revision 1.267 2017/05/13 10:25:05 brouard
174: Summary: temporary save for backprojection
175:
1.267 brouard 176: Revision 1.266 2017/05/13 07:26:12 brouard
177: Summary: Version 0.99r13 (improvements and bugs fixed)
178:
1.266 brouard 179: Revision 1.265 2017/04/26 16:22:11 brouard
180: Summary: imach 0.99r13 Some bugs fixed
181:
1.265 brouard 182: Revision 1.264 2017/04/26 06:01:29 brouard
183: Summary: Labels in graphs
184:
1.264 brouard 185: Revision 1.263 2017/04/24 15:23:15 brouard
186: Summary: to save
187:
1.263 brouard 188: Revision 1.262 2017/04/18 16:48:12 brouard
189: *** empty log message ***
190:
1.262 brouard 191: Revision 1.261 2017/04/05 10:14:09 brouard
192: Summary: Bug in E_ as well as in T_ fixed nres-1 vs k1-1
193:
1.261 brouard 194: Revision 1.260 2017/04/04 17:46:59 brouard
195: Summary: Gnuplot indexations fixed (humm)
196:
1.260 brouard 197: Revision 1.259 2017/04/04 13:01:16 brouard
198: Summary: Some errors to warnings only if date of death is unknown but status is death we could set to pi3
199:
1.259 brouard 200: Revision 1.258 2017/04/03 10:17:47 brouard
201: Summary: Version 0.99r12
202:
203: Some cleanings, conformed with updated documentation.
204:
1.258 brouard 205: Revision 1.257 2017/03/29 16:53:30 brouard
206: Summary: Temp
207:
1.257 brouard 208: Revision 1.256 2017/03/27 05:50:23 brouard
209: Summary: Temporary
210:
1.256 brouard 211: Revision 1.255 2017/03/08 16:02:28 brouard
212: Summary: IMaCh version 0.99r10 bugs in gnuplot fixed
213:
1.255 brouard 214: Revision 1.254 2017/03/08 07:13:00 brouard
215: Summary: Fixing data parameter line
216:
1.254 brouard 217: Revision 1.253 2016/12/15 11:59:41 brouard
218: Summary: 0.99 in progress
219:
1.253 brouard 220: Revision 1.252 2016/09/15 21:15:37 brouard
221: *** empty log message ***
222:
1.252 brouard 223: Revision 1.251 2016/09/15 15:01:13 brouard
224: Summary: not working
225:
1.251 brouard 226: Revision 1.250 2016/09/08 16:07:27 brouard
227: Summary: continue
228:
1.250 brouard 229: Revision 1.249 2016/09/07 17:14:18 brouard
230: Summary: Starting values from frequencies
231:
1.249 brouard 232: Revision 1.248 2016/09/07 14:10:18 brouard
233: *** empty log message ***
234:
1.248 brouard 235: Revision 1.247 2016/09/02 11:11:21 brouard
236: *** empty log message ***
237:
1.247 brouard 238: Revision 1.246 2016/09/02 08:49:22 brouard
239: *** empty log message ***
240:
1.246 brouard 241: Revision 1.245 2016/09/02 07:25:01 brouard
242: *** empty log message ***
243:
1.245 brouard 244: Revision 1.244 2016/09/02 07:17:34 brouard
245: *** empty log message ***
246:
1.244 brouard 247: Revision 1.243 2016/09/02 06:45:35 brouard
248: *** empty log message ***
249:
1.243 brouard 250: Revision 1.242 2016/08/30 15:01:20 brouard
251: Summary: Fixing a lots
252:
1.242 brouard 253: Revision 1.241 2016/08/29 17:17:25 brouard
254: Summary: gnuplot problem in Back projection to fix
255:
1.241 brouard 256: Revision 1.240 2016/08/29 07:53:18 brouard
257: Summary: Better
258:
1.240 brouard 259: Revision 1.239 2016/08/26 15:51:03 brouard
260: Summary: Improvement in Powell output in order to copy and paste
261:
262: Author:
263:
1.239 brouard 264: Revision 1.238 2016/08/26 14:23:35 brouard
265: Summary: Starting tests of 0.99
266:
1.238 brouard 267: Revision 1.237 2016/08/26 09:20:19 brouard
268: Summary: to valgrind
269:
1.237 brouard 270: Revision 1.236 2016/08/25 10:50:18 brouard
271: *** empty log message ***
272:
1.236 brouard 273: Revision 1.235 2016/08/25 06:59:23 brouard
274: *** empty log message ***
275:
1.235 brouard 276: Revision 1.234 2016/08/23 16:51:20 brouard
277: *** empty log message ***
278:
1.234 brouard 279: Revision 1.233 2016/08/23 07:40:50 brouard
280: Summary: not working
281:
1.233 brouard 282: Revision 1.232 2016/08/22 14:20:21 brouard
283: Summary: not working
284:
1.232 brouard 285: Revision 1.231 2016/08/22 07:17:15 brouard
286: Summary: not working
287:
1.231 brouard 288: Revision 1.230 2016/08/22 06:55:53 brouard
289: Summary: Not working
290:
1.230 brouard 291: Revision 1.229 2016/07/23 09:45:53 brouard
292: Summary: Completing for func too
293:
1.229 brouard 294: Revision 1.228 2016/07/22 17:45:30 brouard
295: Summary: Fixing some arrays, still debugging
296:
1.227 brouard 297: Revision 1.226 2016/07/12 18:42:34 brouard
298: Summary: temp
299:
1.226 brouard 300: Revision 1.225 2016/07/12 08:40:03 brouard
301: Summary: saving but not running
302:
1.225 brouard 303: Revision 1.224 2016/07/01 13:16:01 brouard
304: Summary: Fixes
305:
1.224 brouard 306: Revision 1.223 2016/02/19 09:23:35 brouard
307: Summary: temporary
308:
1.223 brouard 309: Revision 1.222 2016/02/17 08:14:50 brouard
310: Summary: Probably last 0.98 stable version 0.98r6
311:
1.222 brouard 312: Revision 1.221 2016/02/15 23:35:36 brouard
313: Summary: minor bug
314:
1.220 brouard 315: Revision 1.219 2016/02/15 00:48:12 brouard
316: *** empty log message ***
317:
1.219 brouard 318: Revision 1.218 2016/02/12 11:29:23 brouard
319: Summary: 0.99 Back projections
320:
1.218 brouard 321: Revision 1.217 2015/12/23 17:18:31 brouard
322: Summary: Experimental backcast
323:
1.217 brouard 324: Revision 1.216 2015/12/18 17:32:11 brouard
325: Summary: 0.98r4 Warning and status=-2
326:
327: Version 0.98r4 is now:
328: - displaying an error when status is -1, date of interview unknown and date of death known;
329: - permitting a status -2 when the vital status is unknown at a known date of right truncation.
330: Older changes concerning s=-2, dating from 2005 have been supersed.
331:
1.216 brouard 332: Revision 1.215 2015/12/16 08:52:24 brouard
333: Summary: 0.98r4 working
334:
1.215 brouard 335: Revision 1.214 2015/12/16 06:57:54 brouard
336: Summary: temporary not working
337:
1.214 brouard 338: Revision 1.213 2015/12/11 18:22:17 brouard
339: Summary: 0.98r4
340:
1.213 brouard 341: Revision 1.212 2015/11/21 12:47:24 brouard
342: Summary: minor typo
343:
1.212 brouard 344: Revision 1.211 2015/11/21 12:41:11 brouard
345: Summary: 0.98r3 with some graph of projected cross-sectional
346:
347: Author: Nicolas Brouard
348:
1.211 brouard 349: Revision 1.210 2015/11/18 17:41:20 brouard
1.252 brouard 350: Summary: Start working on projected prevalences Revision 1.209 2015/11/17 22:12:03 brouard
1.210 brouard 351: Summary: Adding ftolpl parameter
352: Author: N Brouard
353:
354: We had difficulties to get smoothed confidence intervals. It was due
355: to the period prevalence which wasn't computed accurately. The inner
356: parameter ftolpl is now an outer parameter of the .imach parameter
357: file after estepm. If ftolpl is small 1.e-4 and estepm too,
358: computation are long.
359:
1.209 brouard 360: Revision 1.208 2015/11/17 14:31:57 brouard
361: Summary: temporary
362:
1.208 brouard 363: Revision 1.207 2015/10/27 17:36:57 brouard
364: *** empty log message ***
365:
1.207 brouard 366: Revision 1.206 2015/10/24 07:14:11 brouard
367: *** empty log message ***
368:
1.206 brouard 369: Revision 1.205 2015/10/23 15:50:53 brouard
370: Summary: 0.98r3 some clarification for graphs on likelihood contributions
371:
1.205 brouard 372: Revision 1.204 2015/10/01 16:20:26 brouard
373: Summary: Some new graphs of contribution to likelihood
374:
1.204 brouard 375: Revision 1.203 2015/09/30 17:45:14 brouard
376: Summary: looking at better estimation of the hessian
377:
378: Also a better criteria for convergence to the period prevalence And
379: therefore adding the number of years needed to converge. (The
380: prevalence in any alive state shold sum to one
381:
1.203 brouard 382: Revision 1.202 2015/09/22 19:45:16 brouard
383: Summary: Adding some overall graph on contribution to likelihood. Might change
384:
1.202 brouard 385: Revision 1.201 2015/09/15 17:34:58 brouard
386: Summary: 0.98r0
387:
388: - Some new graphs like suvival functions
389: - Some bugs fixed like model=1+age+V2.
390:
1.201 brouard 391: Revision 1.200 2015/09/09 16:53:55 brouard
392: Summary: Big bug thanks to Flavia
393:
394: Even model=1+age+V2. did not work anymore
395:
1.200 brouard 396: Revision 1.199 2015/09/07 14:09:23 brouard
397: Summary: 0.98q6 changing default small png format for graph to vectorized svg.
398:
1.199 brouard 399: Revision 1.198 2015/09/03 07:14:39 brouard
400: Summary: 0.98q5 Flavia
401:
1.198 brouard 402: Revision 1.197 2015/09/01 18:24:39 brouard
403: *** empty log message ***
404:
1.197 brouard 405: Revision 1.196 2015/08/18 23:17:52 brouard
406: Summary: 0.98q5
407:
1.196 brouard 408: Revision 1.195 2015/08/18 16:28:39 brouard
409: Summary: Adding a hack for testing purpose
410:
411: After reading the title, ftol and model lines, if the comment line has
412: a q, starting with #q, the answer at the end of the run is quit. It
413: permits to run test files in batch with ctest. The former workaround was
414: $ echo q | imach foo.imach
415:
1.195 brouard 416: Revision 1.194 2015/08/18 13:32:00 brouard
417: Summary: Adding error when the covariance matrix doesn't contain the exact number of lines required by the model line.
418:
1.194 brouard 419: Revision 1.193 2015/08/04 07:17:42 brouard
420: Summary: 0.98q4
421:
1.193 brouard 422: Revision 1.192 2015/07/16 16:49:02 brouard
423: Summary: Fixing some outputs
424:
1.192 brouard 425: Revision 1.191 2015/07/14 10:00:33 brouard
426: Summary: Some fixes
427:
1.191 brouard 428: Revision 1.190 2015/05/05 08:51:13 brouard
429: Summary: Adding digits in output parameters (7 digits instead of 6)
430:
431: Fix 1+age+.
432:
1.190 brouard 433: Revision 1.189 2015/04/30 14:45:16 brouard
434: Summary: 0.98q2
435:
1.189 brouard 436: Revision 1.188 2015/04/30 08:27:53 brouard
437: *** empty log message ***
438:
1.188 brouard 439: Revision 1.187 2015/04/29 09:11:15 brouard
440: *** empty log message ***
441:
1.187 brouard 442: Revision 1.186 2015/04/23 12:01:52 brouard
443: Summary: V1*age is working now, version 0.98q1
444:
445: Some codes had been disabled in order to simplify and Vn*age was
446: working in the optimization phase, ie, giving correct MLE parameters,
447: but, as usual, outputs were not correct and program core dumped.
448:
1.186 brouard 449: Revision 1.185 2015/03/11 13:26:42 brouard
450: Summary: Inclusion of compile and links command line for Intel Compiler
451:
1.185 brouard 452: Revision 1.184 2015/03/11 11:52:39 brouard
453: Summary: Back from Windows 8. Intel Compiler
454:
1.184 brouard 455: Revision 1.183 2015/03/10 20:34:32 brouard
456: Summary: 0.98q0, trying with directest, mnbrak fixed
457:
458: We use directest instead of original Powell test; probably no
459: incidence on the results, but better justifications;
460: We fixed Numerical Recipes mnbrak routine which was wrong and gave
461: wrong results.
462:
1.183 brouard 463: Revision 1.182 2015/02/12 08:19:57 brouard
464: Summary: Trying to keep directest which seems simpler and more general
465: Author: Nicolas Brouard
466:
1.182 brouard 467: Revision 1.181 2015/02/11 23:22:24 brouard
468: Summary: Comments on Powell added
469:
470: Author:
471:
1.181 brouard 472: Revision 1.180 2015/02/11 17:33:45 brouard
473: Summary: Finishing move from main to function (hpijx and prevalence_limit)
474:
1.180 brouard 475: Revision 1.179 2015/01/04 09:57:06 brouard
476: Summary: back to OS/X
477:
1.179 brouard 478: Revision 1.178 2015/01/04 09:35:48 brouard
479: *** empty log message ***
480:
1.178 brouard 481: Revision 1.177 2015/01/03 18:40:56 brouard
482: Summary: Still testing ilc32 on OSX
483:
1.177 brouard 484: Revision 1.176 2015/01/03 16:45:04 brouard
485: *** empty log message ***
486:
1.176 brouard 487: Revision 1.175 2015/01/03 16:33:42 brouard
488: *** empty log message ***
489:
1.175 brouard 490: Revision 1.174 2015/01/03 16:15:49 brouard
491: Summary: Still in cross-compilation
492:
1.174 brouard 493: Revision 1.173 2015/01/03 12:06:26 brouard
494: Summary: trying to detect cross-compilation
495:
1.173 brouard 496: Revision 1.172 2014/12/27 12:07:47 brouard
497: Summary: Back from Visual Studio and Intel, options for compiling for Windows XP
498:
1.172 brouard 499: Revision 1.171 2014/12/23 13:26:59 brouard
500: Summary: Back from Visual C
501:
502: Still problem with utsname.h on Windows
503:
1.171 brouard 504: Revision 1.170 2014/12/23 11:17:12 brouard
505: Summary: Cleaning some \%% back to %%
506:
507: The escape was mandatory for a specific compiler (which one?), but too many warnings.
508:
1.170 brouard 509: Revision 1.169 2014/12/22 23:08:31 brouard
510: Summary: 0.98p
511:
512: Outputs some informations on compiler used, OS etc. Testing on different platforms.
513:
1.169 brouard 514: Revision 1.168 2014/12/22 15:17:42 brouard
1.170 brouard 515: Summary: update
1.169 brouard 516:
1.168 brouard 517: Revision 1.167 2014/12/22 13:50:56 brouard
518: Summary: Testing uname and compiler version and if compiled 32 or 64
519:
520: Testing on Linux 64
521:
1.167 brouard 522: Revision 1.166 2014/12/22 11:40:47 brouard
523: *** empty log message ***
524:
1.166 brouard 525: Revision 1.165 2014/12/16 11:20:36 brouard
526: Summary: After compiling on Visual C
527:
528: * imach.c (Module): Merging 1.61 to 1.162
529:
1.165 brouard 530: Revision 1.164 2014/12/16 10:52:11 brouard
531: Summary: Merging with Visual C after suppressing some warnings for unused variables. Also fixing Saito's bug 0.98Xn
532:
533: * imach.c (Module): Merging 1.61 to 1.162
534:
1.164 brouard 535: Revision 1.163 2014/12/16 10:30:11 brouard
536: * imach.c (Module): Merging 1.61 to 1.162
537:
1.163 brouard 538: Revision 1.162 2014/09/25 11:43:39 brouard
539: Summary: temporary backup 0.99!
540:
1.162 brouard 541: Revision 1.1 2014/09/16 11:06:58 brouard
542: Summary: With some code (wrong) for nlopt
543:
544: Author:
545:
546: Revision 1.161 2014/09/15 20:41:41 brouard
547: Summary: Problem with macro SQR on Intel compiler
548:
1.161 brouard 549: Revision 1.160 2014/09/02 09:24:05 brouard
550: *** empty log message ***
551:
1.160 brouard 552: Revision 1.159 2014/09/01 10:34:10 brouard
553: Summary: WIN32
554: Author: Brouard
555:
1.159 brouard 556: Revision 1.158 2014/08/27 17:11:51 brouard
557: *** empty log message ***
558:
1.158 brouard 559: Revision 1.157 2014/08/27 16:26:55 brouard
560: Summary: Preparing windows Visual studio version
561: Author: Brouard
562:
563: In order to compile on Visual studio, time.h is now correct and time_t
564: and tm struct should be used. difftime should be used but sometimes I
565: just make the differences in raw time format (time(&now).
566: Trying to suppress #ifdef LINUX
567: Add xdg-open for __linux in order to open default browser.
568:
1.157 brouard 569: Revision 1.156 2014/08/25 20:10:10 brouard
570: *** empty log message ***
571:
1.156 brouard 572: Revision 1.155 2014/08/25 18:32:34 brouard
573: Summary: New compile, minor changes
574: Author: Brouard
575:
1.155 brouard 576: Revision 1.154 2014/06/20 17:32:08 brouard
577: Summary: Outputs now all graphs of convergence to period prevalence
578:
1.154 brouard 579: Revision 1.153 2014/06/20 16:45:46 brouard
580: Summary: If 3 live state, convergence to period prevalence on same graph
581: Author: Brouard
582:
1.153 brouard 583: Revision 1.152 2014/06/18 17:54:09 brouard
584: Summary: open browser, use gnuplot on same dir than imach if not found in the path
585:
1.152 brouard 586: Revision 1.151 2014/06/18 16:43:30 brouard
587: *** empty log message ***
588:
1.151 brouard 589: Revision 1.150 2014/06/18 16:42:35 brouard
590: Summary: If gnuplot is not in the path try on same directory than imach binary (OSX)
591: Author: brouard
592:
1.150 brouard 593: Revision 1.149 2014/06/18 15:51:14 brouard
594: Summary: Some fixes in parameter files errors
595: Author: Nicolas Brouard
596:
1.149 brouard 597: Revision 1.148 2014/06/17 17:38:48 brouard
598: Summary: Nothing new
599: Author: Brouard
600:
601: Just a new packaging for OS/X version 0.98nS
602:
1.148 brouard 603: Revision 1.147 2014/06/16 10:33:11 brouard
604: *** empty log message ***
605:
1.147 brouard 606: Revision 1.146 2014/06/16 10:20:28 brouard
607: Summary: Merge
608: Author: Brouard
609:
610: Merge, before building revised version.
611:
1.146 brouard 612: Revision 1.145 2014/06/10 21:23:15 brouard
613: Summary: Debugging with valgrind
614: Author: Nicolas Brouard
615:
616: Lot of changes in order to output the results with some covariates
617: After the Edimburgh REVES conference 2014, it seems mandatory to
618: improve the code.
619: No more memory valgrind error but a lot has to be done in order to
620: continue the work of splitting the code into subroutines.
621: Also, decodemodel has been improved. Tricode is still not
622: optimal. nbcode should be improved. Documentation has been added in
623: the source code.
624:
1.144 brouard 625: Revision 1.143 2014/01/26 09:45:38 brouard
626: Summary: Version 0.98nR (to be improved, but gives same optimization results as 0.98k. Nice, promising
627:
628: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
629: (Module): Version 0.98nR Running ok, but output format still only works for three covariates.
630:
1.143 brouard 631: Revision 1.142 2014/01/26 03:57:36 brouard
632: Summary: gnuplot changed plot w l 1 has to be changed to plot w l lt 2
633:
634: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
635:
1.142 brouard 636: Revision 1.141 2014/01/26 02:42:01 brouard
637: * imach.c (Module): Trying to merge old staffs together while being at Tokyo. Not tested...
638:
1.141 brouard 639: Revision 1.140 2011/09/02 10:37:54 brouard
640: Summary: times.h is ok with mingw32 now.
641:
1.140 brouard 642: Revision 1.139 2010/06/14 07:50:17 brouard
643: After the theft of my laptop, I probably lost some lines of codes which were not uploaded to the CVS tree.
644: I remember having already fixed agemin agemax which are pointers now but not cvs saved.
645:
1.139 brouard 646: Revision 1.138 2010/04/30 18:19:40 brouard
647: *** empty log message ***
648:
1.138 brouard 649: Revision 1.137 2010/04/29 18:11:38 brouard
650: (Module): Checking covariates for more complex models
651: than V1+V2. A lot of change to be done. Unstable.
652:
1.137 brouard 653: Revision 1.136 2010/04/26 20:30:53 brouard
654: (Module): merging some libgsl code. Fixing computation
655: of likelione (using inter/intrapolation if mle = 0) in order to
656: get same likelihood as if mle=1.
657: Some cleaning of code and comments added.
658:
1.136 brouard 659: Revision 1.135 2009/10/29 15:33:14 brouard
660: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
661:
1.135 brouard 662: Revision 1.134 2009/10/29 13:18:53 brouard
663: (Module): Now imach stops if date of birth, at least year of birth, is not given. Some cleaning of the code.
664:
1.134 brouard 665: Revision 1.133 2009/07/06 10:21:25 brouard
666: just nforces
667:
1.133 brouard 668: Revision 1.132 2009/07/06 08:22:05 brouard
669: Many tings
670:
1.132 brouard 671: Revision 1.131 2009/06/20 16:22:47 brouard
672: Some dimensions resccaled
673:
1.131 brouard 674: Revision 1.130 2009/05/26 06:44:34 brouard
675: (Module): Max Covariate is now set to 20 instead of 8. A
676: lot of cleaning with variables initialized to 0. Trying to make
677: V2+V3*age+V1+V4 strb=V3*age+V1+V4 working better.
678:
1.130 brouard 679: Revision 1.129 2007/08/31 13:49:27 lievre
680: Modification of the way of exiting when the covariate is not binary in order to see on the window the error message before exiting
681:
1.129 lievre 682: Revision 1.128 2006/06/30 13:02:05 brouard
683: (Module): Clarifications on computing e.j
684:
1.128 brouard 685: Revision 1.127 2006/04/28 18:11:50 brouard
686: (Module): Yes the sum of survivors was wrong since
687: imach-114 because nhstepm was no more computed in the age
688: loop. Now we define nhstepma in the age loop.
689: (Module): In order to speed up (in case of numerous covariates) we
690: compute health expectancies (without variances) in a first step
691: and then all the health expectancies with variances or standard
692: deviation (needs data from the Hessian matrices) which slows the
693: computation.
694: In the future we should be able to stop the program is only health
695: expectancies and graph are needed without standard deviations.
696:
1.127 brouard 697: Revision 1.126 2006/04/28 17:23:28 brouard
698: (Module): Yes the sum of survivors was wrong since
699: imach-114 because nhstepm was no more computed in the age
700: loop. Now we define nhstepma in the age loop.
701: Version 0.98h
702:
1.126 brouard 703: Revision 1.125 2006/04/04 15:20:31 lievre
704: Errors in calculation of health expectancies. Age was not initialized.
705: Forecasting file added.
706:
707: Revision 1.124 2006/03/22 17:13:53 lievre
708: Parameters are printed with %lf instead of %f (more numbers after the comma).
709: The log-likelihood is printed in the log file
710:
711: Revision 1.123 2006/03/20 10:52:43 brouard
712: * imach.c (Module): <title> changed, corresponds to .htm file
713: name. <head> headers where missing.
714:
715: * imach.c (Module): Weights can have a decimal point as for
716: English (a comma might work with a correct LC_NUMERIC environment,
717: otherwise the weight is truncated).
718: Modification of warning when the covariates values are not 0 or
719: 1.
720: Version 0.98g
721:
722: Revision 1.122 2006/03/20 09:45:41 brouard
723: (Module): Weights can have a decimal point as for
724: English (a comma might work with a correct LC_NUMERIC environment,
725: otherwise the weight is truncated).
726: Modification of warning when the covariates values are not 0 or
727: 1.
728: Version 0.98g
729:
730: Revision 1.121 2006/03/16 17:45:01 lievre
731: * imach.c (Module): Comments concerning covariates added
732:
733: * imach.c (Module): refinements in the computation of lli if
734: status=-2 in order to have more reliable computation if stepm is
735: not 1 month. Version 0.98f
736:
737: Revision 1.120 2006/03/16 15:10:38 lievre
738: (Module): refinements in the computation of lli if
739: status=-2 in order to have more reliable computation if stepm is
740: not 1 month. Version 0.98f
741:
742: Revision 1.119 2006/03/15 17:42:26 brouard
743: (Module): Bug if status = -2, the loglikelihood was
744: computed as likelihood omitting the logarithm. Version O.98e
745:
746: Revision 1.118 2006/03/14 18:20:07 brouard
747: (Module): varevsij Comments added explaining the second
748: table of variances if popbased=1 .
749: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
750: (Module): Function pstamp added
751: (Module): Version 0.98d
752:
753: Revision 1.117 2006/03/14 17:16:22 brouard
754: (Module): varevsij Comments added explaining the second
755: table of variances if popbased=1 .
756: (Module): Covariances of eij, ekl added, graphs fixed, new html link.
757: (Module): Function pstamp added
758: (Module): Version 0.98d
759:
760: Revision 1.116 2006/03/06 10:29:27 brouard
761: (Module): Variance-covariance wrong links and
762: varian-covariance of ej. is needed (Saito).
763:
764: Revision 1.115 2006/02/27 12:17:45 brouard
765: (Module): One freematrix added in mlikeli! 0.98c
766:
767: Revision 1.114 2006/02/26 12:57:58 brouard
768: (Module): Some improvements in processing parameter
769: filename with strsep.
770:
771: Revision 1.113 2006/02/24 14:20:24 brouard
772: (Module): Memory leaks checks with valgrind and:
773: datafile was not closed, some imatrix were not freed and on matrix
774: allocation too.
775:
776: Revision 1.112 2006/01/30 09:55:26 brouard
777: (Module): Back to gnuplot.exe instead of wgnuplot.exe
778:
779: Revision 1.111 2006/01/25 20:38:18 brouard
780: (Module): Lots of cleaning and bugs added (Gompertz)
781: (Module): Comments can be added in data file. Missing date values
782: can be a simple dot '.'.
783:
784: Revision 1.110 2006/01/25 00:51:50 brouard
785: (Module): Lots of cleaning and bugs added (Gompertz)
786:
787: Revision 1.109 2006/01/24 19:37:15 brouard
788: (Module): Comments (lines starting with a #) are allowed in data.
789:
790: Revision 1.108 2006/01/19 18:05:42 lievre
791: Gnuplot problem appeared...
792: To be fixed
793:
794: Revision 1.107 2006/01/19 16:20:37 brouard
795: Test existence of gnuplot in imach path
796:
797: Revision 1.106 2006/01/19 13:24:36 brouard
798: Some cleaning and links added in html output
799:
800: Revision 1.105 2006/01/05 20:23:19 lievre
801: *** empty log message ***
802:
803: Revision 1.104 2005/09/30 16:11:43 lievre
804: (Module): sump fixed, loop imx fixed, and simplifications.
805: (Module): If the status is missing at the last wave but we know
806: that the person is alive, then we can code his/her status as -2
807: (instead of missing=-1 in earlier versions) and his/her
808: contributions to the likelihood is 1 - Prob of dying from last
809: health status (= 1-p13= p11+p12 in the easiest case of somebody in
810: the healthy state at last known wave). Version is 0.98
811:
812: Revision 1.103 2005/09/30 15:54:49 lievre
813: (Module): sump fixed, loop imx fixed, and simplifications.
814:
815: Revision 1.102 2004/09/15 17:31:30 brouard
816: Add the possibility to read data file including tab characters.
817:
818: Revision 1.101 2004/09/15 10:38:38 brouard
819: Fix on curr_time
820:
821: Revision 1.100 2004/07/12 18:29:06 brouard
822: Add version for Mac OS X. Just define UNIX in Makefile
823:
824: Revision 1.99 2004/06/05 08:57:40 brouard
825: *** empty log message ***
826:
827: Revision 1.98 2004/05/16 15:05:56 brouard
828: New version 0.97 . First attempt to estimate force of mortality
829: directly from the data i.e. without the need of knowing the health
830: state at each age, but using a Gompertz model: log u =a + b*age .
831: This is the basic analysis of mortality and should be done before any
832: other analysis, in order to test if the mortality estimated from the
833: cross-longitudinal survey is different from the mortality estimated
834: from other sources like vital statistic data.
835:
836: The same imach parameter file can be used but the option for mle should be -3.
837:
1.133 brouard 838: Agnès, who wrote this part of the code, tried to keep most of the
1.126 brouard 839: former routines in order to include the new code within the former code.
840:
841: The output is very simple: only an estimate of the intercept and of
842: the slope with 95% confident intervals.
843:
844: Current limitations:
845: A) Even if you enter covariates, i.e. with the
846: model= V1+V2 equation for example, the programm does only estimate a unique global model without covariates.
847: B) There is no computation of Life Expectancy nor Life Table.
848:
849: Revision 1.97 2004/02/20 13:25:42 lievre
850: Version 0.96d. Population forecasting command line is (temporarily)
851: suppressed.
852:
853: Revision 1.96 2003/07/15 15:38:55 brouard
854: * imach.c (Repository): Errors in subdirf, 2, 3 while printing tmpout is
855: rewritten within the same printf. Workaround: many printfs.
856:
857: Revision 1.95 2003/07/08 07:54:34 brouard
858: * imach.c (Repository):
859: (Repository): Using imachwizard code to output a more meaningful covariance
860: matrix (cov(a12,c31) instead of numbers.
861:
862: Revision 1.94 2003/06/27 13:00:02 brouard
863: Just cleaning
864:
865: Revision 1.93 2003/06/25 16:33:55 brouard
866: (Module): On windows (cygwin) function asctime_r doesn't
867: exist so I changed back to asctime which exists.
868: (Module): Version 0.96b
869:
870: Revision 1.92 2003/06/25 16:30:45 brouard
871: (Module): On windows (cygwin) function asctime_r doesn't
872: exist so I changed back to asctime which exists.
873:
874: Revision 1.91 2003/06/25 15:30:29 brouard
875: * imach.c (Repository): Duplicated warning errors corrected.
876: (Repository): Elapsed time after each iteration is now output. It
877: helps to forecast when convergence will be reached. Elapsed time
878: is stamped in powell. We created a new html file for the graphs
879: concerning matrix of covariance. It has extension -cov.htm.
880:
881: Revision 1.90 2003/06/24 12:34:15 brouard
882: (Module): Some bugs corrected for windows. Also, when
883: mle=-1 a template is output in file "or"mypar.txt with the design
884: of the covariance matrix to be input.
885:
886: Revision 1.89 2003/06/24 12:30:52 brouard
887: (Module): Some bugs corrected for windows. Also, when
888: mle=-1 a template is output in file "or"mypar.txt with the design
889: of the covariance matrix to be input.
890:
891: Revision 1.88 2003/06/23 17:54:56 brouard
892: * 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.
893:
894: Revision 1.87 2003/06/18 12:26:01 brouard
895: Version 0.96
896:
897: Revision 1.86 2003/06/17 20:04:08 brouard
898: (Module): Change position of html and gnuplot routines and added
899: routine fileappend.
900:
901: Revision 1.85 2003/06/17 13:12:43 brouard
902: * imach.c (Repository): Check when date of death was earlier that
903: current date of interview. It may happen when the death was just
904: prior to the death. In this case, dh was negative and likelihood
905: was wrong (infinity). We still send an "Error" but patch by
906: assuming that the date of death was just one stepm after the
907: interview.
908: (Repository): Because some people have very long ID (first column)
909: we changed int to long in num[] and we added a new lvector for
910: memory allocation. But we also truncated to 8 characters (left
911: truncation)
912: (Repository): No more line truncation errors.
913:
914: Revision 1.84 2003/06/13 21:44:43 brouard
915: * imach.c (Repository): Replace "freqsummary" at a correct
916: place. It differs from routine "prevalence" which may be called
917: many times. Probs is memory consuming and must be used with
918: parcimony.
919: Version 0.95a3 (should output exactly the same maximization than 0.8a2)
920:
921: Revision 1.83 2003/06/10 13:39:11 lievre
922: *** empty log message ***
923:
924: Revision 1.82 2003/06/05 15:57:20 brouard
925: Add log in imach.c and fullversion number is now printed.
926:
927: */
928: /*
929: Interpolated Markov Chain
930:
931: Short summary of the programme:
932:
1.227 brouard 933: This program computes Healthy Life Expectancies or State-specific
934: (if states aren't health statuses) Expectancies from
935: cross-longitudinal data. Cross-longitudinal data consist in:
936:
937: -1- a first survey ("cross") where individuals from different ages
938: are interviewed on their health status or degree of disability (in
939: the case of a health survey which is our main interest)
940:
941: -2- at least a second wave of interviews ("longitudinal") which
942: measure each change (if any) in individual health status. Health
943: expectancies are computed from the time spent in each health state
944: according to a model. More health states you consider, more time is
945: necessary to reach the Maximum Likelihood of the parameters involved
946: in the model. The simplest model is the multinomial logistic model
947: where pij is the probability to be observed in state j at the second
948: wave conditional to be observed in state i at the first
949: wave. Therefore the model is: log(pij/pii)= aij + bij*age+ cij*sex +
950: etc , where 'age' is age and 'sex' is a covariate. If you want to
951: have a more complex model than "constant and age", you should modify
952: the program where the markup *Covariates have to be included here
953: again* invites you to do it. More covariates you add, slower the
1.126 brouard 954: convergence.
955:
956: The advantage of this computer programme, compared to a simple
957: multinomial logistic model, is clear when the delay between waves is not
958: identical for each individual. Also, if a individual missed an
959: intermediate interview, the information is lost, but taken into
960: account using an interpolation or extrapolation.
961:
962: hPijx is the probability to be observed in state i at age x+h
963: conditional to the observed state i at age x. The delay 'h' can be
964: split into an exact number (nh*stepm) of unobserved intermediate
965: states. This elementary transition (by month, quarter,
966: semester or year) is modelled as a multinomial logistic. The hPx
967: matrix is simply the matrix product of nh*stepm elementary matrices
968: and the contribution of each individual to the likelihood is simply
969: hPijx.
970:
971: Also this programme outputs the covariance matrix of the parameters but also
1.218 brouard 972: of the life expectancies. It also computes the period (stable) prevalence.
973:
974: Back prevalence and projections:
1.227 brouard 975:
976: - back_prevalence_limit(double *p, double **bprlim, double ageminpar,
977: double agemaxpar, double ftolpl, int *ncvyearp, double
978: dateprev1,double dateprev2, int firstpass, int lastpass, int
979: mobilavproj)
980:
981: Computes the back prevalence limit for any combination of
982: covariate values k at any age between ageminpar and agemaxpar and
983: returns it in **bprlim. In the loops,
984:
985: - **bprevalim(**bprlim, ***mobaverage, nlstate, *p, age, **oldm,
986: **savm, **dnewm, **doldm, **dsavm, ftolpl, ncvyearp, k);
987:
988: - hBijx Back Probability to be in state i at age x-h being in j at x
1.218 brouard 989: Computes for any combination of covariates k and any age between bage and fage
990: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
991: oldm=oldms;savm=savms;
1.227 brouard 992:
1.267 brouard 993: - hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.218 brouard 994: Computes the transition matrix starting at age 'age' over
995: 'nhstepm*hstepm*stepm' months (i.e. until
996: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
1.227 brouard 997: nhstepm*hstepm matrices.
998:
999: Returns p3mat[i][j][h] after calling
1000: p3mat[i][j][h]=matprod2(newm,
1001: bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm,
1002: dsavm,ij),\ 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
1003: oldm);
1.226 brouard 1004:
1005: Important routines
1006:
1007: - func (or funcone), computes logit (pij) distinguishing
1008: o fixed variables (single or product dummies or quantitative);
1009: o varying variables by:
1010: (1) wave (single, product dummies, quantitative),
1011: (2) by age (can be month) age (done), age*age (done), age*Vn where Vn can be:
1012: % fixed dummy (treated) or quantitative (not done because time-consuming);
1013: % varying dummy (not done) or quantitative (not done);
1014: - Tricode which tests the modality of dummy variables (in order to warn with wrong or empty modalities)
1015: and returns the number of efficient covariates cptcoveff and modalities nbcode[Tvar[k]][1]= 0 and nbcode[Tvar[k]][2]= 1 usually.
1016: - printinghtml which outputs results like life expectancy in and from a state for a combination of modalities of dummy variables
1017: o There are 2*cptcoveff combinations of (0,1) for cptcoveff variables. Outputting only combinations with people, éliminating 1 1 if
1018: race White (0 0), Black vs White (1 0), Hispanic (0 1) and 1 1 being meaningless.
1.218 brouard 1019:
1.226 brouard 1020:
1021:
1.133 brouard 1022: Authors: Nicolas Brouard (brouard@ined.fr) and Agnès Lièvre (lievre@ined.fr).
1023: Institut national d'études démographiques, Paris.
1.126 brouard 1024: This software have been partly granted by Euro-REVES, a concerted action
1025: from the European Union.
1026: It is copyrighted identically to a GNU software product, ie programme and
1027: software can be distributed freely for non commercial use. Latest version
1028: can be accessed at http://euroreves.ined.fr/imach .
1029:
1030: Help to debug: LD_PRELOAD=/usr/local/lib/libnjamd.so ./imach foo.imach
1031: or better on gdb : set env LD_PRELOAD=/usr/local/lib/libnjamd.so
1032:
1033: **********************************************************************/
1034: /*
1035: main
1036: read parameterfile
1037: read datafile
1038: concatwav
1039: freqsummary
1040: if (mle >= 1)
1041: mlikeli
1042: print results files
1043: if mle==1
1044: computes hessian
1045: read end of parameter file: agemin, agemax, bage, fage, estepm
1046: begin-prev-date,...
1047: open gnuplot file
1048: open html file
1.145 brouard 1049: period (stable) prevalence | pl_nom 1-1 2-2 etc by covariate
1050: for age prevalim() | #****** V1=0 V2=1 V3=1 V4=0 ******
1051: | 65 1 0 2 1 3 1 4 0 0.96326 0.03674
1052: freexexit2 possible for memory heap.
1053:
1054: h Pij x | pij_nom ficrestpij
1055: # Cov Agex agex+h hpijx with i,j= 1-1 1-2 1-3 2-1 2-2 2-3
1056: 1 85 85 1.00000 0.00000 0.00000 0.00000 1.00000 0.00000
1057: 1 85 86 0.68299 0.22291 0.09410 0.71093 0.00000 0.28907
1058:
1059: 1 65 99 0.00364 0.00322 0.99314 0.00350 0.00310 0.99340
1060: 1 65 100 0.00214 0.00204 0.99581 0.00206 0.00196 0.99597
1061: variance of p one-step probabilities varprob | prob_nom ficresprob #One-step probabilities and stand. devi in ()
1062: Standard deviation of one-step probabilities | probcor_nom ficresprobcor #One-step probabilities and correlation matrix
1063: Matrix of variance covariance of one-step probabilities | probcov_nom ficresprobcov #One-step probabilities and covariance matrix
1064:
1.126 brouard 1065: forecasting if prevfcast==1 prevforecast call prevalence()
1066: health expectancies
1067: Variance-covariance of DFLE
1068: prevalence()
1069: movingaverage()
1070: varevsij()
1071: if popbased==1 varevsij(,popbased)
1072: total life expectancies
1073: Variance of period (stable) prevalence
1074: end
1075: */
1076:
1.187 brouard 1077: /* #define DEBUG */
1078: /* #define DEBUGBRENT */
1.203 brouard 1079: /* #define DEBUGLINMIN */
1080: /* #define DEBUGHESS */
1081: #define DEBUGHESSIJ
1.224 brouard 1082: /* #define LINMINORIGINAL /\* Don't use loop on scale in linmin (accepting nan) *\/ */
1.165 brouard 1083: #define POWELL /* Instead of NLOPT */
1.224 brouard 1084: #define POWELLNOF3INFF1TEST /* Skip test */
1.186 brouard 1085: /* #define POWELLORIGINAL /\* Don't use Directest to decide new direction but original Powell test *\/ */
1086: /* #define MNBRAKORIGINAL /\* Don't use mnbrak fix *\/ */
1.126 brouard 1087:
1088: #include <math.h>
1089: #include <stdio.h>
1090: #include <stdlib.h>
1091: #include <string.h>
1.226 brouard 1092: #include <ctype.h>
1.159 brouard 1093:
1094: #ifdef _WIN32
1095: #include <io.h>
1.172 brouard 1096: #include <windows.h>
1097: #include <tchar.h>
1.159 brouard 1098: #else
1.126 brouard 1099: #include <unistd.h>
1.159 brouard 1100: #endif
1.126 brouard 1101:
1102: #include <limits.h>
1103: #include <sys/types.h>
1.171 brouard 1104:
1105: #if defined(__GNUC__)
1106: #include <sys/utsname.h> /* Doesn't work on Windows */
1107: #endif
1108:
1.126 brouard 1109: #include <sys/stat.h>
1110: #include <errno.h>
1.159 brouard 1111: /* extern int errno; */
1.126 brouard 1112:
1.157 brouard 1113: /* #ifdef LINUX */
1114: /* #include <time.h> */
1115: /* #include "timeval.h" */
1116: /* #else */
1117: /* #include <sys/time.h> */
1118: /* #endif */
1119:
1.126 brouard 1120: #include <time.h>
1121:
1.136 brouard 1122: #ifdef GSL
1123: #include <gsl/gsl_errno.h>
1124: #include <gsl/gsl_multimin.h>
1125: #endif
1126:
1.167 brouard 1127:
1.162 brouard 1128: #ifdef NLOPT
1129: #include <nlopt.h>
1130: typedef struct {
1131: double (* function)(double [] );
1132: } myfunc_data ;
1133: #endif
1134:
1.126 brouard 1135: /* #include <libintl.h> */
1136: /* #define _(String) gettext (String) */
1137:
1.251 brouard 1138: #define MAXLINE 2048 /* Was 256 and 1024. Overflow with 312 with 2 states and 4 covariates. Should be ok */
1.126 brouard 1139:
1140: #define GNUPLOTPROGRAM "gnuplot"
1141: /*#define GNUPLOTPROGRAM "..\\gp37mgw\\wgnuplot"*/
1142: #define FILENAMELENGTH 132
1143:
1144: #define GLOCK_ERROR_NOPATH -1 /* empty path */
1145: #define GLOCK_ERROR_GETCWD -2 /* cannot get cwd */
1146:
1.144 brouard 1147: #define MAXPARM 128 /**< Maximum number of parameters for the optimization */
1148: #define NPARMAX 64 /**< (nlstate+ndeath-1)*nlstate*ncovmodel */
1.126 brouard 1149:
1150: #define NINTERVMAX 8
1.144 brouard 1151: #define NLSTATEMAX 8 /**< Maximum number of live states (for func) */
1152: #define NDEATHMAX 8 /**< Maximum number of dead states (for func) */
1.291 brouard 1153: #define NCOVMAX 20 /**< Maximum number of covariates, including generated covariates V1*V2 */
1.197 brouard 1154: #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.211 brouard 1155: /*#define decodtabm(h,k,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (k-1)) & 1) +1 : -1)*/
1156: #define decodtabm(h,k,cptcoveff) (((h-1) >> (k-1)) & 1) +1
1.290 brouard 1157: /*#define MAXN 20000 */ /* Should by replaced by nobs, real number of observations and unlimited */
1.144 brouard 1158: #define YEARM 12. /**< Number of months per year */
1.218 brouard 1159: /* #define AGESUP 130 */
1.288 brouard 1160: /* #define AGESUP 150 */
1161: #define AGESUP 200
1.268 brouard 1162: #define AGEINF 0
1.218 brouard 1163: #define AGEMARGE 25 /* Marge for agemin and agemax for(iage=agemin-AGEMARGE; iage <= agemax+3+AGEMARGE; iage++) */
1.126 brouard 1164: #define AGEBASE 40
1.194 brouard 1165: #define AGEOVERFLOW 1.e20
1.164 brouard 1166: #define AGEGOMP 10 /**< Minimal age for Gompertz adjustment */
1.157 brouard 1167: #ifdef _WIN32
1168: #define DIRSEPARATOR '\\'
1169: #define CHARSEPARATOR "\\"
1170: #define ODIRSEPARATOR '/'
1171: #else
1.126 brouard 1172: #define DIRSEPARATOR '/'
1173: #define CHARSEPARATOR "/"
1174: #define ODIRSEPARATOR '\\'
1175: #endif
1176:
1.314 ! brouard 1177: /* $Id: imach.c,v 1.313 2022/04/11 15:57:42 brouard Exp $ */
1.126 brouard 1178: /* $State: Exp $ */
1.196 brouard 1179: #include "version.h"
1180: char version[]=__IMACH_VERSION__;
1.308 brouard 1181: char copyright[]="March 2021,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, INED 2000-2021";
1.314 ! brouard 1182: char fullversion[]="$Revision: 1.313 $ $Date: 2022/04/11 15:57:42 $";
1.126 brouard 1183: char strstart[80];
1184: char optionfilext[10], optionfilefiname[FILENAMELENGTH];
1.130 brouard 1185: int erreur=0, nberr=0, nbwarn=0; /* Error number, number of errors number of warnings */
1.187 brouard 1186: int nagesqr=0, nforce=0; /* nagesqr=1 if model is including age*age, number of forces */
1.145 brouard 1187: /* Number of covariates model=V2+V1+ V3*age+V2*V4 */
1188: int cptcovn=0; /**< cptcovn number of covariates added in the model (excepting constant and age and age*product) */
1189: int cptcovt=0; /**< cptcovt number of covariates added in the model (excepting constant and age) */
1.225 brouard 1190: int cptcovs=0; /**< cptcovs number of simple covariates in the model V2+V1 =2 */
1191: int cptcovsnq=0; /**< cptcovsnq number of simple covariates in the model but non quantitative V2+V1 =2 */
1.145 brouard 1192: int cptcovage=0; /**< Number of covariates with age: V3*age only =1 */
1193: int cptcovprodnoage=0; /**< Number of covariate products without age */
1194: int cptcoveff=0; /* Total number of covariates to vary for printing results */
1.233 brouard 1195: int ncovf=0; /* Total number of effective fixed covariates (dummy or quantitative) in the model */
1196: int ncovv=0; /* Total number of effective (wave) varying covariates (dummy or quantitative) in the model */
1.232 brouard 1197: int ncova=0; /* Total number of effective (wave and stepm) varying with age covariates (dummy of quantitative) in the model */
1.234 brouard 1198: int nsd=0; /**< Total number of single dummy variables (output) */
1199: int nsq=0; /**< Total number of single quantitative variables (output) */
1.232 brouard 1200: int ncoveff=0; /* Total number of effective fixed dummy covariates in the model */
1.225 brouard 1201: int nqfveff=0; /**< nqfveff Number of Quantitative Fixed Variables Effective */
1.224 brouard 1202: int ntveff=0; /**< ntveff number of effective time varying variables */
1203: int nqtveff=0; /**< ntqveff number of effective time varying quantitative variables */
1.145 brouard 1204: int cptcov=0; /* Working variable */
1.290 brouard 1205: int nobs=10; /* Number of observations in the data lastobs-firstobs */
1.218 brouard 1206: int ncovcombmax=NCOVMAX; /* Maximum calculated number of covariate combination = pow(2, cptcoveff) */
1.302 brouard 1207: int npar=NPARMAX; /* Number of parameters (nlstate+ndeath-1)*nlstate*ncovmodel; */
1.126 brouard 1208: int nlstate=2; /* Number of live states */
1209: int ndeath=1; /* Number of dead states */
1.130 brouard 1210: int ncovmodel=0, ncovcol=0; /* Total number of covariables including constant a12*1 +b12*x ncovmodel=2 */
1.223 brouard 1211: int nqv=0, ntv=0, nqtv=0; /* Total number of quantitative variables, time variable (dummy), quantitative and time variable */
1.126 brouard 1212: int popbased=0;
1213:
1214: int *wav; /* Number of waves for this individuual 0 is possible */
1.130 brouard 1215: int maxwav=0; /* Maxim number of waves */
1216: int jmin=0, jmax=0; /* min, max spacing between 2 waves */
1217: int ijmin=0, ijmax=0; /* Individuals having jmin and jmax */
1218: int gipmx=0, gsw=0; /* Global variables on the number of contributions
1.126 brouard 1219: to the likelihood and the sum of weights (done by funcone)*/
1.130 brouard 1220: int mle=1, weightopt=0;
1.126 brouard 1221: int **mw; /* mw[mi][i] is number of the mi wave for this individual */
1222: int **dh; /* dh[mi][i] is number of steps between mi,mi+1 for this individual */
1223: int **bh; /* bh[mi][i] is the bias (+ or -) for this individual if the delay between
1224: * wave mi and wave mi+1 is not an exact multiple of stepm. */
1.162 brouard 1225: int countcallfunc=0; /* Count the number of calls to func */
1.230 brouard 1226: int selected(int kvar); /* Is covariate kvar selected for printing results */
1227:
1.130 brouard 1228: double jmean=1; /* Mean space between 2 waves */
1.145 brouard 1229: double **matprod2(); /* test */
1.126 brouard 1230: double **oldm, **newm, **savm; /* Working pointers to matrices */
1231: double **oldms, **newms, **savms; /* Fixed working pointers to matrices */
1.218 brouard 1232: double **ddnewms, **ddoldms, **ddsavms; /* for freeing later */
1233:
1.136 brouard 1234: /*FILE *fic ; */ /* Used in readdata only */
1.217 brouard 1235: FILE *ficpar, *ficparo,*ficres, *ficresp, *ficresphtm, *ficresphtmfr, *ficrespl, *ficresplb,*ficrespij, *ficrespijb, *ficrest,*ficresf, *ficresfb,*ficrespop;
1.126 brouard 1236: FILE *ficlog, *ficrespow;
1.130 brouard 1237: int globpr=0; /* Global variable for printing or not */
1.126 brouard 1238: double fretone; /* Only one call to likelihood */
1.130 brouard 1239: long ipmx=0; /* Number of contributions */
1.126 brouard 1240: double sw; /* Sum of weights */
1241: char filerespow[FILENAMELENGTH];
1242: char fileresilk[FILENAMELENGTH]; /* File of individual contributions to the likelihood */
1243: FILE *ficresilk;
1244: FILE *ficgp,*ficresprob,*ficpop, *ficresprobcov, *ficresprobcor;
1245: FILE *ficresprobmorprev;
1246: FILE *fichtm, *fichtmcov; /* Html File */
1247: FILE *ficreseij;
1248: char filerese[FILENAMELENGTH];
1249: FILE *ficresstdeij;
1250: char fileresstde[FILENAMELENGTH];
1251: FILE *ficrescveij;
1252: char filerescve[FILENAMELENGTH];
1253: FILE *ficresvij;
1254: char fileresv[FILENAMELENGTH];
1.269 brouard 1255:
1.126 brouard 1256: char title[MAXLINE];
1.234 brouard 1257: char model[MAXLINE]; /**< The model line */
1.217 brouard 1258: char optionfile[FILENAMELENGTH], datafile[FILENAMELENGTH], filerespl[FILENAMELENGTH], fileresplb[FILENAMELENGTH];
1.126 brouard 1259: char plotcmd[FILENAMELENGTH], pplotcmd[FILENAMELENGTH];
1260: char tmpout[FILENAMELENGTH], tmpout2[FILENAMELENGTH];
1261: char command[FILENAMELENGTH];
1262: int outcmd=0;
1263:
1.217 brouard 1264: char fileres[FILENAMELENGTH], filerespij[FILENAMELENGTH], filerespijb[FILENAMELENGTH], filereso[FILENAMELENGTH], rfileres[FILENAMELENGTH];
1.202 brouard 1265: char fileresu[FILENAMELENGTH]; /* fileres without r in front */
1.126 brouard 1266: char filelog[FILENAMELENGTH]; /* Log file */
1267: char filerest[FILENAMELENGTH];
1268: char fileregp[FILENAMELENGTH];
1269: char popfile[FILENAMELENGTH];
1270:
1271: char optionfilegnuplot[FILENAMELENGTH], optionfilehtm[FILENAMELENGTH], optionfilehtmcov[FILENAMELENGTH] ;
1272:
1.157 brouard 1273: /* struct timeval start_time, end_time, curr_time, last_time, forecast_time; */
1274: /* struct timezone tzp; */
1275: /* extern int gettimeofday(); */
1276: struct tm tml, *gmtime(), *localtime();
1277:
1278: extern time_t time();
1279:
1280: struct tm start_time, end_time, curr_time, last_time, forecast_time;
1281: time_t rstart_time, rend_time, rcurr_time, rlast_time, rforecast_time; /* raw time */
1282: struct tm tm;
1283:
1.126 brouard 1284: char strcurr[80], strfor[80];
1285:
1286: char *endptr;
1287: long lval;
1288: double dval;
1289:
1290: #define NR_END 1
1291: #define FREE_ARG char*
1292: #define FTOL 1.0e-10
1293:
1294: #define NRANSI
1.240 brouard 1295: #define ITMAX 200
1296: #define ITPOWMAX 20 /* This is now multiplied by the number of parameters */
1.126 brouard 1297:
1298: #define TOL 2.0e-4
1299:
1300: #define CGOLD 0.3819660
1301: #define ZEPS 1.0e-10
1302: #define SHFT(a,b,c,d) (a)=(b);(b)=(c);(c)=(d);
1303:
1304: #define GOLD 1.618034
1305: #define GLIMIT 100.0
1306: #define TINY 1.0e-20
1307:
1308: static double maxarg1,maxarg2;
1309: #define FMAX(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)>(maxarg2)? (maxarg1):(maxarg2))
1310: #define FMIN(a,b) (maxarg1=(a),maxarg2=(b),(maxarg1)<(maxarg2)? (maxarg1):(maxarg2))
1311:
1312: #define SIGN(a,b) ((b)>0.0 ? fabs(a) : -fabs(a))
1313: #define rint(a) floor(a+0.5)
1.166 brouard 1314: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/myutils_8h-source.html */
1.183 brouard 1315: #define mytinydouble 1.0e-16
1.166 brouard 1316: /* #define DEQUAL(a,b) (fabs((a)-(b))<mytinydouble) */
1317: /* http://www.thphys.uni-heidelberg.de/~robbers/cmbeasy/doc/html/mynrutils_8h-source.html */
1318: /* static double dsqrarg; */
1319: /* #define DSQR(a) (DEQUAL((dsqrarg=(a)),0.0) ? 0.0 : dsqrarg*dsqrarg) */
1.126 brouard 1320: static double sqrarg;
1321: #define SQR(a) ((sqrarg=(a)) == 0.0 ? 0.0 :sqrarg*sqrarg)
1322: #define SWAP(a,b) {temp=(a);(a)=(b);(b)=temp;}
1323: int agegomp= AGEGOMP;
1324:
1325: int imx;
1326: int stepm=1;
1327: /* Stepm, step in month: minimum step interpolation*/
1328:
1329: int estepm;
1330: /* Estepm, step in month to interpolate survival function in order to approximate Life Expectancy*/
1331:
1332: int m,nb;
1333: long *num;
1.197 brouard 1334: int firstpass=0, lastpass=4,*cod, *cens;
1.192 brouard 1335: int *ncodemax; /* ncodemax[j]= Number of modalities of the j th
1336: covariate for which somebody answered excluding
1337: undefined. Usually 2: 0 and 1. */
1338: int *ncodemaxwundef; /* ncodemax[j]= Number of modalities of the j th
1339: covariate for which somebody answered including
1340: undefined. Usually 3: -1, 0 and 1. */
1.126 brouard 1341: double **agev,*moisnais, *annais, *moisdc, *andc,**mint, **anint;
1.218 brouard 1342: double **pmmij, ***probs; /* Global pointer */
1.219 brouard 1343: double ***mobaverage, ***mobaverages; /* New global variable */
1.126 brouard 1344: double *ageexmed,*agecens;
1345: double dateintmean=0;
1.296 brouard 1346: double anprojd, mprojd, jprojd; /* For eventual projections */
1347: double anprojf, mprojf, jprojf;
1.126 brouard 1348:
1.296 brouard 1349: double anbackd, mbackd, jbackd; /* For eventual backprojections */
1350: double anbackf, mbackf, jbackf;
1351: double jintmean,mintmean,aintmean;
1.126 brouard 1352: double *weight;
1353: int **s; /* Status */
1.141 brouard 1354: double *agedc;
1.145 brouard 1355: double **covar; /**< covar[j,i], value of jth covariate for individual i,
1.141 brouard 1356: * covar=matrix(0,NCOVMAX,1,n);
1.187 brouard 1357: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*age; */
1.268 brouard 1358: double **coqvar; /* Fixed quantitative covariate nqv */
1359: double ***cotvar; /* Time varying covariate ntv */
1.225 brouard 1360: double ***cotqvar; /* Time varying quantitative covariate itqv */
1.141 brouard 1361: double idx;
1362: int **nbcode, *Tvar; /**< model=V2 => Tvar[1]= 2 */
1.234 brouard 1363: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1364: /*k 1 2 3 4 5 6 7 8 9 */
1365: /*Tvar[k]= 5 4 3 6 5 2 7 1 1 */
1366: /* Tndvar[k] 1 2 3 4 5 */
1367: /*TDvar 4 3 6 7 1 */ /* For outputs only; combination of dummies fixed or varying */
1368: /* Tns[k] 1 2 2 4 5 */ /* Number of single cova */
1369: /* TvarsD[k] 1 2 3 */ /* Number of single dummy cova */
1370: /* TvarsDind 2 3 9 */ /* position K of single dummy cova */
1371: /* TvarsQ[k] 1 2 */ /* Number of single quantitative cova */
1372: /* TvarsQind 1 6 */ /* position K of single quantitative cova */
1373: /* Tprod[i]=k 4 7 */
1374: /* Tage[i]=k 5 8 */
1375: /* */
1376: /* Type */
1377: /* V 1 2 3 4 5 */
1378: /* F F V V V */
1379: /* D Q D D Q */
1380: /* */
1381: int *TvarsD;
1382: int *TvarsDind;
1383: int *TvarsQ;
1384: int *TvarsQind;
1385:
1.235 brouard 1386: #define MAXRESULTLINES 10
1387: int nresult=0;
1.258 brouard 1388: int parameterline=0; /* # of the parameter (type) line */
1.235 brouard 1389: int TKresult[MAXRESULTLINES];
1.237 brouard 1390: int Tresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1391: int Tinvresult[MAXRESULTLINES][NCOVMAX];/* For dummy variable , value (output) */
1.235 brouard 1392: int Tvresult[MAXRESULTLINES][NCOVMAX]; /* For dummy variable , variable # (output) */
1393: double Tqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.237 brouard 1394: double Tqinvresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , value (output) */
1.235 brouard 1395: int Tvqresult[MAXRESULTLINES][NCOVMAX]; /* For quantitative variable , variable # (output) */
1396:
1.234 brouard 1397: /* 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 1398: 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 */
1399: 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 */
1400: 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 */
1401: int *TvarVind; /**< TvarVind[1]=1, TvarVind[2]=2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1402: 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 */
1403: 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 1404: int *TvarFD; /**< TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1405: int *TvarFDind; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1406: int *TvarFQ; /* TvarFQ[1]=V2 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1407: int *TvarFQind; /* TvarFQind[1]=6 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1408: int *TvarVD; /* TvarVD[1]=V5 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1409: int *TvarVDind; /* TvarVDind[1]=1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */ /* Only simple fixed quantitative variable */
1410: 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 */
1411: 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 */
1412:
1.230 brouard 1413: int *Tvarsel; /**< Selected covariates for output */
1414: double *Tvalsel; /**< Selected modality value of covariate for output */
1.226 brouard 1415: int *Typevar; /**< 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product */
1.227 brouard 1416: int *Fixed; /** Fixed[k] 0=fixed, 1 varying, 2 fixed with age product, 3 varying with age product */
1417: 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 1418: int *DummyV; /** Dummy[v] 0=dummy (0 1), 1 quantitative */
1419: int *FixedV; /** FixedV[v] 0 fixed, 1 varying */
1.197 brouard 1420: int *Tage;
1.227 brouard 1421: int anyvaryingduminmodel=0; /**< Any varying dummy in Model=1 yes, 0 no, to avoid a loop on waves in freq */
1.228 brouard 1422: 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 1423: 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*/
1424: 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 1425: int *Ndum; /** Freq of modality (tricode */
1.200 brouard 1426: /* int **codtab;*/ /**< codtab=imatrix(1,100,1,10); */
1.227 brouard 1427: int **Tvard;
1428: int *Tprod;/**< Gives the k position of the k1 product */
1.238 brouard 1429: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3 */
1.227 brouard 1430: int *Tposprod; /**< Gives the k1 product from the k position */
1.238 brouard 1431: /* if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2) */
1432: /* Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5(V3*V2)]=2 (2nd product without age) */
1.227 brouard 1433: int cptcovprod, *Tvaraff, *invalidvarcomb;
1.126 brouard 1434: double *lsurv, *lpop, *tpop;
1435:
1.231 brouard 1436: #define FD 1; /* Fixed dummy covariate */
1437: #define FQ 2; /* Fixed quantitative covariate */
1438: #define FP 3; /* Fixed product covariate */
1439: #define FPDD 7; /* Fixed product dummy*dummy covariate */
1440: #define FPDQ 8; /* Fixed product dummy*quantitative covariate */
1441: #define FPQQ 9; /* Fixed product quantitative*quantitative covariate */
1442: #define VD 10; /* Varying dummy covariate */
1443: #define VQ 11; /* Varying quantitative covariate */
1444: #define VP 12; /* Varying product covariate */
1445: #define VPDD 13; /* Varying product dummy*dummy covariate */
1446: #define VPDQ 14; /* Varying product dummy*quantitative covariate */
1447: #define VPQQ 15; /* Varying product quantitative*quantitative covariate */
1448: #define APFD 16; /* Age product * fixed dummy covariate */
1449: #define APFQ 17; /* Age product * fixed quantitative covariate */
1450: #define APVD 18; /* Age product * varying dummy covariate */
1451: #define APVQ 19; /* Age product * varying quantitative covariate */
1452:
1453: #define FTYPE 1; /* Fixed covariate */
1454: #define VTYPE 2; /* Varying covariate (loop in wave) */
1455: #define ATYPE 2; /* Age product covariate (loop in dh within wave)*/
1456:
1457: struct kmodel{
1458: int maintype; /* main type */
1459: int subtype; /* subtype */
1460: };
1461: struct kmodel modell[NCOVMAX];
1462:
1.143 brouard 1463: double ftol=FTOL; /**< Tolerance for computing Max Likelihood */
1464: double ftolhess; /**< Tolerance for computing hessian */
1.126 brouard 1465:
1466: /**************** split *************************/
1467: static int split( char *path, char *dirc, char *name, char *ext, char *finame )
1468: {
1469: /* From a file name with (full) path (either Unix or Windows) we extract the directory (dirc)
1470: the name of the file (name), its extension only (ext) and its first part of the name (finame)
1471: */
1472: char *ss; /* pointer */
1.186 brouard 1473: int l1=0, l2=0; /* length counters */
1.126 brouard 1474:
1475: l1 = strlen(path ); /* length of path */
1476: if ( l1 == 0 ) return( GLOCK_ERROR_NOPATH );
1477: ss= strrchr( path, DIRSEPARATOR ); /* find last / */
1478: if ( ss == NULL ) { /* no directory, so determine current directory */
1479: strcpy( name, path ); /* we got the fullname name because no directory */
1480: /*if(strrchr(path, ODIRSEPARATOR )==NULL)
1481: printf("Warning you should use %s as a separator\n",DIRSEPARATOR);*/
1482: /* get current working directory */
1483: /* extern char* getcwd ( char *buf , int len);*/
1.184 brouard 1484: #ifdef WIN32
1485: if (_getcwd( dirc, FILENAME_MAX ) == NULL ) {
1486: #else
1487: if (getcwd(dirc, FILENAME_MAX) == NULL) {
1488: #endif
1.126 brouard 1489: return( GLOCK_ERROR_GETCWD );
1490: }
1491: /* got dirc from getcwd*/
1492: printf(" DIRC = %s \n",dirc);
1.205 brouard 1493: } else { /* strip directory from path */
1.126 brouard 1494: ss++; /* after this, the filename */
1495: l2 = strlen( ss ); /* length of filename */
1496: if ( l2 == 0 ) return( GLOCK_ERROR_NOPATH );
1497: strcpy( name, ss ); /* save file name */
1498: strncpy( dirc, path, l1 - l2 ); /* now the directory */
1.186 brouard 1499: dirc[l1-l2] = '\0'; /* add zero */
1.126 brouard 1500: printf(" DIRC2 = %s \n",dirc);
1501: }
1502: /* We add a separator at the end of dirc if not exists */
1503: l1 = strlen( dirc ); /* length of directory */
1504: if( dirc[l1-1] != DIRSEPARATOR ){
1505: dirc[l1] = DIRSEPARATOR;
1506: dirc[l1+1] = 0;
1507: printf(" DIRC3 = %s \n",dirc);
1508: }
1509: ss = strrchr( name, '.' ); /* find last / */
1510: if (ss >0){
1511: ss++;
1512: strcpy(ext,ss); /* save extension */
1513: l1= strlen( name);
1514: l2= strlen(ss)+1;
1515: strncpy( finame, name, l1-l2);
1516: finame[l1-l2]= 0;
1517: }
1518:
1519: return( 0 ); /* we're done */
1520: }
1521:
1522:
1523: /******************************************/
1524:
1525: void replace_back_to_slash(char *s, char*t)
1526: {
1527: int i;
1528: int lg=0;
1529: i=0;
1530: lg=strlen(t);
1531: for(i=0; i<= lg; i++) {
1532: (s[i] = t[i]);
1533: if (t[i]== '\\') s[i]='/';
1534: }
1535: }
1536:
1.132 brouard 1537: char *trimbb(char *out, char *in)
1.137 brouard 1538: { /* Trim multiple blanks in line but keeps first blanks if line starts with blanks */
1.132 brouard 1539: char *s;
1540: s=out;
1541: while (*in != '\0'){
1.137 brouard 1542: while( *in == ' ' && *(in+1) == ' '){ /* && *(in+1) != '\0'){*/
1.132 brouard 1543: in++;
1544: }
1545: *out++ = *in++;
1546: }
1547: *out='\0';
1548: return s;
1549: }
1550:
1.187 brouard 1551: /* char *substrchaine(char *out, char *in, char *chain) */
1552: /* { */
1553: /* /\* Substract chain 'chain' from 'in', return and output 'out' *\/ */
1554: /* char *s, *t; */
1555: /* t=in;s=out; */
1556: /* while ((*in != *chain) && (*in != '\0')){ */
1557: /* *out++ = *in++; */
1558: /* } */
1559:
1560: /* /\* *in matches *chain *\/ */
1561: /* while ((*in++ == *chain++) && (*in != '\0')){ */
1562: /* printf("*in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1563: /* } */
1564: /* in--; chain--; */
1565: /* while ( (*in != '\0')){ */
1566: /* printf("Bef *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1567: /* *out++ = *in++; */
1568: /* printf("Aft *in = %c, *out= %c *chain= %c \n", *in, *out, *chain); */
1569: /* } */
1570: /* *out='\0'; */
1571: /* out=s; */
1572: /* return out; */
1573: /* } */
1574: char *substrchaine(char *out, char *in, char *chain)
1575: {
1576: /* Substract chain 'chain' from 'in', return and output 'out' */
1577: /* in="V1+V1*age+age*age+V2", chain="age*age" */
1578:
1579: char *strloc;
1580:
1581: strcpy (out, in);
1582: strloc = strstr(out, chain); /* strloc points to out at age*age+V2 */
1583: printf("Bef strloc=%s chain=%s out=%s \n", strloc, chain, out);
1584: if(strloc != NULL){
1585: /* will affect out */ /* strloc+strlenc(chain)=+V2 */ /* Will also work in Unicode */
1586: memmove(strloc,strloc+strlen(chain), strlen(strloc+strlen(chain))+1);
1587: /* strcpy (strloc, strloc +strlen(chain));*/
1588: }
1589: printf("Aft strloc=%s chain=%s in=%s out=%s \n", strloc, chain, in, out);
1590: return out;
1591: }
1592:
1593:
1.145 brouard 1594: char *cutl(char *blocc, char *alocc, char *in, char occ)
1595: {
1.187 brouard 1596: /* cuts string in into blocc and alocc where blocc ends before FIRST occurence of char 'occ'
1.145 brouard 1597: and alocc starts after first occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1.310 brouard 1598: gives alocc="abcdef" and blocc="ghi2j".
1.145 brouard 1599: If occ is not found blocc is null and alocc is equal to in. Returns blocc
1600: */
1.160 brouard 1601: char *s, *t;
1.145 brouard 1602: t=in;s=in;
1603: while ((*in != occ) && (*in != '\0')){
1604: *alocc++ = *in++;
1605: }
1606: if( *in == occ){
1607: *(alocc)='\0';
1608: s=++in;
1609: }
1610:
1611: if (s == t) {/* occ not found */
1612: *(alocc-(in-s))='\0';
1613: in=s;
1614: }
1615: while ( *in != '\0'){
1616: *blocc++ = *in++;
1617: }
1618:
1619: *blocc='\0';
1620: return t;
1621: }
1.137 brouard 1622: char *cutv(char *blocc, char *alocc, char *in, char occ)
1623: {
1.187 brouard 1624: /* cuts string in into blocc and alocc where blocc ends before LAST occurence of char 'occ'
1.137 brouard 1625: and alocc starts after last occurence of char 'occ' : ex cutv(blocc,alocc,"abcdef2ghi2j",'2')
1626: gives blocc="abcdef2ghi" and alocc="j".
1627: If occ is not found blocc is null and alocc is equal to in. Returns alocc
1628: */
1629: char *s, *t;
1630: t=in;s=in;
1631: while (*in != '\0'){
1632: while( *in == occ){
1633: *blocc++ = *in++;
1634: s=in;
1635: }
1636: *blocc++ = *in++;
1637: }
1638: if (s == t) /* occ not found */
1639: *(blocc-(in-s))='\0';
1640: else
1641: *(blocc-(in-s)-1)='\0';
1642: in=s;
1643: while ( *in != '\0'){
1644: *alocc++ = *in++;
1645: }
1646:
1647: *alocc='\0';
1648: return s;
1649: }
1650:
1.126 brouard 1651: int nbocc(char *s, char occ)
1652: {
1653: int i,j=0;
1654: int lg=20;
1655: i=0;
1656: lg=strlen(s);
1657: for(i=0; i<= lg; i++) {
1.234 brouard 1658: if (s[i] == occ ) j++;
1.126 brouard 1659: }
1660: return j;
1661: }
1662:
1.137 brouard 1663: /* void cutv(char *u,char *v, char*t, char occ) */
1664: /* { */
1665: /* /\* cuts string t into u and v where u ends before last occurence of char 'occ' */
1666: /* and v starts after last occurence of char 'occ' : ex cutv(u,v,"abcdef2ghi2j",'2') */
1667: /* gives u="abcdef2ghi" and v="j" *\/ */
1668: /* int i,lg,j,p=0; */
1669: /* i=0; */
1670: /* lg=strlen(t); */
1671: /* for(j=0; j<=lg-1; j++) { */
1672: /* if((t[j]!= occ) && (t[j+1]== occ)) p=j+1; */
1673: /* } */
1.126 brouard 1674:
1.137 brouard 1675: /* for(j=0; j<p; j++) { */
1676: /* (u[j] = t[j]); */
1677: /* } */
1678: /* u[p]='\0'; */
1.126 brouard 1679:
1.137 brouard 1680: /* for(j=0; j<= lg; j++) { */
1681: /* if (j>=(p+1))(v[j-p-1] = t[j]); */
1682: /* } */
1683: /* } */
1.126 brouard 1684:
1.160 brouard 1685: #ifdef _WIN32
1686: char * strsep(char **pp, const char *delim)
1687: {
1688: char *p, *q;
1689:
1690: if ((p = *pp) == NULL)
1691: return 0;
1692: if ((q = strpbrk (p, delim)) != NULL)
1693: {
1694: *pp = q + 1;
1695: *q = '\0';
1696: }
1697: else
1698: *pp = 0;
1699: return p;
1700: }
1701: #endif
1702:
1.126 brouard 1703: /********************** nrerror ********************/
1704:
1705: void nrerror(char error_text[])
1706: {
1707: fprintf(stderr,"ERREUR ...\n");
1708: fprintf(stderr,"%s\n",error_text);
1709: exit(EXIT_FAILURE);
1710: }
1711: /*********************** vector *******************/
1712: double *vector(int nl, int nh)
1713: {
1714: double *v;
1715: v=(double *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(double)));
1716: if (!v) nrerror("allocation failure in vector");
1717: return v-nl+NR_END;
1718: }
1719:
1720: /************************ free vector ******************/
1721: void free_vector(double*v, int nl, int nh)
1722: {
1723: free((FREE_ARG)(v+nl-NR_END));
1724: }
1725:
1726: /************************ivector *******************************/
1727: int *ivector(long nl,long nh)
1728: {
1729: int *v;
1730: v=(int *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(int)));
1731: if (!v) nrerror("allocation failure in ivector");
1732: return v-nl+NR_END;
1733: }
1734:
1735: /******************free ivector **************************/
1736: void free_ivector(int *v, long nl, long nh)
1737: {
1738: free((FREE_ARG)(v+nl-NR_END));
1739: }
1740:
1741: /************************lvector *******************************/
1742: long *lvector(long nl,long nh)
1743: {
1744: long *v;
1745: v=(long *) malloc((size_t)((nh-nl+1+NR_END)*sizeof(long)));
1746: if (!v) nrerror("allocation failure in ivector");
1747: return v-nl+NR_END;
1748: }
1749:
1750: /******************free lvector **************************/
1751: void free_lvector(long *v, long nl, long nh)
1752: {
1753: free((FREE_ARG)(v+nl-NR_END));
1754: }
1755:
1756: /******************* imatrix *******************************/
1757: int **imatrix(long nrl, long nrh, long ncl, long nch)
1758: /* allocate a int matrix with subscript range m[nrl..nrh][ncl..nch] */
1759: {
1760: long i, nrow=nrh-nrl+1,ncol=nch-ncl+1;
1761: int **m;
1762:
1763: /* allocate pointers to rows */
1764: m=(int **) malloc((size_t)((nrow+NR_END)*sizeof(int*)));
1765: if (!m) nrerror("allocation failure 1 in matrix()");
1766: m += NR_END;
1767: m -= nrl;
1768:
1769:
1770: /* allocate rows and set pointers to them */
1771: m[nrl]=(int *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(int)));
1772: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1773: m[nrl] += NR_END;
1774: m[nrl] -= ncl;
1775:
1776: for(i=nrl+1;i<=nrh;i++) m[i]=m[i-1]+ncol;
1777:
1778: /* return pointer to array of pointers to rows */
1779: return m;
1780: }
1781:
1782: /****************** free_imatrix *************************/
1783: void free_imatrix(m,nrl,nrh,ncl,nch)
1784: int **m;
1785: long nch,ncl,nrh,nrl;
1786: /* free an int matrix allocated by imatrix() */
1787: {
1788: free((FREE_ARG) (m[nrl]+ncl-NR_END));
1789: free((FREE_ARG) (m+nrl-NR_END));
1790: }
1791:
1792: /******************* matrix *******************************/
1793: double **matrix(long nrl, long nrh, long ncl, long nch)
1794: {
1795: long i, nrow=nrh-nrl+1, ncol=nch-ncl+1;
1796: double **m;
1797:
1798: m=(double **) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1799: if (!m) nrerror("allocation failure 1 in matrix()");
1800: m += NR_END;
1801: m -= nrl;
1802:
1803: m[nrl]=(double *) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1804: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1805: m[nrl] += NR_END;
1806: m[nrl] -= ncl;
1807:
1808: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1809: return m;
1.145 brouard 1810: /* print *(*(m+1)+70) or print m[1][70]; print m+1 or print &(m[1]) or &(m[1][0])
1811: m[i] = address of ith row of the table. &(m[i]) is its value which is another adress
1812: that of m[i][0]. In order to get the value p m[i][0] but it is unitialized.
1.126 brouard 1813: */
1814: }
1815:
1816: /*************************free matrix ************************/
1817: void free_matrix(double **m, long nrl, long nrh, long ncl, long nch)
1818: {
1819: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1820: free((FREE_ARG)(m+nrl-NR_END));
1821: }
1822:
1823: /******************* ma3x *******************************/
1824: double ***ma3x(long nrl, long nrh, long ncl, long nch, long nll, long nlh)
1825: {
1826: long i, j, nrow=nrh-nrl+1, ncol=nch-ncl+1, nlay=nlh-nll+1;
1827: double ***m;
1828:
1829: m=(double ***) malloc((size_t)((nrow+NR_END)*sizeof(double*)));
1830: if (!m) nrerror("allocation failure 1 in matrix()");
1831: m += NR_END;
1832: m -= nrl;
1833:
1834: m[nrl]=(double **) malloc((size_t)((nrow*ncol+NR_END)*sizeof(double)));
1835: if (!m[nrl]) nrerror("allocation failure 2 in matrix()");
1836: m[nrl] += NR_END;
1837: m[nrl] -= ncl;
1838:
1839: for (i=nrl+1; i<=nrh; i++) m[i]=m[i-1]+ncol;
1840:
1841: m[nrl][ncl]=(double *) malloc((size_t)((nrow*ncol*nlay+NR_END)*sizeof(double)));
1842: if (!m[nrl][ncl]) nrerror("allocation failure 3 in matrix()");
1843: m[nrl][ncl] += NR_END;
1844: m[nrl][ncl] -= nll;
1845: for (j=ncl+1; j<=nch; j++)
1846: m[nrl][j]=m[nrl][j-1]+nlay;
1847:
1848: for (i=nrl+1; i<=nrh; i++) {
1849: m[i][ncl]=m[i-1l][ncl]+ncol*nlay;
1850: for (j=ncl+1; j<=nch; j++)
1851: m[i][j]=m[i][j-1]+nlay;
1852: }
1853: return m;
1854: /* gdb: p *(m+1) <=> p m[1] and p (m+1) <=> p (m+1) <=> p &(m[1])
1855: &(m[i][j][k]) <=> *((*(m+i) + j)+k)
1856: */
1857: }
1858:
1859: /*************************free ma3x ************************/
1860: void free_ma3x(double ***m, long nrl, long nrh, long ncl, long nch,long nll, long nlh)
1861: {
1862: free((FREE_ARG)(m[nrl][ncl]+ nll-NR_END));
1863: free((FREE_ARG)(m[nrl]+ncl-NR_END));
1864: free((FREE_ARG)(m+nrl-NR_END));
1865: }
1866:
1867: /*************** function subdirf ***********/
1868: char *subdirf(char fileres[])
1869: {
1870: /* Caution optionfilefiname is hidden */
1871: strcpy(tmpout,optionfilefiname);
1872: strcat(tmpout,"/"); /* Add to the right */
1873: strcat(tmpout,fileres);
1874: return tmpout;
1875: }
1876:
1877: /*************** function subdirf2 ***********/
1878: char *subdirf2(char fileres[], char *preop)
1879: {
1.314 ! brouard 1880: /* Example subdirf2(optionfilefiname,"FB_") with optionfilefiname="texte", result="texte/FB_texte"
! 1881: Errors in subdirf, 2, 3 while printing tmpout is
! 1882: rewritten within the same printf. Workaround: many printfs
1.126 brouard 1883: /* Caution optionfilefiname is hidden */
1884: strcpy(tmpout,optionfilefiname);
1885: strcat(tmpout,"/");
1886: strcat(tmpout,preop);
1887: strcat(tmpout,fileres);
1888: return tmpout;
1889: }
1890:
1891: /*************** function subdirf3 ***********/
1892: char *subdirf3(char fileres[], char *preop, char *preop2)
1893: {
1894:
1895: /* Caution optionfilefiname is hidden */
1896: strcpy(tmpout,optionfilefiname);
1897: strcat(tmpout,"/");
1898: strcat(tmpout,preop);
1899: strcat(tmpout,preop2);
1900: strcat(tmpout,fileres);
1901: return tmpout;
1902: }
1.213 brouard 1903:
1904: /*************** function subdirfext ***********/
1905: char *subdirfext(char fileres[], char *preop, char *postop)
1906: {
1907:
1908: strcpy(tmpout,preop);
1909: strcat(tmpout,fileres);
1910: strcat(tmpout,postop);
1911: return tmpout;
1912: }
1.126 brouard 1913:
1.213 brouard 1914: /*************** function subdirfext3 ***********/
1915: char *subdirfext3(char fileres[], char *preop, char *postop)
1916: {
1917:
1918: /* Caution optionfilefiname is hidden */
1919: strcpy(tmpout,optionfilefiname);
1920: strcat(tmpout,"/");
1921: strcat(tmpout,preop);
1922: strcat(tmpout,fileres);
1923: strcat(tmpout,postop);
1924: return tmpout;
1925: }
1926:
1.162 brouard 1927: char *asc_diff_time(long time_sec, char ascdiff[])
1928: {
1929: long sec_left, days, hours, minutes;
1930: days = (time_sec) / (60*60*24);
1931: sec_left = (time_sec) % (60*60*24);
1932: hours = (sec_left) / (60*60) ;
1933: sec_left = (sec_left) %(60*60);
1934: minutes = (sec_left) /60;
1935: sec_left = (sec_left) % (60);
1936: sprintf(ascdiff,"%ld day(s) %ld hour(s) %ld minute(s) %ld second(s)",days, hours, minutes, sec_left);
1937: return ascdiff;
1938: }
1939:
1.126 brouard 1940: /***************** f1dim *************************/
1941: extern int ncom;
1942: extern double *pcom,*xicom;
1943: extern double (*nrfunc)(double []);
1944:
1945: double f1dim(double x)
1946: {
1947: int j;
1948: double f;
1949: double *xt;
1950:
1951: xt=vector(1,ncom);
1952: for (j=1;j<=ncom;j++) xt[j]=pcom[j]+x*xicom[j];
1953: f=(*nrfunc)(xt);
1954: free_vector(xt,1,ncom);
1955: return f;
1956: }
1957:
1958: /*****************brent *************************/
1959: double brent(double ax, double bx, double cx, double (*f)(double), double tol, double *xmin)
1.187 brouard 1960: {
1961: /* Given a function f, and given a bracketing triplet of abscissas ax, bx, cx (such that bx is
1962: * between ax and cx, and f(bx) is less than both f(ax) and f(cx) ), this routine isolates
1963: * the minimum to a fractional precision of about tol using Brent’s method. The abscissa of
1964: * the minimum is returned as xmin, and the minimum function value is returned as brent , the
1965: * returned function value.
1966: */
1.126 brouard 1967: int iter;
1968: double a,b,d,etemp;
1.159 brouard 1969: double fu=0,fv,fw,fx;
1.164 brouard 1970: double ftemp=0.;
1.126 brouard 1971: double p,q,r,tol1,tol2,u,v,w,x,xm;
1972: double e=0.0;
1973:
1974: a=(ax < cx ? ax : cx);
1975: b=(ax > cx ? ax : cx);
1976: x=w=v=bx;
1977: fw=fv=fx=(*f)(x);
1978: for (iter=1;iter<=ITMAX;iter++) {
1979: xm=0.5*(a+b);
1980: tol2=2.0*(tol1=tol*fabs(x)+ZEPS);
1981: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret)))*/
1982: printf(".");fflush(stdout);
1983: fprintf(ficlog,".");fflush(ficlog);
1.162 brouard 1984: #ifdef DEBUGBRENT
1.126 brouard 1985: 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);
1986: 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);
1987: /* if ((fabs(x-xm) <= (tol2-0.5*(b-a)))||(2.0*fabs(fu-ftemp) <= ftol*1.e-2*(fabs(fu)+fabs(ftemp)))) { */
1988: #endif
1989: if (fabs(x-xm) <= (tol2-0.5*(b-a))){
1990: *xmin=x;
1991: return fx;
1992: }
1993: ftemp=fu;
1994: if (fabs(e) > tol1) {
1995: r=(x-w)*(fx-fv);
1996: q=(x-v)*(fx-fw);
1997: p=(x-v)*q-(x-w)*r;
1998: q=2.0*(q-r);
1999: if (q > 0.0) p = -p;
2000: q=fabs(q);
2001: etemp=e;
2002: e=d;
2003: if (fabs(p) >= fabs(0.5*q*etemp) || p <= q*(a-x) || p >= q*(b-x))
1.224 brouard 2004: d=CGOLD*(e=(x >= xm ? a-x : b-x));
1.126 brouard 2005: else {
1.224 brouard 2006: d=p/q;
2007: u=x+d;
2008: if (u-a < tol2 || b-u < tol2)
2009: d=SIGN(tol1,xm-x);
1.126 brouard 2010: }
2011: } else {
2012: d=CGOLD*(e=(x >= xm ? a-x : b-x));
2013: }
2014: u=(fabs(d) >= tol1 ? x+d : x+SIGN(tol1,d));
2015: fu=(*f)(u);
2016: if (fu <= fx) {
2017: if (u >= x) a=x; else b=x;
2018: SHFT(v,w,x,u)
1.183 brouard 2019: SHFT(fv,fw,fx,fu)
2020: } else {
2021: if (u < x) a=u; else b=u;
2022: if (fu <= fw || w == x) {
1.224 brouard 2023: v=w;
2024: w=u;
2025: fv=fw;
2026: fw=fu;
1.183 brouard 2027: } else if (fu <= fv || v == x || v == w) {
1.224 brouard 2028: v=u;
2029: fv=fu;
1.183 brouard 2030: }
2031: }
1.126 brouard 2032: }
2033: nrerror("Too many iterations in brent");
2034: *xmin=x;
2035: return fx;
2036: }
2037:
2038: /****************** mnbrak ***********************/
2039:
2040: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb, double *fc,
2041: double (*func)(double))
1.183 brouard 2042: { /* Given a function func , and given distinct initial points ax and bx , this routine searches in
2043: the downhill direction (defined by the function as evaluated at the initial points) and returns
2044: new points ax , bx , cx that bracket a minimum of the function. Also returned are the function
2045: values at the three points, fa, fb , and fc such that fa > fb and fb < fc.
2046: */
1.126 brouard 2047: double ulim,u,r,q, dum;
2048: double fu;
1.187 brouard 2049:
2050: double scale=10.;
2051: int iterscale=0;
2052:
2053: *fa=(*func)(*ax); /* xta[j]=pcom[j]+(*ax)*xicom[j]; fa=f(xta[j])*/
2054: *fb=(*func)(*bx); /* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) */
2055:
2056:
2057: /* while(*fb != *fb){ /\* *ax should be ok, reducing distance to *ax *\/ */
2058: /* printf("Warning mnbrak *fb = %lf, *bx=%lf *ax=%lf *fa==%lf iter=%d\n",*fb, *bx, *ax, *fa, iterscale++); */
2059: /* *bx = *ax - (*ax - *bx)/scale; */
2060: /* *fb=(*func)(*bx); /\* xtb[j]=pcom[j]+(*bx)*xicom[j]; fb=f(xtb[j]) *\/ */
2061: /* } */
2062:
1.126 brouard 2063: if (*fb > *fa) {
2064: SHFT(dum,*ax,*bx,dum)
1.183 brouard 2065: SHFT(dum,*fb,*fa,dum)
2066: }
1.126 brouard 2067: *cx=(*bx)+GOLD*(*bx-*ax);
2068: *fc=(*func)(*cx);
1.183 brouard 2069: #ifdef DEBUG
1.224 brouard 2070: printf("mnbrak0 a=%lf *fa=%lf, b=%lf *fb=%lf, c=%lf *fc=%lf\n",*ax,*fa,*bx,*fb,*cx, *fc);
2071: 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 2072: #endif
1.224 brouard 2073: 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 2074: r=(*bx-*ax)*(*fb-*fc);
1.224 brouard 2075: q=(*bx-*cx)*(*fb-*fa); /* What if fa=inf */
1.126 brouard 2076: u=(*bx)-((*bx-*cx)*q-(*bx-*ax)*r)/
1.183 brouard 2077: (2.0*SIGN(FMAX(fabs(q-r),TINY),q-r)); /* Minimum abscissa of a parabolic estimated from (a,fa), (b,fb) and (c,fc). */
2078: ulim=(*bx)+GLIMIT*(*cx-*bx); /* Maximum abscissa where function should be evaluated */
2079: if ((*bx-u)*(u-*cx) > 0.0) { /* if u_p is between b and c */
1.126 brouard 2080: fu=(*func)(u);
1.163 brouard 2081: #ifdef DEBUG
2082: /* f(x)=A(x-u)**2+f(u) */
2083: double A, fparabu;
2084: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2085: fparabu= *fa - A*(*ax-u)*(*ax-u);
1.224 brouard 2086: 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);
2087: 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 2088: /* And thus,it can be that fu > *fc even if fparabu < *fc */
2089: /* mnbrak (*ax=7.666299858533, *fa=299039.693133272231), (*bx=8.595447774979, *fb=298976.598289369489),
2090: (*cx=10.098840694817, *fc=298946.631474258087), (*u=9.852501168332, fu=298948.773013752128, fparabu=298945.434711494134) */
2091: /* In that case, there is no bracket in the output! Routine is wrong with many consequences.*/
1.163 brouard 2092: #endif
1.184 brouard 2093: #ifdef MNBRAKORIGINAL
1.183 brouard 2094: #else
1.191 brouard 2095: /* if (fu > *fc) { */
2096: /* #ifdef DEBUG */
2097: /* printf("mnbrak4 fu > fc \n"); */
2098: /* fprintf(ficlog, "mnbrak4 fu > fc\n"); */
2099: /* #endif */
2100: /* /\* 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 *\\/ *\/ */
2101: /* /\* SHFT(*fa,*fc,fu,*fc) /\\* (b, u, c) is a bracket while test fb > fc will be fu > fc will exit *\\/ *\/ */
2102: /* dum=u; /\* Shifting c and u *\/ */
2103: /* u = *cx; */
2104: /* *cx = dum; */
2105: /* dum = fu; */
2106: /* fu = *fc; */
2107: /* *fc =dum; */
2108: /* } else { /\* end *\/ */
2109: /* #ifdef DEBUG */
2110: /* printf("mnbrak3 fu < fc \n"); */
2111: /* fprintf(ficlog, "mnbrak3 fu < fc\n"); */
2112: /* #endif */
2113: /* dum=u; /\* Shifting c and u *\/ */
2114: /* u = *cx; */
2115: /* *cx = dum; */
2116: /* dum = fu; */
2117: /* fu = *fc; */
2118: /* *fc =dum; */
2119: /* } */
1.224 brouard 2120: #ifdef DEBUGMNBRAK
2121: double A, fparabu;
2122: A= (*fb - *fa)/(*bx-*ax)/(*bx+*ax-2*u);
2123: fparabu= *fa - A*(*ax-u)*(*ax-u);
2124: 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);
2125: 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 2126: #endif
1.191 brouard 2127: dum=u; /* Shifting c and u */
2128: u = *cx;
2129: *cx = dum;
2130: dum = fu;
2131: fu = *fc;
2132: *fc =dum;
1.183 brouard 2133: #endif
1.162 brouard 2134: } else if ((*cx-u)*(u-ulim) > 0.0) { /* u is after c but before ulim */
1.183 brouard 2135: #ifdef DEBUG
1.224 brouard 2136: printf("\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
2137: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim\n",u,*cx);
1.183 brouard 2138: #endif
1.126 brouard 2139: fu=(*func)(u);
2140: if (fu < *fc) {
1.183 brouard 2141: #ifdef DEBUG
1.224 brouard 2142: printf("\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2143: fprintf(ficlog,"\nmnbrak2 u=%lf after c=%lf but before ulim=%lf AND fu=%lf < %lf=fc\n",u,*cx,ulim,fu, *fc);
2144: #endif
2145: SHFT(*bx,*cx,u,*cx+GOLD*(*cx-*bx))
2146: SHFT(*fb,*fc,fu,(*func)(u))
2147: #ifdef DEBUG
2148: printf("\nmnbrak2 shift GOLD c=%lf",*cx+GOLD*(*cx-*bx));
1.183 brouard 2149: #endif
2150: }
1.162 brouard 2151: } else if ((u-ulim)*(ulim-*cx) >= 0.0) { /* u outside ulim (verifying that ulim is beyond c) */
1.183 brouard 2152: #ifdef DEBUG
1.224 brouard 2153: printf("\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
2154: fprintf(ficlog,"\nmnbrak2 u=%lf outside ulim=%lf (verifying that ulim is beyond c=%lf)\n",u,ulim,*cx);
1.183 brouard 2155: #endif
1.126 brouard 2156: u=ulim;
2157: fu=(*func)(u);
1.183 brouard 2158: } else { /* u could be left to b (if r > q parabola has a maximum) */
2159: #ifdef DEBUG
1.224 brouard 2160: printf("\nmnbrak2 u=%lf could be left to b=%lf (if r=%lf > q=%lf parabola has a maximum)\n",u,*bx,r,q);
2161: 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 2162: #endif
1.126 brouard 2163: u=(*cx)+GOLD*(*cx-*bx);
2164: fu=(*func)(u);
1.224 brouard 2165: #ifdef DEBUG
2166: printf("\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2167: fprintf(ficlog,"\nmnbrak2 new u=%lf fu=%lf shifted gold left from c=%lf and b=%lf \n",u,fu,*cx,*bx);
2168: #endif
1.183 brouard 2169: } /* end tests */
1.126 brouard 2170: SHFT(*ax,*bx,*cx,u)
1.183 brouard 2171: SHFT(*fa,*fb,*fc,fu)
2172: #ifdef DEBUG
1.224 brouard 2173: printf("\nmnbrak2 shift (*ax=%.12f, *fa=%.12lf), (*bx=%.12f, *fb=%.12lf), (*cx=%.12f, *fc=%.12lf)\n",*ax,*fa,*bx,*fb,*cx,*fc);
2174: 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 2175: #endif
2176: } /* 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 2177: }
2178:
2179: /*************** linmin ************************/
1.162 brouard 2180: /* Given an n -dimensional point p[1..n] and an n -dimensional direction xi[1..n] , moves and
2181: resets p to where the function func(p) takes on a minimum along the direction xi from p ,
2182: and replaces xi by the actual vector displacement that p was moved. Also returns as fret
2183: the value of func at the returned location p . This is actually all accomplished by calling the
2184: routines mnbrak and brent .*/
1.126 brouard 2185: int ncom;
2186: double *pcom,*xicom;
2187: double (*nrfunc)(double []);
2188:
1.224 brouard 2189: #ifdef LINMINORIGINAL
1.126 brouard 2190: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []))
1.224 brouard 2191: #else
2192: void linmin(double p[], double xi[], int n, double *fret,double (*func)(double []), int *flat)
2193: #endif
1.126 brouard 2194: {
2195: double brent(double ax, double bx, double cx,
2196: double (*f)(double), double tol, double *xmin);
2197: double f1dim(double x);
2198: void mnbrak(double *ax, double *bx, double *cx, double *fa, double *fb,
2199: double *fc, double (*func)(double));
2200: int j;
2201: double xx,xmin,bx,ax;
2202: double fx,fb,fa;
1.187 brouard 2203:
1.203 brouard 2204: #ifdef LINMINORIGINAL
2205: #else
2206: double scale=10., axs, xxs; /* Scale added for infinity */
2207: #endif
2208:
1.126 brouard 2209: ncom=n;
2210: pcom=vector(1,n);
2211: xicom=vector(1,n);
2212: nrfunc=func;
2213: for (j=1;j<=n;j++) {
2214: pcom[j]=p[j];
1.202 brouard 2215: xicom[j]=xi[j]; /* Former scale xi[j] of currrent direction i */
1.126 brouard 2216: }
1.187 brouard 2217:
1.203 brouard 2218: #ifdef LINMINORIGINAL
2219: xx=1.;
2220: #else
2221: axs=0.0;
2222: xxs=1.;
2223: do{
2224: xx= xxs;
2225: #endif
1.187 brouard 2226: ax=0.;
2227: mnbrak(&ax,&xx,&bx,&fa,&fx,&fb,f1dim); /* Outputs: xtx[j]=pcom[j]+(*xx)*xicom[j]; fx=f(xtx[j]) */
2228: /* brackets with inputs ax=0 and xx=1, but points, pcom=p, and directions values, xicom=xi, are sent via f1dim(x) */
2229: /* 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)) */
2230: /* Outputs: fa=f(p(j)) and fx=f(p(j) + xxs * xi(j) ) and f(bx)= f(p(j)+ bx* xi(j)) */
2231: /* Given input ax=axs and xx=xxs, xx might be too far from ax to get a finite f(xx) */
2232: /* Searches on line, outputs (ax, xx, bx) such that fx < min(fa and fb) */
2233: /* 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 2234: #ifdef LINMINORIGINAL
2235: #else
2236: if (fx != fx){
1.224 brouard 2237: xxs=xxs/scale; /* Trying a smaller xx, closer to initial ax=0 */
2238: printf("|");
2239: fprintf(ficlog,"|");
1.203 brouard 2240: #ifdef DEBUGLINMIN
1.224 brouard 2241: 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 2242: #endif
2243: }
1.224 brouard 2244: }while(fx != fx && xxs > 1.e-5);
1.203 brouard 2245: #endif
2246:
1.191 brouard 2247: #ifdef DEBUGLINMIN
2248: 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 2249: 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 2250: #endif
1.224 brouard 2251: #ifdef LINMINORIGINAL
2252: #else
2253: if(fb == fx){ /* Flat function in the direction */
2254: xmin=xx;
2255: *flat=1;
2256: }else{
2257: *flat=0;
2258: #endif
2259: /*Flat mnbrak2 shift (*ax=0.000000000000, *fa=51626.272983130431), (*bx=-1.618034000000, *fb=51590.149499362531), (*cx=-4.236068025156, *fc=51590.149499362531) */
1.187 brouard 2260: *fret=brent(ax,xx,bx,f1dim,TOL,&xmin); /* Giving a bracketting triplet (ax, xx, bx), find a minimum, xmin, according to f1dim, *fret(xmin),*/
2261: /* fa = f(p[j] + ax * xi[j]), fx = f(p[j] + xx * xi[j]), fb = f(p[j] + bx * xi[j]) */
2262: /* fmin = f(p[j] + xmin * xi[j]) */
2263: /* P+lambda n in that direction (lambdamin), with TOL between abscisses */
2264: /* f1dim(xmin): for (j=1;j<=ncom;j++) xt[j]=pcom[j]+xmin*xicom[j]; */
1.126 brouard 2265: #ifdef DEBUG
1.224 brouard 2266: 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);
2267: 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);
2268: #endif
2269: #ifdef LINMINORIGINAL
2270: #else
2271: }
1.126 brouard 2272: #endif
1.191 brouard 2273: #ifdef DEBUGLINMIN
2274: printf("linmin end ");
1.202 brouard 2275: fprintf(ficlog,"linmin end ");
1.191 brouard 2276: #endif
1.126 brouard 2277: for (j=1;j<=n;j++) {
1.203 brouard 2278: #ifdef LINMINORIGINAL
2279: xi[j] *= xmin;
2280: #else
2281: #ifdef DEBUGLINMIN
2282: if(xxs <1.0)
2283: printf(" before xi[%d]=%12.8f", j,xi[j]);
2284: #endif
2285: 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) */
2286: #ifdef DEBUGLINMIN
2287: if(xxs <1.0)
2288: 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 );
2289: #endif
2290: #endif
1.187 brouard 2291: p[j] += xi[j]; /* Parameters values are updated accordingly */
1.126 brouard 2292: }
1.191 brouard 2293: #ifdef DEBUGLINMIN
1.203 brouard 2294: printf("\n");
1.191 brouard 2295: printf("Comparing last *frec(xmin=%12.8f)=%12.8f from Brent and frec(0.)=%12.8f \n", xmin, *fret, (*func)(p));
1.202 brouard 2296: 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 2297: for (j=1;j<=n;j++) {
1.202 brouard 2298: printf(" xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2299: fprintf(ficlog," xi[%d]= %14.10f p[%d]= %12.7f",j,xi[j],j,p[j]);
2300: if(j % ncovmodel == 0){
1.191 brouard 2301: printf("\n");
1.202 brouard 2302: fprintf(ficlog,"\n");
2303: }
1.191 brouard 2304: }
1.203 brouard 2305: #else
1.191 brouard 2306: #endif
1.126 brouard 2307: free_vector(xicom,1,n);
2308: free_vector(pcom,1,n);
2309: }
2310:
2311:
2312: /*************** powell ************************/
1.162 brouard 2313: /*
2314: Minimization of a function func of n variables. Input consists of an initial starting point
2315: p[1..n] ; an initial matrix xi[1..n][1..n] , whose columns contain the initial set of di-
2316: rections (usually the n unit vectors); and ftol , the fractional tolerance in the function value
2317: such that failure to decrease by more than this amount on one iteration signals doneness. On
2318: output, p is set to the best point found, xi is the then-current direction set, fret is the returned
2319: function value at p , and iter is the number of iterations taken. The routine linmin is used.
2320: */
1.224 brouard 2321: #ifdef LINMINORIGINAL
2322: #else
2323: int *flatdir; /* Function is vanishing in that direction */
1.225 brouard 2324: int flat=0, flatd=0; /* Function is vanishing in that direction */
1.224 brouard 2325: #endif
1.126 brouard 2326: void powell(double p[], double **xi, int n, double ftol, int *iter, double *fret,
2327: double (*func)(double []))
2328: {
1.224 brouard 2329: #ifdef LINMINORIGINAL
2330: void linmin(double p[], double xi[], int n, double *fret,
1.126 brouard 2331: double (*func)(double []));
1.224 brouard 2332: #else
1.241 brouard 2333: void linmin(double p[], double xi[], int n, double *fret,
2334: double (*func)(double []),int *flat);
1.224 brouard 2335: #endif
1.239 brouard 2336: int i,ibig,j,jk,k;
1.126 brouard 2337: double del,t,*pt,*ptt,*xit;
1.181 brouard 2338: double directest;
1.126 brouard 2339: double fp,fptt;
2340: double *xits;
2341: int niterf, itmp;
1.224 brouard 2342: #ifdef LINMINORIGINAL
2343: #else
2344:
2345: flatdir=ivector(1,n);
2346: for (j=1;j<=n;j++) flatdir[j]=0;
2347: #endif
1.126 brouard 2348:
2349: pt=vector(1,n);
2350: ptt=vector(1,n);
2351: xit=vector(1,n);
2352: xits=vector(1,n);
2353: *fret=(*func)(p);
2354: for (j=1;j<=n;j++) pt[j]=p[j];
1.202 brouard 2355: rcurr_time = time(NULL);
1.126 brouard 2356: for (*iter=1;;++(*iter)) {
1.187 brouard 2357: fp=(*fret); /* From former iteration or initial value */
1.126 brouard 2358: ibig=0;
2359: del=0.0;
1.157 brouard 2360: rlast_time=rcurr_time;
2361: /* (void) gettimeofday(&curr_time,&tzp); */
2362: rcurr_time = time(NULL);
2363: curr_time = *localtime(&rcurr_time);
2364: printf("\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret, rcurr_time-rlast_time, rcurr_time-rstart_time);fflush(stdout);
2365: fprintf(ficlog,"\nPowell iter=%d -2*LL=%.12f %ld sec. %ld sec.",*iter,*fret,rcurr_time-rlast_time, rcurr_time-rstart_time); fflush(ficlog);
2366: /* fprintf(ficrespow,"%d %.12f %ld",*iter,*fret,curr_time.tm_sec-start_time.tm_sec); */
1.192 brouard 2367: for (i=1;i<=n;i++) {
1.126 brouard 2368: fprintf(ficrespow," %.12lf", p[i]);
2369: }
1.239 brouard 2370: fprintf(ficrespow,"\n");fflush(ficrespow);
2371: printf("\n#model= 1 + age ");
2372: fprintf(ficlog,"\n#model= 1 + age ");
2373: if(nagesqr==1){
1.241 brouard 2374: printf(" + age*age ");
2375: fprintf(ficlog," + age*age ");
1.239 brouard 2376: }
2377: for(j=1;j <=ncovmodel-2;j++){
2378: if(Typevar[j]==0) {
2379: printf(" + V%d ",Tvar[j]);
2380: fprintf(ficlog," + V%d ",Tvar[j]);
2381: }else if(Typevar[j]==1) {
2382: printf(" + V%d*age ",Tvar[j]);
2383: fprintf(ficlog," + V%d*age ",Tvar[j]);
2384: }else if(Typevar[j]==2) {
2385: printf(" + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2386: fprintf(ficlog," + V%d*V%d ",Tvard[Tposprod[j]][1],Tvard[Tposprod[j]][2]);
2387: }
2388: }
1.126 brouard 2389: printf("\n");
1.239 brouard 2390: /* printf("12 47.0114589 0.0154322 33.2424412 0.3279905 2.3731903 */
2391: /* 13 -21.5392400 0.1118147 1.2680506 1.2973408 -1.0663662 */
1.126 brouard 2392: fprintf(ficlog,"\n");
1.239 brouard 2393: for(i=1,jk=1; i <=nlstate; i++){
2394: for(k=1; k <=(nlstate+ndeath); k++){
2395: if (k != i) {
2396: printf("%d%d ",i,k);
2397: fprintf(ficlog,"%d%d ",i,k);
2398: for(j=1; j <=ncovmodel; j++){
2399: printf("%12.7f ",p[jk]);
2400: fprintf(ficlog,"%12.7f ",p[jk]);
2401: jk++;
2402: }
2403: printf("\n");
2404: fprintf(ficlog,"\n");
2405: }
2406: }
2407: }
1.241 brouard 2408: if(*iter <=3 && *iter >1){
1.157 brouard 2409: tml = *localtime(&rcurr_time);
2410: strcpy(strcurr,asctime(&tml));
2411: rforecast_time=rcurr_time;
1.126 brouard 2412: itmp = strlen(strcurr);
2413: if(strcurr[itmp-1]=='\n') /* Windows outputs with a new line */
1.241 brouard 2414: strcurr[itmp-1]='\0';
1.162 brouard 2415: printf("\nConsidering the time needed for the last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.157 brouard 2416: fprintf(ficlog,"\nConsidering the time needed for this last iteration #%d: %ld seconds,\n",*iter,rcurr_time-rlast_time);
1.126 brouard 2417: for(niterf=10;niterf<=30;niterf+=10){
1.241 brouard 2418: rforecast_time=rcurr_time+(niterf-*iter)*(rcurr_time-rlast_time);
2419: forecast_time = *localtime(&rforecast_time);
2420: strcpy(strfor,asctime(&forecast_time));
2421: itmp = strlen(strfor);
2422: if(strfor[itmp-1]=='\n')
2423: strfor[itmp-1]='\0';
2424: 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);
2425: 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 2426: }
2427: }
1.187 brouard 2428: for (i=1;i<=n;i++) { /* For each direction i */
2429: for (j=1;j<=n;j++) xit[j]=xi[j][i]; /* Directions stored from previous iteration with previous scales */
1.126 brouard 2430: fptt=(*fret);
2431: #ifdef DEBUG
1.203 brouard 2432: printf("fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
2433: fprintf(ficlog, "fret=%lf, %lf, %lf \n", *fret, *fret, *fret);
1.126 brouard 2434: #endif
1.203 brouard 2435: printf("%d",i);fflush(stdout); /* print direction (parameter) i */
1.126 brouard 2436: fprintf(ficlog,"%d",i);fflush(ficlog);
1.224 brouard 2437: #ifdef LINMINORIGINAL
1.188 brouard 2438: linmin(p,xit,n,fret,func); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
1.224 brouard 2439: #else
2440: linmin(p,xit,n,fret,func,&flat); /* Point p[n]. xit[n] has been loaded for direction i as input.*/
2441: flatdir[i]=flat; /* Function is vanishing in that direction i */
2442: #endif
2443: /* Outputs are fret(new point p) p is updated and xit rescaled */
1.188 brouard 2444: if (fabs(fptt-(*fret)) > del) { /* We are keeping the max gain on each of the n directions */
1.224 brouard 2445: /* because that direction will be replaced unless the gain del is small */
2446: /* in comparison with the 'probable' gain, mu^2, with the last average direction. */
2447: /* Unless the n directions are conjugate some gain in the determinant may be obtained */
2448: /* with the new direction. */
2449: del=fabs(fptt-(*fret));
2450: ibig=i;
1.126 brouard 2451: }
2452: #ifdef DEBUG
2453: printf("%d %.12e",i,(*fret));
2454: fprintf(ficlog,"%d %.12e",i,(*fret));
2455: for (j=1;j<=n;j++) {
1.224 brouard 2456: xits[j]=FMAX(fabs(p[j]-pt[j]),1.e-5);
2457: printf(" x(%d)=%.12e",j,xit[j]);
2458: fprintf(ficlog," x(%d)=%.12e",j,xit[j]);
1.126 brouard 2459: }
2460: for(j=1;j<=n;j++) {
1.225 brouard 2461: printf(" p(%d)=%.12e",j,p[j]);
2462: fprintf(ficlog," p(%d)=%.12e",j,p[j]);
1.126 brouard 2463: }
2464: printf("\n");
2465: fprintf(ficlog,"\n");
2466: #endif
1.187 brouard 2467: } /* end loop on each direction i */
2468: /* Convergence test will use last linmin estimation (fret) and compare former iteration (fp) */
1.188 brouard 2469: /* But p and xit have been updated at the end of linmin, *fret corresponds to new p, xit */
1.187 brouard 2470: /* New value of last point Pn is not computed, P(n-1) */
1.224 brouard 2471: for(j=1;j<=n;j++) {
1.302 brouard 2472: if(flatdir[j] >0){
2473: printf(" p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2474: fprintf(ficlog," p(%d)=%lf flat=%d ",j,p[j],flatdir[j]);
2475: }
2476: /* printf("\n"); */
2477: /* fprintf(ficlog,"\n"); */
2478: }
1.243 brouard 2479: /* if (2.0*fabs(fp-(*fret)) <= ftol*(fabs(fp)+fabs(*fret))) { /\* Did we reach enough precision? *\/ */
2480: if (2.0*fabs(fp-(*fret)) <= ftol) { /* Did we reach enough precision? */
1.188 brouard 2481: /* We could compare with a chi^2. chisquare(0.95,ddl=1)=3.84 */
2482: /* By adding age*age in a model, the new -2LL should be lower and the difference follows a */
2483: /* a chisquare statistics with 1 degree. To be significant at the 95% level, it should have */
2484: /* decreased of more than 3.84 */
2485: /* By adding age*age and V1*age the gain (-2LL) should be more than 5.99 (ddl=2) */
2486: /* By using V1+V2+V3, the gain should be 7.82, compared with basic 1+age. */
2487: /* By adding 10 parameters more the gain should be 18.31 */
1.224 brouard 2488:
1.188 brouard 2489: /* Starting the program with initial values given by a former maximization will simply change */
2490: /* the scales of the directions and the directions, because the are reset to canonical directions */
2491: /* Thus the first calls to linmin will give new points and better maximizations until fp-(*fret) is */
2492: /* under the tolerance value. If the tolerance is very small 1.e-9, it could last long. */
1.126 brouard 2493: #ifdef DEBUG
2494: int k[2],l;
2495: k[0]=1;
2496: k[1]=-1;
2497: printf("Max: %.12e",(*func)(p));
2498: fprintf(ficlog,"Max: %.12e",(*func)(p));
2499: for (j=1;j<=n;j++) {
2500: printf(" %.12e",p[j]);
2501: fprintf(ficlog," %.12e",p[j]);
2502: }
2503: printf("\n");
2504: fprintf(ficlog,"\n");
2505: for(l=0;l<=1;l++) {
2506: for (j=1;j<=n;j++) {
2507: ptt[j]=p[j]+(p[j]-pt[j])*k[l];
2508: printf("l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2509: fprintf(ficlog,"l=%d j=%d ptt=%.12e, xits=%.12e, p=%.12e, xit=%.12e", l,j,ptt[j],xits[j],p[j],xit[j]);
2510: }
2511: printf("func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2512: fprintf(ficlog,"func(ptt)=%.12e, deriv=%.12e\n",(*func)(ptt),(ptt[j]-p[j])/((*func)(ptt)-(*func)(p)));
2513: }
2514: #endif
2515:
1.224 brouard 2516: #ifdef LINMINORIGINAL
2517: #else
2518: free_ivector(flatdir,1,n);
2519: #endif
1.126 brouard 2520: free_vector(xit,1,n);
2521: free_vector(xits,1,n);
2522: free_vector(ptt,1,n);
2523: free_vector(pt,1,n);
2524: return;
1.192 brouard 2525: } /* enough precision */
1.240 brouard 2526: if (*iter == ITMAX*n) nrerror("powell exceeding maximum iterations.");
1.181 brouard 2527: for (j=1;j<=n;j++) { /* Computes the extrapolated point P_0 + 2 (P_n-P_0) */
1.126 brouard 2528: ptt[j]=2.0*p[j]-pt[j];
2529: xit[j]=p[j]-pt[j];
2530: pt[j]=p[j];
2531: }
1.181 brouard 2532: fptt=(*func)(ptt); /* f_3 */
1.224 brouard 2533: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
2534: if (*iter <=4) {
1.225 brouard 2535: #else
2536: #endif
1.224 brouard 2537: #ifdef POWELLNOF3INFF1TEST /* skips test F3 <F1 */
1.192 brouard 2538: #else
1.161 brouard 2539: if (fptt < fp) { /* If extrapolated point is better, decide if we keep that new direction or not */
1.192 brouard 2540: #endif
1.162 brouard 2541: /* (x1 f1=fp), (x2 f2=*fret), (x3 f3=fptt), (xm fm) */
1.161 brouard 2542: /* From x1 (P0) distance of x2 is at h and x3 is 2h */
1.162 brouard 2543: /* Let f"(x2) be the 2nd derivative equal everywhere. */
2544: /* Then the parabolic through (x1,f1), (x2,f2) and (x3,f3) */
2545: /* will reach at f3 = fm + h^2/2 f"m ; f" = (f1 -2f2 +f3 ) / h**2 */
1.224 brouard 2546: /* Conditional for using this new direction is that mu^2 = (f1-2f2+f3)^2 /2 < del or directest <0 */
2547: /* also lamda^2=(f1-f2)^2/mu² is a parasite solution of powell */
2548: /* For powell, inclusion of this average direction is only if t(del)<0 or del inbetween mu^2 and lambda^2 */
1.161 brouard 2549: /* t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)-del*SQR(fp-fptt); */
1.224 brouard 2550: /* Even if f3 <f1, directest can be negative and t >0 */
2551: /* mu² and del² are equal when f3=f1 */
2552: /* f3 < f1 : mu² < del <= lambda^2 both test are equivalent */
2553: /* f3 < f1 : mu² < lambda^2 < del then directtest is negative and powell t is positive */
2554: /* f3 > f1 : lambda² < mu^2 < del then t is negative and directest >0 */
2555: /* f3 > f1 : lambda² < del < mu^2 then t is positive and directest >0 */
1.183 brouard 2556: #ifdef NRCORIGINAL
2557: t=2.0*(fp-2.0*(*fret)+fptt)*SQR(fp-(*fret)-del)- del*SQR(fp-fptt); /* Original Numerical Recipes in C*/
2558: #else
2559: 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 2560: t= t- del*SQR(fp-fptt);
1.183 brouard 2561: #endif
1.202 brouard 2562: directest = fp-2.0*(*fret)+fptt - 2.0 * del; /* If delta was big enough we change it for a new direction */
1.161 brouard 2563: #ifdef DEBUG
1.181 brouard 2564: 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);
2565: 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 2566: printf("t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2567: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2568: fprintf(ficlog,"t3= %.12lf, t4= %.12lf, t3*= %.12lf, t4*= %.12lf\n",SQR(fp-(*fret)-del),SQR(fp-fptt),
2569: (fp-(*fret)-del)*(fp-(*fret)-del),(fp-fptt)*(fp-fptt));
2570: 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);
2571: 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);
2572: #endif
1.183 brouard 2573: #ifdef POWELLORIGINAL
2574: if (t < 0.0) { /* Then we use it for new direction */
2575: #else
1.182 brouard 2576: if (directest*t < 0.0) { /* Contradiction between both tests */
1.224 brouard 2577: 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 2578: 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 2579: 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 2580: fprintf(ficlog,"f1-2f2+f3= %.12lf, f1-f2-del= %.12lf, f1-f3= %.12lf\n",fp-2.0*(*fret)+fptt, fp -(*fret) -del, fp-fptt);
2581: }
1.181 brouard 2582: if (directest < 0.0) { /* Then we use it for new direction */
2583: #endif
1.191 brouard 2584: #ifdef DEBUGLINMIN
1.234 brouard 2585: printf("Before linmin in direction P%d-P0\n",n);
2586: for (j=1;j<=n;j++) {
2587: printf(" Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2588: fprintf(ficlog," Before xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2589: if(j % ncovmodel == 0){
2590: printf("\n");
2591: fprintf(ficlog,"\n");
2592: }
2593: }
1.224 brouard 2594: #endif
2595: #ifdef LINMINORIGINAL
1.234 brouard 2596: linmin(p,xit,n,fret,func); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
1.224 brouard 2597: #else
1.234 brouard 2598: linmin(p,xit,n,fret,func,&flat); /* computes minimum on the extrapolated direction: changes p and rescales xit.*/
2599: flatdir[i]=flat; /* Function is vanishing in that direction i */
1.191 brouard 2600: #endif
1.234 brouard 2601:
1.191 brouard 2602: #ifdef DEBUGLINMIN
1.234 brouard 2603: for (j=1;j<=n;j++) {
2604: printf("After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2605: fprintf(ficlog,"After xit[%d]= %12.7f p[%d]= %12.7f",j,xit[j],j,p[j]);
2606: if(j % ncovmodel == 0){
2607: printf("\n");
2608: fprintf(ficlog,"\n");
2609: }
2610: }
1.224 brouard 2611: #endif
1.234 brouard 2612: for (j=1;j<=n;j++) {
2613: xi[j][ibig]=xi[j][n]; /* Replace direction with biggest decrease by last direction n */
2614: xi[j][n]=xit[j]; /* and this nth direction by the by the average p_0 p_n */
2615: }
1.224 brouard 2616: #ifdef LINMINORIGINAL
2617: #else
1.234 brouard 2618: for (j=1, flatd=0;j<=n;j++) {
2619: if(flatdir[j]>0)
2620: flatd++;
2621: }
2622: if(flatd >0){
1.255 brouard 2623: printf("%d flat directions: ",flatd);
2624: fprintf(ficlog,"%d flat directions :",flatd);
1.234 brouard 2625: for (j=1;j<=n;j++) {
2626: if(flatdir[j]>0){
2627: printf("%d ",j);
2628: fprintf(ficlog,"%d ",j);
2629: }
2630: }
2631: printf("\n");
2632: fprintf(ficlog,"\n");
2633: }
1.191 brouard 2634: #endif
1.234 brouard 2635: printf("Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2636: fprintf(ficlog,"Gaining to use new average direction of P0 P%d instead of biggest increase direction %d :\n",n,ibig);
2637:
1.126 brouard 2638: #ifdef DEBUG
1.234 brouard 2639: printf("Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2640: fprintf(ficlog,"Direction changed last moved %d in place of ibig=%d, new last is the average:\n",n,ibig);
2641: for(j=1;j<=n;j++){
2642: printf(" %lf",xit[j]);
2643: fprintf(ficlog," %lf",xit[j]);
2644: }
2645: printf("\n");
2646: fprintf(ficlog,"\n");
1.126 brouard 2647: #endif
1.192 brouard 2648: } /* end of t or directest negative */
1.224 brouard 2649: #ifdef POWELLNOF3INFF1TEST
1.192 brouard 2650: #else
1.234 brouard 2651: } /* end if (fptt < fp) */
1.192 brouard 2652: #endif
1.225 brouard 2653: #ifdef NODIRECTIONCHANGEDUNTILNITER /* No change in drections until some iterations are done */
1.234 brouard 2654: } /*NODIRECTIONCHANGEDUNTILNITER No change in drections until some iterations are done */
1.225 brouard 2655: #else
1.224 brouard 2656: #endif
1.234 brouard 2657: } /* loop iteration */
1.126 brouard 2658: }
1.234 brouard 2659:
1.126 brouard 2660: /**** Prevalence limit (stable or period prevalence) ****************/
1.234 brouard 2661:
1.235 brouard 2662: 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 2663: {
1.279 brouard 2664: /**< Computes the prevalence limit in each live state at age x and for covariate combination ij
2665: * (and selected quantitative values in nres)
2666: * by left multiplying the unit
2667: * matrix by transitions matrix until convergence is reached with precision ftolpl
2668: * Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I
2669: * Wx is row vector: population in state 1, population in state 2, population dead
2670: * or prevalence in state 1, prevalence in state 2, 0
2671: * newm is the matrix after multiplications, its rows are identical at a factor.
2672: * Inputs are the parameter, age, a tolerance for the prevalence limit ftolpl.
2673: * Output is prlim.
2674: * Initial matrix pimij
2675: */
1.206 brouard 2676: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2677: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2678: /* 0, 0 , 1} */
2679: /*
2680: * and after some iteration: */
2681: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2682: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2683: /* 0, 0 , 1} */
2684: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2685: /* {0.51571254859325999, 0.4842874514067399, */
2686: /* 0.51326036147820708, 0.48673963852179264} */
2687: /* If we start from prlim again, prlim tends to a constant matrix */
1.234 brouard 2688:
1.126 brouard 2689: int i, ii,j,k;
1.209 brouard 2690: double *min, *max, *meandiff, maxmax,sumnew=0.;
1.145 brouard 2691: /* double **matprod2(); */ /* test */
1.218 brouard 2692: double **out, cov[NCOVMAX+1], **pmij(); /* **pmmij is a global variable feeded with oldms etc */
1.126 brouard 2693: double **newm;
1.209 brouard 2694: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
1.203 brouard 2695: int ncvloop=0;
1.288 brouard 2696: int first=0;
1.169 brouard 2697:
1.209 brouard 2698: min=vector(1,nlstate);
2699: max=vector(1,nlstate);
2700: meandiff=vector(1,nlstate);
2701:
1.218 brouard 2702: /* Starting with matrix unity */
1.126 brouard 2703: for (ii=1;ii<=nlstate+ndeath;ii++)
2704: for (j=1;j<=nlstate+ndeath;j++){
2705: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2706: }
1.169 brouard 2707:
2708: cov[1]=1.;
2709:
2710: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
1.202 brouard 2711: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.126 brouard 2712: for(agefin=age-stepm/YEARM; agefin>=age-delaymax; agefin=agefin-stepm/YEARM){
1.202 brouard 2713: ncvloop++;
1.126 brouard 2714: newm=savm;
2715: /* Covariates have to be included here again */
1.138 brouard 2716: cov[2]=agefin;
1.187 brouard 2717: if(nagesqr==1)
2718: cov[3]= agefin*agefin;;
1.234 brouard 2719: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2720: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2721: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.235 brouard 2722: /* printf("prevalim Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,Tvar[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); */
1.234 brouard 2723: }
2724: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2725: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
1.235 brouard 2726: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2727: /* printf("prevalim Quantitative k=%d TvarsQind[%d]=%d, TvarsQ[%d]=V%d,Tqresult[%d][%d]=%f\n",k,k,TvarsQind[k],k,TvarsQ[k],nres,k,Tqresult[nres][k]); */
1.138 brouard 2728: }
1.237 brouard 2729: for (k=1; k<=cptcovage;k++){ /* For product with age */
1.234 brouard 2730: if(Dummy[Tvar[Tage[k]]]){
2731: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2732: } else{
1.235 brouard 2733: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
1.234 brouard 2734: }
1.235 brouard 2735: /* printf("prevalim Age combi=%d k=%d Tage[%d]=V%d Tqresult[%d][%d]=%f\n",ij,k,k,Tage[k],nres,k,Tqresult[nres][k]); */
1.234 brouard 2736: }
1.237 brouard 2737: for (k=1; k<=cptcovprod;k++){ /* For product without age */
1.235 brouard 2738: /* printf("prevalim Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2]); */
1.237 brouard 2739: if(Dummy[Tvard[k][1]==0]){
2740: if(Dummy[Tvard[k][2]==0]){
2741: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2742: }else{
2743: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2744: }
2745: }else{
2746: if(Dummy[Tvard[k][2]==0]){
2747: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2748: }else{
2749: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2750: }
2751: }
1.234 brouard 2752: }
1.138 brouard 2753: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2754: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2755: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
1.145 brouard 2756: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2757: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2758: /* age and covariate values of ij are in 'cov' */
1.142 brouard 2759: out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /* Bug Valgrind */
1.138 brouard 2760:
1.126 brouard 2761: savm=oldm;
2762: oldm=newm;
1.209 brouard 2763:
2764: for(j=1; j<=nlstate; j++){
2765: max[j]=0.;
2766: min[j]=1.;
2767: }
2768: for(i=1;i<=nlstate;i++){
2769: sumnew=0;
2770: for(k=1; k<=ndeath; k++) sumnew+=newm[i][nlstate+k];
2771: for(j=1; j<=nlstate; j++){
2772: prlim[i][j]= newm[i][j]/(1-sumnew);
2773: max[j]=FMAX(max[j],prlim[i][j]);
2774: min[j]=FMIN(min[j],prlim[i][j]);
2775: }
2776: }
2777:
1.126 brouard 2778: maxmax=0.;
1.209 brouard 2779: for(j=1; j<=nlstate; j++){
2780: meandiff[j]=(max[j]-min[j])/(max[j]+min[j])*2.; /* mean difference for each column */
2781: maxmax=FMAX(maxmax,meandiff[j]);
2782: /* 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 2783: } /* j loop */
1.203 brouard 2784: *ncvyear= (int)age- (int)agefin;
1.208 brouard 2785: /* 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 2786: if(maxmax < ftolpl){
1.209 brouard 2787: /* printf("maxmax=%lf ncvloop=%ld, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
2788: free_vector(min,1,nlstate);
2789: free_vector(max,1,nlstate);
2790: free_vector(meandiff,1,nlstate);
1.126 brouard 2791: return prlim;
2792: }
1.288 brouard 2793: } /* agefin loop */
1.208 brouard 2794: /* After some age loop it doesn't converge */
1.288 brouard 2795: if(!first){
2796: first=1;
2797: 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);
2798: }
2799: 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);
2800:
1.209 brouard 2801: /* 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); */
2802: free_vector(min,1,nlstate);
2803: free_vector(max,1,nlstate);
2804: free_vector(meandiff,1,nlstate);
1.208 brouard 2805:
1.169 brouard 2806: return prlim; /* should not reach here */
1.126 brouard 2807: }
2808:
1.217 brouard 2809:
2810: /**** Back Prevalence limit (stable or period prevalence) ****************/
2811:
1.218 brouard 2812: /* 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) */
2813: /* 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 2814: double **bprevalim(double **bprlim, double ***prevacurrent, int nlstate, double x[], double age, double ftolpl, int *ncvyear, int ij, int nres)
1.217 brouard 2815: {
1.264 brouard 2816: /* 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 2817: matrix by transitions matrix until convergence is reached with precision ftolpl */
2818: /* Wx= Wx-1 Px-1= Wx-2 Px-2 Px-1 = Wx-n Px-n ... Px-2 Px-1 I */
2819: /* Wx is row vector: population in state 1, population in state 2, population dead */
2820: /* or prevalence in state 1, prevalence in state 2, 0 */
2821: /* newm is the matrix after multiplications, its rows are identical at a factor */
2822: /* Initial matrix pimij */
2823: /* {0.85204250825084937, 0.13044499163996345, 0.017512500109187184, */
2824: /* 0.090851990222114765, 0.88271245433047185, 0.026435555447413338, */
2825: /* 0, 0 , 1} */
2826: /*
2827: * and after some iteration: */
2828: /* {0.45504275246439968, 0.42731458730878791, 0.11764266022681241, */
2829: /* 0.45201005341706885, 0.42865420071559901, 0.11933574586733192, */
2830: /* 0, 0 , 1} */
2831: /* And prevalence by suppressing the deaths are close to identical rows in prlim: */
2832: /* {0.51571254859325999, 0.4842874514067399, */
2833: /* 0.51326036147820708, 0.48673963852179264} */
2834: /* If we start from prlim again, prlim tends to a constant matrix */
2835:
2836: int i, ii,j,k;
1.247 brouard 2837: int first=0;
1.217 brouard 2838: double *min, *max, *meandiff, maxmax,sumnew=0.;
2839: /* double **matprod2(); */ /* test */
2840: double **out, cov[NCOVMAX+1], **bmij();
2841: double **newm;
1.218 brouard 2842: double **dnewm, **doldm, **dsavm; /* for use */
2843: double **oldm, **savm; /* for use */
2844:
1.217 brouard 2845: double agefin, delaymax=200. ; /* 100 Max number of years to converge */
2846: int ncvloop=0;
2847:
2848: min=vector(1,nlstate);
2849: max=vector(1,nlstate);
2850: meandiff=vector(1,nlstate);
2851:
1.266 brouard 2852: dnewm=ddnewms; doldm=ddoldms; dsavm=ddsavms;
2853: oldm=oldms; savm=savms;
2854:
2855: /* Starting with matrix unity */
2856: for (ii=1;ii<=nlstate+ndeath;ii++)
2857: for (j=1;j<=nlstate+ndeath;j++){
1.217 brouard 2858: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
2859: }
2860:
2861: cov[1]=1.;
2862:
2863: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
2864: /* Start at agefin= age, computes the matrix of passage and loops decreasing agefin until convergence is reached */
1.218 brouard 2865: /* for(agefin=age+stepm/YEARM; agefin<=age+delaymax; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
1.288 brouard 2866: /* for(agefin=age; agefin<AGESUP; agefin=agefin+stepm/YEARM){ /\* A changer en age *\/ */
2867: for(agefin=age; agefin<FMIN(AGESUP,age+delaymax); agefin=agefin+stepm/YEARM){ /* A changer en age */
1.217 brouard 2868: ncvloop++;
1.218 brouard 2869: newm=savm; /* oldm should be kept from previous iteration or unity at start */
2870: /* newm points to the allocated table savm passed by the function it can be written, savm could be reallocated */
1.217 brouard 2871: /* Covariates have to be included here again */
2872: cov[2]=agefin;
2873: if(nagesqr==1)
2874: cov[3]= agefin*agefin;;
1.242 brouard 2875: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
2876: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
2877: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
1.264 brouard 2878: /* printf("bprevalim Dummy agefin=%.0f combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov[%d]=%lf codtabm(%d,Tvar[%d])=%d \n",agefin,ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],2+nagesqr+TvarsDind[k],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); */
1.242 brouard 2879: }
2880: /* for (k=1; k<=cptcovn;k++) { */
2881: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
2882: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
2883: /* /\* 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])]); *\/ */
2884: /* } */
2885: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
2886: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
2887: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
2888: /* 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]); */
2889: }
2890: /* for (k=1; k<=cptcovage;k++) cov[2+nagesqr+Tage[k]]=nbcode[Tvar[k]][codtabm(ij,k)]*cov[2]; */
2891: /* for (k=1; k<=cptcovprod;k++) /\* Useless *\/ */
2892: /* /\* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,Tvard[k][1])] * nbcode[Tvard[k][2]][codtabm(ij,Tvard[k][2])]; *\/ */
2893: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)]; */
2894: for (k=1; k<=cptcovage;k++){ /* For product with age */
2895: if(Dummy[Tvar[Tage[k]]]){
2896: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
2897: } else{
2898: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
2899: }
2900: /* 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]); */
2901: }
2902: for (k=1; k<=cptcovprod;k++){ /* For product without age */
2903: /* 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]); */
2904: if(Dummy[Tvard[k][1]==0]){
2905: if(Dummy[Tvard[k][2]==0]){
2906: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
2907: }else{
2908: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * Tqresult[nres][k];
2909: }
2910: }else{
2911: if(Dummy[Tvard[k][2]==0]){
2912: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][2]][codtabm(ij,k)] * Tqinvresult[nres][Tvard[k][1]];
2913: }else{
2914: cov[2+nagesqr+Tprod[k]]=Tqinvresult[nres][Tvard[k][1]]* Tqinvresult[nres][Tvard[k][2]];
2915: }
2916: }
1.217 brouard 2917: }
2918:
2919: /*printf("ij=%d cptcovprod=%d tvar=%d ", ij, cptcovprod, Tvar[1]);*/
2920: /*printf("ij=%d cov[3]=%lf cov[4]=%lf \n",ij, cov[3],cov[4]);*/
2921: /*printf("ij=%d cov[3]=%lf \n",ij, cov[3]);*/
2922: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
2923: /* out=matprod2(newm, pmij(pmmij,cov,ncovmodel,x,nlstate),1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); /\* Bug Valgrind *\/ */
1.218 brouard 2924: /* ij should be linked to the correct index of cov */
2925: /* age and covariate values ij are in 'cov', but we need to pass
2926: * ij for the observed prevalence at age and status and covariate
2927: * number: prevacurrent[(int)agefin][ii][ij]
2928: */
2929: /* 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 *\/ */
2930: /* 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 *\/ */
2931: 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 2932: /* if((int)age == 86 || (int)age == 87){ */
1.266 brouard 2933: /* printf(" Backward prevalim age=%d agefin=%d \n", (int) age, (int) agefin); */
2934: /* for(i=1; i<=nlstate+ndeath; i++) { */
2935: /* printf("%d newm= ",i); */
2936: /* for(j=1;j<=nlstate+ndeath;j++) { */
2937: /* printf("%f ",newm[i][j]); */
2938: /* } */
2939: /* printf("oldm * "); */
2940: /* for(j=1;j<=nlstate+ndeath;j++) { */
2941: /* printf("%f ",oldm[i][j]); */
2942: /* } */
1.268 brouard 2943: /* printf(" bmmij "); */
1.266 brouard 2944: /* for(j=1;j<=nlstate+ndeath;j++) { */
2945: /* printf("%f ",pmmij[i][j]); */
2946: /* } */
2947: /* printf("\n"); */
2948: /* } */
2949: /* } */
1.217 brouard 2950: savm=oldm;
2951: oldm=newm;
1.266 brouard 2952:
1.217 brouard 2953: for(j=1; j<=nlstate; j++){
2954: max[j]=0.;
2955: min[j]=1.;
2956: }
2957: for(j=1; j<=nlstate; j++){
2958: for(i=1;i<=nlstate;i++){
1.234 brouard 2959: /* bprlim[i][j]= newm[i][j]/(1-sumnew); */
2960: bprlim[i][j]= newm[i][j];
2961: max[i]=FMAX(max[i],bprlim[i][j]); /* Max in line */
2962: min[i]=FMIN(min[i],bprlim[i][j]);
1.217 brouard 2963: }
2964: }
1.218 brouard 2965:
1.217 brouard 2966: maxmax=0.;
2967: for(i=1; i<=nlstate; i++){
2968: meandiff[i]=(max[i]-min[i])/(max[i]+min[i])*2.; /* mean difference for each column */
2969: maxmax=FMAX(maxmax,meandiff[i]);
2970: /* 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 2971: } /* i loop */
1.217 brouard 2972: *ncvyear= -( (int)age- (int)agefin);
1.268 brouard 2973: /* printf("Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2974: if(maxmax < ftolpl){
1.220 brouard 2975: /* printf("OK Back maxmax=%lf ncvloop=%d, age=%d, agefin=%d ncvyear=%d \n", maxmax, ncvloop, (int)age, (int)agefin, *ncvyear); */
1.217 brouard 2976: free_vector(min,1,nlstate);
2977: free_vector(max,1,nlstate);
2978: free_vector(meandiff,1,nlstate);
2979: return bprlim;
2980: }
1.288 brouard 2981: } /* agefin loop */
1.217 brouard 2982: /* After some age loop it doesn't converge */
1.288 brouard 2983: if(!first){
1.247 brouard 2984: first=1;
2985: 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\
2986: 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);
2987: }
2988: 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 2989: 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);
2990: /* 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); */
2991: free_vector(min,1,nlstate);
2992: free_vector(max,1,nlstate);
2993: free_vector(meandiff,1,nlstate);
2994:
2995: return bprlim; /* should not reach here */
2996: }
2997:
1.126 brouard 2998: /*************** transition probabilities ***************/
2999:
3000: double **pmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
3001: {
1.138 brouard 3002: /* According to parameters values stored in x and the covariate's values stored in cov,
1.266 brouard 3003: computes the probability to be observed in state j (after stepm years) being in state i by appying the
1.138 brouard 3004: model to the ncovmodel covariates (including constant and age).
3005: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3006: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3007: ncth covariate in the global vector x is given by the formula:
3008: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3009: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3010: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3011: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
1.266 brouard 3012: Outputs ps[i][j] or probability to be observed in j being in i according to
1.138 brouard 3013: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
1.266 brouard 3014: Sum on j ps[i][j] should equal to 1.
1.138 brouard 3015: */
3016: double s1, lnpijopii;
1.126 brouard 3017: /*double t34;*/
1.164 brouard 3018: int i,j, nc, ii, jj;
1.126 brouard 3019:
1.223 brouard 3020: for(i=1; i<= nlstate; i++){
3021: for(j=1; j<i;j++){
3022: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3023: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3024: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3025: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3026: }
3027: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3028: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3029: }
3030: for(j=i+1; j<=nlstate+ndeath;j++){
3031: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3032: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3033: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3034: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3035: }
3036: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3037: }
3038: }
1.218 brouard 3039:
1.223 brouard 3040: for(i=1; i<= nlstate; i++){
3041: s1=0;
3042: for(j=1; j<i; j++){
3043: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3044: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3045: }
3046: for(j=i+1; j<=nlstate+ndeath; j++){
3047: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3048: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3049: }
3050: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3051: ps[i][i]=1./(s1+1.);
3052: /* Computing other pijs */
3053: for(j=1; j<i; j++)
3054: ps[i][j]= exp(ps[i][j])*ps[i][i];
3055: for(j=i+1; j<=nlstate+ndeath; j++)
3056: ps[i][j]= exp(ps[i][j])*ps[i][i];
3057: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3058: } /* end i */
1.218 brouard 3059:
1.223 brouard 3060: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3061: for(jj=1; jj<= nlstate+ndeath; jj++){
3062: ps[ii][jj]=0;
3063: ps[ii][ii]=1;
3064: }
3065: }
1.294 brouard 3066:
3067:
1.223 brouard 3068: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3069: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3070: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3071: /* } */
3072: /* printf("\n "); */
3073: /* } */
3074: /* printf("\n ");printf("%lf ",cov[2]);*/
3075: /*
3076: for(i=1; i<= npar; i++) printf("%f ",x[i]);
1.218 brouard 3077: goto end;*/
1.266 brouard 3078: return ps; /* Pointer is unchanged since its call */
1.126 brouard 3079: }
3080:
1.218 brouard 3081: /*************** backward transition probabilities ***************/
3082:
3083: /* 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 ) */
3084: /* double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, double ***dnewm, double **doldm, double **dsavm, int ij ) */
3085: double **bmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate, double ***prevacurrent, int ij )
3086: {
1.302 brouard 3087: /* 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 3088: * 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 3089: */
1.218 brouard 3090: int i, ii, j,k;
1.222 brouard 3091:
3092: double **out, **pmij();
3093: double sumnew=0.;
1.218 brouard 3094: double agefin;
1.292 brouard 3095: 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 3096: double **dnewm, **dsavm, **doldm;
3097: double **bbmij;
3098:
1.218 brouard 3099: doldm=ddoldms; /* global pointers */
1.222 brouard 3100: dnewm=ddnewms;
3101: dsavm=ddsavms;
3102:
3103: agefin=cov[2];
1.268 brouard 3104: /* Bx = Diag(w_x) P_x Diag(Sum_i w^i_x p^ij_x */
1.222 brouard 3105: /* bmij *//* age is cov[2], ij is included in cov, but we need for
1.266 brouard 3106: the observed prevalence (with this covariate ij) at beginning of transition */
3107: /* dsavm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
1.268 brouard 3108:
3109: /* P_x */
1.266 brouard 3110: pmmij=pmij(pmmij,cov,ncovmodel,x,nlstate); /*This is forward probability from agefin to agefin + stepm */
1.268 brouard 3111: /* outputs pmmij which is a stochastic matrix in row */
3112:
3113: /* Diag(w_x) */
1.292 brouard 3114: /* Rescaling the cross-sectional prevalence: Problem with prevacurrent which can be zero */
1.268 brouard 3115: sumnew=0.;
1.269 brouard 3116: /*for (ii=1;ii<=nlstate+ndeath;ii++){*/
1.268 brouard 3117: for (ii=1;ii<=nlstate;ii++){ /* Only on live states */
1.297 brouard 3118: /* printf(" agefin=%d, ii=%d, ij=%d, prev=%f\n",(int)agefin,ii, ij, prevacurrent[(int)agefin][ii][ij]); */
1.268 brouard 3119: sumnew+=prevacurrent[(int)agefin][ii][ij];
3120: }
3121: if(sumnew >0.01){ /* At least some value in the prevalence */
3122: for (ii=1;ii<=nlstate+ndeath;ii++){
3123: for (j=1;j<=nlstate+ndeath;j++)
1.269 brouard 3124: doldm[ii][j]=(ii==j ? prevacurrent[(int)agefin][ii][ij]/sumnew : 0.0);
1.268 brouard 3125: }
3126: }else{
3127: for (ii=1;ii<=nlstate+ndeath;ii++){
3128: for (j=1;j<=nlstate+ndeath;j++)
3129: doldm[ii][j]=(ii==j ? 1./nlstate : 0.0);
3130: }
3131: /* if(sumnew <0.9){ */
3132: /* printf("Problem internal bmij B: sum on i wi <0.9: j=%d, sum_i wi=%lf,agefin=%d\n",j,sumnew, (int)agefin); */
3133: /* } */
3134: }
3135: k3=0.0; /* We put the last diagonal to 0 */
3136: for (ii=nlstate+1;ii<=nlstate+ndeath;ii++){
3137: doldm[ii][ii]= k3;
3138: }
3139: /* End doldm, At the end doldm is diag[(w_i)] */
3140:
1.292 brouard 3141: /* Left product of this diag matrix by pmmij=Px (dnewm=dsavm*doldm): diag[(w_i)*Px */
3142: bbmij=matprod2(dnewm, doldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, pmmij); /* was a Bug Valgrind */
1.268 brouard 3143:
1.292 brouard 3144: /* Diag(Sum_i w^i_x p^ij_x, should be the prevalence at age x+stepm */
1.268 brouard 3145: /* 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 3146: for (j=1;j<=nlstate+ndeath;j++){
1.268 brouard 3147: sumnew=0.;
1.222 brouard 3148: for (ii=1;ii<=nlstate;ii++){
1.266 brouard 3149: /* sumnew+=dsavm[ii][j]*prevacurrent[(int)agefin][ii][ij]; */
1.268 brouard 3150: sumnew+=pmmij[ii][j]*doldm[ii][ii]; /* Yes prevalence at beginning of transition */
1.222 brouard 3151: } /* sumnew is (N11+N21)/N..= N.1/N.. = sum on i of w_i pij */
1.268 brouard 3152: for (ii=1;ii<=nlstate+ndeath;ii++){
1.222 brouard 3153: /* if(agefin >= agemaxpar && agefin <= agemaxpar+stepm/YEARM){ */
1.268 brouard 3154: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3155: /* }else if(agefin >= agemaxpar+stepm/YEARM){ */
1.268 brouard 3156: /* dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0); */
1.222 brouard 3157: /* }else */
1.268 brouard 3158: dsavm[ii][j]=(ii==j ? 1./sumnew : 0.0);
3159: } /*End ii */
3160: } /* 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 */
3161:
1.292 brouard 3162: ps=matprod2(ps, dnewm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, dsavm); /* was a Bug Valgrind */
1.268 brouard 3163: /* ps is now diag[w_i] * Px * diag [1/(w_1p1i+w_2 p2i)] */
1.222 brouard 3164: /* end bmij */
1.266 brouard 3165: return ps; /*pointer is unchanged */
1.218 brouard 3166: }
1.217 brouard 3167: /*************** transition probabilities ***************/
3168:
1.218 brouard 3169: double **bpmij(double **ps, double *cov, int ncovmodel, double *x, int nlstate )
1.217 brouard 3170: {
3171: /* According to parameters values stored in x and the covariate's values stored in cov,
3172: computes the probability to be observed in state j being in state i by appying the
3173: model to the ncovmodel covariates (including constant and age).
3174: lnpijopii=ln(pij/pii)= aij+bij*age+cij*v1+dij*v2+... = sum_nc=1^ncovmodel xij(nc)*cov[nc]
3175: and, according on how parameters are entered, the position of the coefficient xij(nc) of the
3176: ncth covariate in the global vector x is given by the formula:
3177: j<i nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel
3178: j>=i nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel
3179: Computes ln(pij/pii) (lnpijopii), deduces pij/pii by exponentiation,
3180: sums on j different of i to get 1-pii/pii, deduces pii, and then all pij.
3181: Outputs ps[i][j] the probability to be observed in j being in j according to
3182: the values of the covariates cov[nc] and corresponding parameter values x[nc+shiftij]
3183: */
3184: double s1, lnpijopii;
3185: /*double t34;*/
3186: int i,j, nc, ii, jj;
3187:
1.234 brouard 3188: for(i=1; i<= nlstate; i++){
3189: for(j=1; j<i;j++){
3190: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3191: /*lnpijopii += param[i][j][nc]*cov[nc];*/
3192: lnpijopii += x[nc+((i-1)*(nlstate+ndeath-1)+j-1)*ncovmodel]*cov[nc];
3193: /* printf("Int j<i s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3194: }
3195: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3196: /* printf("s1=%.17e, lnpijopii=%.17e\n",s1,lnpijopii); */
3197: }
3198: for(j=i+1; j<=nlstate+ndeath;j++){
3199: for (nc=1, lnpijopii=0.;nc <=ncovmodel; nc++){
3200: /*lnpijopii += x[(i-1)*nlstate*ncovmodel+(j-2)*ncovmodel+nc+(i-1)*(ndeath-1)*ncovmodel]*cov[nc];*/
3201: lnpijopii += x[nc + ((i-1)*(nlstate+ndeath-1)+(j-2))*ncovmodel]*cov[nc];
3202: /* printf("Int j>i s1=%.17e, lnpijopii=%.17e %lx %lx\n",s1,lnpijopii,s1,lnpijopii); */
3203: }
3204: ps[i][j]=lnpijopii; /* In fact ln(pij/pii) */
3205: }
3206: }
3207:
3208: for(i=1; i<= nlstate; i++){
3209: s1=0;
3210: for(j=1; j<i; j++){
3211: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3212: /*printf("debug1 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3213: }
3214: for(j=i+1; j<=nlstate+ndeath; j++){
3215: s1+=exp(ps[i][j]); /* In fact sums pij/pii */
3216: /*printf("debug2 %d %d ps=%lf exp(ps)=%lf s1+=%lf\n",i,j,ps[i][j],exp(ps[i][j]),s1); */
3217: }
3218: /* s1= sum_{j<>i} pij/pii=(1-pii)/pii and thus pii is known from s1 */
3219: ps[i][i]=1./(s1+1.);
3220: /* Computing other pijs */
3221: for(j=1; j<i; j++)
3222: ps[i][j]= exp(ps[i][j])*ps[i][i];
3223: for(j=i+1; j<=nlstate+ndeath; j++)
3224: ps[i][j]= exp(ps[i][j])*ps[i][i];
3225: /* ps[i][nlstate+1]=1.-s1- ps[i][i];*/ /* Sum should be 1 */
3226: } /* end i */
3227:
3228: for(ii=nlstate+1; ii<= nlstate+ndeath; ii++){
3229: for(jj=1; jj<= nlstate+ndeath; jj++){
3230: ps[ii][jj]=0;
3231: ps[ii][ii]=1;
3232: }
3233: }
1.296 brouard 3234: /* Added for prevbcast */ /* Transposed matrix too */
1.234 brouard 3235: for(jj=1; jj<= nlstate+ndeath; jj++){
3236: s1=0.;
3237: for(ii=1; ii<= nlstate+ndeath; ii++){
3238: s1+=ps[ii][jj];
3239: }
3240: for(ii=1; ii<= nlstate; ii++){
3241: ps[ii][jj]=ps[ii][jj]/s1;
3242: }
3243: }
3244: /* Transposition */
3245: for(jj=1; jj<= nlstate+ndeath; jj++){
3246: for(ii=jj; ii<= nlstate+ndeath; ii++){
3247: s1=ps[ii][jj];
3248: ps[ii][jj]=ps[jj][ii];
3249: ps[jj][ii]=s1;
3250: }
3251: }
3252: /* for(ii=1; ii<= nlstate+ndeath; ii++){ */
3253: /* for(jj=1; jj<= nlstate+ndeath; jj++){ */
3254: /* printf(" pmij ps[%d][%d]=%lf ",ii,jj,ps[ii][jj]); */
3255: /* } */
3256: /* printf("\n "); */
3257: /* } */
3258: /* printf("\n ");printf("%lf ",cov[2]);*/
3259: /*
3260: for(i=1; i<= npar; i++) printf("%f ",x[i]);
3261: goto end;*/
3262: return ps;
1.217 brouard 3263: }
3264:
3265:
1.126 brouard 3266: /**************** Product of 2 matrices ******************/
3267:
1.145 brouard 3268: double **matprod2(double **out, double **in,int nrl, int nrh, int ncl, int nch, int ncolol, int ncoloh, double **b)
1.126 brouard 3269: {
3270: /* Computes the matrix product of in(1,nrh-nrl+1)(1,nch-ncl+1) times
3271: b(1,nch-ncl+1)(1,ncoloh-ncolol+1) into out(...) */
3272: /* in, b, out are matrice of pointers which should have been initialized
3273: before: only the contents of out is modified. The function returns
3274: a pointer to pointers identical to out */
1.145 brouard 3275: int i, j, k;
1.126 brouard 3276: for(i=nrl; i<= nrh; i++)
1.145 brouard 3277: for(k=ncolol; k<=ncoloh; k++){
3278: out[i][k]=0.;
3279: for(j=ncl; j<=nch; j++)
3280: out[i][k] +=in[i][j]*b[j][k];
3281: }
1.126 brouard 3282: return out;
3283: }
3284:
3285:
3286: /************* Higher Matrix Product ***************/
3287:
1.235 brouard 3288: 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 3289: {
1.218 brouard 3290: /* Computes the transition matrix starting at age 'age' and combination of covariate values corresponding to ij over
1.126 brouard 3291: 'nhstepm*hstepm*stepm' months (i.e. until
3292: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3293: nhstepm*hstepm matrices.
3294: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3295: (typically every 2 years instead of every month which is too big
3296: for the memory).
3297: Model is determined by parameters x and covariates have to be
3298: included manually here.
3299:
3300: */
3301:
3302: int i, j, d, h, k;
1.131 brouard 3303: double **out, cov[NCOVMAX+1];
1.126 brouard 3304: double **newm;
1.187 brouard 3305: double agexact;
1.214 brouard 3306: double agebegin, ageend;
1.126 brouard 3307:
3308: /* Hstepm could be zero and should return the unit matrix */
3309: for (i=1;i<=nlstate+ndeath;i++)
3310: for (j=1;j<=nlstate+ndeath;j++){
3311: oldm[i][j]=(i==j ? 1.0 : 0.0);
3312: po[i][j][0]=(i==j ? 1.0 : 0.0);
3313: }
3314: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3315: for(h=1; h <=nhstepm; h++){
3316: for(d=1; d <=hstepm; d++){
3317: newm=savm;
3318: /* Covariates have to be included here again */
3319: cov[1]=1.;
1.214 brouard 3320: agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /* age just before transition */
1.187 brouard 3321: cov[2]=agexact;
3322: if(nagesqr==1)
1.227 brouard 3323: cov[3]= agexact*agexact;
1.235 brouard 3324: for (k=1; k<=nsd;k++) { /* For single dummy covariates only */
3325: /* Here comes the value of the covariate 'ij' after renumbering k with single dummy covariates */
3326: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3327: /* printf("hpxij Dummy combi=%d k=%d TvarsD[%d]=V%d TvarsDind[%d]=%d nbcode=%d cov=%lf codtabm(%d,Tvar[%d])=%d \n",ij,k, k, TvarsD[k],k,TvarsDind[k],nbcode[TvarsD[k]][codtabm(ij,k)],cov[2+nagesqr+TvarsDind[k]], ij, k, codtabm(ij,k)); */
3328: }
3329: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3330: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3331: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3332: /* 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]); */
3333: }
3334: for (k=1; k<=cptcovage;k++){
3335: if(Dummy[Tvar[Tage[k]]]){
3336: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3337: } else{
3338: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3339: }
3340: /* 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]); */
3341: }
3342: for (k=1; k<=cptcovprod;k++){ /* */
3343: /* printf("hPxij Prod ij=%d k=%d Tprod[%d]=%d Tvard[%d][1]=V%d, Tvard[%d][2]=V%d\n",ij,k,k,Tprod[k], k,Tvard[k][1], k,Tvard[k][2]); */
3344: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)] * nbcode[Tvard[k][2]][codtabm(ij,k)];
3345: }
3346: /* for (k=1; k<=cptcovn;k++) */
3347: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3348: /* for (k=1; k<=cptcovage;k++) /\* Should start at cptcovn+1 *\/ */
3349: /* cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2]; */
3350: /* for (k=1; k<=cptcovprod;k++) /\* Useless because included in cptcovn *\/ */
3351: /* cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)]; */
1.227 brouard 3352:
3353:
1.126 brouard 3354: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3355: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.218 brouard 3356: /* right multiplication of oldm by the current matrix */
1.126 brouard 3357: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath,
3358: pmij(pmmij,cov,ncovmodel,x,nlstate));
1.217 brouard 3359: /* if((int)age == 70){ */
3360: /* printf(" Forward hpxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3361: /* for(i=1; i<=nlstate+ndeath; i++) { */
3362: /* printf("%d pmmij ",i); */
3363: /* for(j=1;j<=nlstate+ndeath;j++) { */
3364: /* printf("%f ",pmmij[i][j]); */
3365: /* } */
3366: /* printf(" oldm "); */
3367: /* for(j=1;j<=nlstate+ndeath;j++) { */
3368: /* printf("%f ",oldm[i][j]); */
3369: /* } */
3370: /* printf("\n"); */
3371: /* } */
3372: /* } */
1.126 brouard 3373: savm=oldm;
3374: oldm=newm;
3375: }
3376: for(i=1; i<=nlstate+ndeath; i++)
3377: for(j=1;j<=nlstate+ndeath;j++) {
1.267 brouard 3378: po[i][j][h]=newm[i][j];
3379: /*if(h==nhstepm) printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]);*/
1.126 brouard 3380: }
1.128 brouard 3381: /*printf("h=%d ",h);*/
1.126 brouard 3382: } /* end h */
1.267 brouard 3383: /* printf("\n H=%d \n",h); */
1.126 brouard 3384: return po;
3385: }
3386:
1.217 brouard 3387: /************* Higher Back Matrix Product ***************/
1.218 brouard 3388: /* 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 3389: 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 3390: {
1.266 brouard 3391: /* For a combination of dummy covariate ij, computes the transition matrix starting at age 'age' over
1.217 brouard 3392: 'nhstepm*hstepm*stepm' months (i.e. until
1.218 brouard 3393: age (in years) age+nhstepm*hstepm*stepm/12) by multiplying
3394: nhstepm*hstepm matrices.
3395: Output is stored in matrix po[i][j][h] for h every 'hstepm' step
3396: (typically every 2 years instead of every month which is too big
1.217 brouard 3397: for the memory).
1.218 brouard 3398: Model is determined by parameters x and covariates have to be
1.266 brouard 3399: included manually here. Then we use a call to bmij(x and cov)
3400: The addresss of po (p3mat allocated to the dimension of nhstepm) should be stored for output
1.222 brouard 3401: */
1.217 brouard 3402:
3403: int i, j, d, h, k;
1.266 brouard 3404: double **out, cov[NCOVMAX+1], **bmij();
3405: double **newm, ***newmm;
1.217 brouard 3406: double agexact;
3407: double agebegin, ageend;
1.222 brouard 3408: double **oldm, **savm;
1.217 brouard 3409:
1.266 brouard 3410: newmm=po; /* To be saved */
3411: oldm=oldms;savm=savms; /* Global pointers */
1.217 brouard 3412: /* Hstepm could be zero and should return the unit matrix */
3413: for (i=1;i<=nlstate+ndeath;i++)
3414: for (j=1;j<=nlstate+ndeath;j++){
3415: oldm[i][j]=(i==j ? 1.0 : 0.0);
3416: po[i][j][0]=(i==j ? 1.0 : 0.0);
3417: }
3418: /* Even if hstepm = 1, at least one multiplication by the unit matrix */
3419: for(h=1; h <=nhstepm; h++){
3420: for(d=1; d <=hstepm; d++){
3421: newm=savm;
3422: /* Covariates have to be included here again */
3423: cov[1]=1.;
1.271 brouard 3424: agexact=age-( (h-1)*hstepm + (d) )*stepm/YEARM; /* age just before transition, d or d-1? */
1.217 brouard 3425: /* agexact=age+((h-1)*hstepm + (d-1))*stepm/YEARM; /\* age just before transition *\/ */
3426: cov[2]=agexact;
3427: if(nagesqr==1)
1.222 brouard 3428: cov[3]= agexact*agexact;
1.266 brouard 3429: for (k=1; k<=cptcovn;k++){
3430: /* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,k)]; */
3431: /* /\* cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; *\/ */
3432: cov[2+nagesqr+TvarsDind[k]]=nbcode[TvarsD[k]][codtabm(ij,k)];
3433: /* 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)); */
3434: }
1.267 brouard 3435: for (k=1; k<=nsq;k++) { /* For single varying covariates only */
3436: /* Here comes the value of quantitative after renumbering k with single quantitative covariates */
3437: cov[2+nagesqr+TvarsQind[k]]=Tqresult[nres][k];
3438: /* 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]); */
3439: }
3440: for (k=1; k<=cptcovage;k++){ /* Should start at cptcovn+1 */
3441: if(Dummy[Tvar[Tage[k]]]){
3442: cov[2+nagesqr+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
3443: } else{
3444: cov[2+nagesqr+Tage[k]]=Tqresult[nres][k];
3445: }
3446: /* 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]); */
3447: }
3448: for (k=1; k<=cptcovprod;k++){ /* Useless because included in cptcovn */
1.222 brouard 3449: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.267 brouard 3450: }
1.217 brouard 3451: /*printf("hxi cptcov=%d cptcode=%d\n",cptcov,cptcode);*/
3452: /*printf("h=%d d=%d age=%f cov=%f\n",h,d,age,cov[2]);*/
1.267 brouard 3453:
1.218 brouard 3454: /* Careful transposed matrix */
1.266 brouard 3455: /* age is in cov[2], prevacurrent at beginning of transition. */
1.218 brouard 3456: /* out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent, dnewm, doldm, dsavm,ij),\ */
1.222 brouard 3457: /* 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm); */
1.218 brouard 3458: out=matprod2(newm, bmij(pmmij,cov,ncovmodel,x,nlstate,prevacurrent,ij),\
1.222 brouard 3459: 1,nlstate+ndeath,1,nlstate+ndeath,1,nlstate+ndeath, oldm);
1.217 brouard 3460: /* if((int)age == 70){ */
3461: /* printf(" Backward hbxij age=%d agexact=%f d=%d nhstepm=%d hstepm=%d\n", (int) age, agexact, d, nhstepm, hstepm); */
3462: /* for(i=1; i<=nlstate+ndeath; i++) { */
3463: /* printf("%d pmmij ",i); */
3464: /* for(j=1;j<=nlstate+ndeath;j++) { */
3465: /* printf("%f ",pmmij[i][j]); */
3466: /* } */
3467: /* printf(" oldm "); */
3468: /* for(j=1;j<=nlstate+ndeath;j++) { */
3469: /* printf("%f ",oldm[i][j]); */
3470: /* } */
3471: /* printf("\n"); */
3472: /* } */
3473: /* } */
3474: savm=oldm;
3475: oldm=newm;
3476: }
3477: for(i=1; i<=nlstate+ndeath; i++)
3478: for(j=1;j<=nlstate+ndeath;j++) {
1.222 brouard 3479: po[i][j][h]=newm[i][j];
1.268 brouard 3480: /* if(h==nhstepm) */
3481: /* printf("po[%d][%d][%d]=%f ",i,j,h,po[i][j][h]); */
1.217 brouard 3482: }
1.268 brouard 3483: /* printf("h=%d %.1f ",h, agexact); */
1.217 brouard 3484: } /* end h */
1.268 brouard 3485: /* printf("\n H=%d nhs=%d \n",h, nhstepm); */
1.217 brouard 3486: return po;
3487: }
3488:
3489:
1.162 brouard 3490: #ifdef NLOPT
3491: double myfunc(unsigned n, const double *p1, double *grad, void *pd){
3492: double fret;
3493: double *xt;
3494: int j;
3495: myfunc_data *d2 = (myfunc_data *) pd;
3496: /* xt = (p1-1); */
3497: xt=vector(1,n);
3498: for (j=1;j<=n;j++) xt[j]=p1[j-1]; /* xt[1]=p1[0] */
3499:
3500: fret=(d2->function)(xt); /* p xt[1]@8 is fine */
3501: /* fret=(*func)(xt); /\* p xt[1]@8 is fine *\/ */
3502: printf("Function = %.12lf ",fret);
3503: for (j=1;j<=n;j++) printf(" %d %.8lf", j, xt[j]);
3504: printf("\n");
3505: free_vector(xt,1,n);
3506: return fret;
3507: }
3508: #endif
1.126 brouard 3509:
3510: /*************** log-likelihood *************/
3511: double func( double *x)
3512: {
1.226 brouard 3513: int i, ii, j, k, mi, d, kk;
3514: int ioffset=0;
3515: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
3516: double **out;
3517: double lli; /* Individual log likelihood */
3518: int s1, s2;
1.228 brouard 3519: 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 3520: double bbh, survp;
3521: long ipmx;
3522: double agexact;
3523: /*extern weight */
3524: /* We are differentiating ll according to initial status */
3525: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3526: /*for(i=1;i<imx;i++)
3527: printf(" %d\n",s[4][i]);
3528: */
1.162 brouard 3529:
1.226 brouard 3530: ++countcallfunc;
1.162 brouard 3531:
1.226 brouard 3532: cov[1]=1.;
1.126 brouard 3533:
1.226 brouard 3534: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3535: ioffset=0;
1.226 brouard 3536: if(mle==1){
3537: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3538: /* Computes the values of the ncovmodel covariates of the model
3539: depending if the covariates are fixed or varying (age dependent) and stores them in cov[]
3540: Then computes with function pmij which return a matrix p[i][j] giving the elementary probability
3541: to be observed in j being in i according to the model.
3542: */
1.243 brouard 3543: ioffset=2+nagesqr ;
1.233 brouard 3544: /* Fixed */
1.234 brouard 3545: for (k=1; k<=ncovf;k++){ /* Simple and product fixed covariates without age* products */
3546: 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)*/
3547: }
1.226 brouard 3548: /* In model V2+V1*V4+age*V3+V3*V2 Tvar[1] is V2, Tvar[2=V1*V4]
3549: is 6, Tvar[3=age*V3] should not be computed because of age Tvar[4=V3*V2]
3550: has been calculated etc */
3551: /* For an individual i, wav[i] gives the number of effective waves */
3552: /* We compute the contribution to Likelihood of each effective transition
3553: mw[mi][i] is real wave of the mi th effectve wave */
3554: /* Then statuses are computed at each begin and end of an effective wave s1=s[ mw[mi][i] ][i];
3555: s2=s[mw[mi+1][i]][i];
3556: And the iv th varying covariate is the cotvar[mw[mi+1][i]][iv][i]
3557: But if the variable is not in the model TTvar[iv] is the real variable effective in the model:
3558: meaning that decodemodel should be used cotvar[mw[mi+1][i]][TTvar[iv]][i]
3559: */
3560: for(mi=1; mi<= wav[i]-1; mi++){
1.234 brouard 3561: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3562: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3563: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
1.234 brouard 3564: }
3565: for (ii=1;ii<=nlstate+ndeath;ii++)
3566: for (j=1;j<=nlstate+ndeath;j++){
3567: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3568: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3569: }
3570: for(d=0; d<dh[mi][i]; d++){
3571: newm=savm;
3572: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3573: cov[2]=agexact;
3574: if(nagesqr==1)
3575: cov[3]= agexact*agexact; /* Should be changed here */
3576: for (kk=1; kk<=cptcovage;kk++) {
1.242 brouard 3577: if(!FixedV[Tvar[Tage[kk]]])
1.234 brouard 3578: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact; /* Tage[kk] gives the data-covariate associated with age */
1.242 brouard 3579: else
3580: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
1.234 brouard 3581: }
3582: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3583: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3584: savm=oldm;
3585: oldm=newm;
3586: } /* end mult */
3587:
3588: /*lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]);*/ /* Original formula */
3589: /* But now since version 0.9 we anticipate for bias at large stepm.
3590: * If stepm is larger than one month (smallest stepm) and if the exact delay
3591: * (in months) between two waves is not a multiple of stepm, we rounded to
3592: * the nearest (and in case of equal distance, to the lowest) interval but now
3593: * we keep into memory the bias bh[mi][i] and also the previous matrix product
3594: * (i.e to dh[mi][i]-1) saved in 'savm'. Then we inter(extra)polate the
3595: * probability in order to take into account the bias as a fraction of the way
1.231 brouard 3596: * from savm to out if bh is negative or even beyond if bh is positive. bh varies
3597: * -stepm/2 to stepm/2 .
3598: * For stepm=1 the results are the same as for previous versions of Imach.
3599: * For stepm > 1 the results are less biased than in previous versions.
3600: */
1.234 brouard 3601: s1=s[mw[mi][i]][i];
3602: s2=s[mw[mi+1][i]][i];
3603: bbh=(double)bh[mi][i]/(double)stepm;
3604: /* bias bh is positive if real duration
3605: * is higher than the multiple of stepm and negative otherwise.
3606: */
3607: /* lli= (savm[s1][s2]>1.e-8 ?(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]):log((1.+bbh)*out[s1][s2]));*/
3608: if( s2 > nlstate){
3609: /* i.e. if s2 is a death state and if the date of death is known
3610: then the contribution to the likelihood is the probability to
3611: die between last step unit time and current step unit time,
3612: which is also equal to probability to die before dh
3613: minus probability to die before dh-stepm .
3614: In version up to 0.92 likelihood was computed
3615: as if date of death was unknown. Death was treated as any other
3616: health state: the date of the interview describes the actual state
3617: and not the date of a change in health state. The former idea was
3618: to consider that at each interview the state was recorded
3619: (healthy, disable or death) and IMaCh was corrected; but when we
3620: introduced the exact date of death then we should have modified
3621: the contribution of an exact death to the likelihood. This new
3622: contribution is smaller and very dependent of the step unit
3623: stepm. It is no more the probability to die between last interview
3624: and month of death but the probability to survive from last
3625: interview up to one month before death multiplied by the
3626: probability to die within a month. Thanks to Chris
3627: Jackson for correcting this bug. Former versions increased
3628: mortality artificially. The bad side is that we add another loop
3629: which slows down the processing. The difference can be up to 10%
3630: lower mortality.
3631: */
3632: /* If, at the beginning of the maximization mostly, the
3633: cumulative probability or probability to be dead is
3634: constant (ie = 1) over time d, the difference is equal to
3635: 0. out[s1][3] = savm[s1][3]: probability, being at state
3636: s1 at precedent wave, to be dead a month before current
3637: wave is equal to probability, being at state s1 at
3638: precedent wave, to be dead at mont of the current
3639: wave. Then the observed probability (that this person died)
3640: is null according to current estimated parameter. In fact,
3641: it should be very low but not zero otherwise the log go to
3642: infinity.
3643: */
1.183 brouard 3644: /* #ifdef INFINITYORIGINAL */
3645: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3646: /* #else */
3647: /* if ((out[s1][s2] - savm[s1][s2]) < mytinydouble) */
3648: /* lli=log(mytinydouble); */
3649: /* else */
3650: /* lli=log(out[s1][s2] - savm[s1][s2]); */
3651: /* #endif */
1.226 brouard 3652: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3653:
1.226 brouard 3654: } else if ( s2==-1 ) { /* alive */
3655: for (j=1,survp=0. ; j<=nlstate; j++)
3656: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3657: /*survp += out[s1][j]; */
3658: lli= log(survp);
3659: }
3660: else if (s2==-4) {
3661: for (j=3,survp=0. ; j<=nlstate; j++)
3662: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3663: lli= log(survp);
3664: }
3665: else if (s2==-5) {
3666: for (j=1,survp=0. ; j<=2; j++)
3667: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3668: lli= log(survp);
3669: }
3670: else{
3671: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3672: /* 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 */
3673: }
3674: /*lli=(1.+bbh)*log(out[s1][s2])- bbh*log(savm[s1][s2]);*/
3675: /*if(lli ==000.0)*/
3676: /*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); */
3677: ipmx +=1;
3678: sw += weight[i];
3679: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3680: /* if (lli < log(mytinydouble)){ */
3681: /* 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); */
3682: /* 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]); */
3683: /* } */
3684: } /* end of wave */
3685: } /* end of individual */
3686: } else if(mle==2){
3687: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3688: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3689: for(mi=1; mi<= wav[i]-1; mi++){
3690: for (ii=1;ii<=nlstate+ndeath;ii++)
3691: for (j=1;j<=nlstate+ndeath;j++){
3692: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3693: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3694: }
3695: for(d=0; d<=dh[mi][i]; d++){
3696: newm=savm;
3697: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3698: cov[2]=agexact;
3699: if(nagesqr==1)
3700: cov[3]= agexact*agexact;
3701: for (kk=1; kk<=cptcovage;kk++) {
3702: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3703: }
3704: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3705: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3706: savm=oldm;
3707: oldm=newm;
3708: } /* end mult */
3709:
3710: s1=s[mw[mi][i]][i];
3711: s2=s[mw[mi+1][i]][i];
3712: bbh=(double)bh[mi][i]/(double)stepm;
3713: 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 */
3714: ipmx +=1;
3715: sw += weight[i];
3716: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3717: } /* end of wave */
3718: } /* end of individual */
3719: } else if(mle==3){ /* exponential inter-extrapolation */
3720: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3721: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3722: for(mi=1; mi<= wav[i]-1; mi++){
3723: for (ii=1;ii<=nlstate+ndeath;ii++)
3724: for (j=1;j<=nlstate+ndeath;j++){
3725: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3726: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3727: }
3728: for(d=0; d<dh[mi][i]; d++){
3729: newm=savm;
3730: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3731: cov[2]=agexact;
3732: if(nagesqr==1)
3733: cov[3]= agexact*agexact;
3734: for (kk=1; kk<=cptcovage;kk++) {
3735: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3736: }
3737: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3738: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3739: savm=oldm;
3740: oldm=newm;
3741: } /* end mult */
3742:
3743: s1=s[mw[mi][i]][i];
3744: s2=s[mw[mi+1][i]][i];
3745: bbh=(double)bh[mi][i]/(double)stepm;
3746: 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 */
3747: ipmx +=1;
3748: sw += weight[i];
3749: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3750: } /* end of wave */
3751: } /* end of individual */
3752: }else if (mle==4){ /* ml=4 no inter-extrapolation */
3753: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3754: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3755: for(mi=1; mi<= wav[i]-1; mi++){
3756: for (ii=1;ii<=nlstate+ndeath;ii++)
3757: for (j=1;j<=nlstate+ndeath;j++){
3758: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3759: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3760: }
3761: for(d=0; d<dh[mi][i]; d++){
3762: newm=savm;
3763: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3764: cov[2]=agexact;
3765: if(nagesqr==1)
3766: cov[3]= agexact*agexact;
3767: for (kk=1; kk<=cptcovage;kk++) {
3768: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3769: }
1.126 brouard 3770:
1.226 brouard 3771: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3772: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3773: savm=oldm;
3774: oldm=newm;
3775: } /* end mult */
3776:
3777: s1=s[mw[mi][i]][i];
3778: s2=s[mw[mi+1][i]][i];
3779: if( s2 > nlstate){
3780: lli=log(out[s1][s2] - savm[s1][s2]);
3781: } else if ( s2==-1 ) { /* alive */
3782: for (j=1,survp=0. ; j<=nlstate; j++)
3783: survp += out[s1][j];
3784: lli= log(survp);
3785: }else{
3786: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3787: }
3788: ipmx +=1;
3789: sw += weight[i];
3790: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.126 brouard 3791: /* 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 3792: } /* end of wave */
3793: } /* end of individual */
3794: }else{ /* ml=5 no inter-extrapolation no jackson =0.8a */
3795: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
3796: for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i];
3797: for(mi=1; mi<= wav[i]-1; mi++){
3798: for (ii=1;ii<=nlstate+ndeath;ii++)
3799: for (j=1;j<=nlstate+ndeath;j++){
3800: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3801: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3802: }
3803: for(d=0; d<dh[mi][i]; d++){
3804: newm=savm;
3805: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM;
3806: cov[2]=agexact;
3807: if(nagesqr==1)
3808: cov[3]= agexact*agexact;
3809: for (kk=1; kk<=cptcovage;kk++) {
3810: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3811: }
1.126 brouard 3812:
1.226 brouard 3813: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3814: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3815: savm=oldm;
3816: oldm=newm;
3817: } /* end mult */
3818:
3819: s1=s[mw[mi][i]][i];
3820: s2=s[mw[mi+1][i]][i];
3821: lli=log(out[s[mw[mi][i]][i]][s[mw[mi+1][i]][i]]); /* Original formula */
3822: ipmx +=1;
3823: sw += weight[i];
3824: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
3825: /*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]);*/
3826: } /* end of wave */
3827: } /* end of individual */
3828: } /* End of if */
3829: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3830: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3831: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3832: return -l;
1.126 brouard 3833: }
3834:
3835: /*************** log-likelihood *************/
3836: double funcone( double *x)
3837: {
1.228 brouard 3838: /* Same as func but slower because of a lot of printf and if */
1.126 brouard 3839: int i, ii, j, k, mi, d, kk;
1.228 brouard 3840: int ioffset=0;
1.131 brouard 3841: double l, ll[NLSTATEMAX+1], cov[NCOVMAX+1];
1.126 brouard 3842: double **out;
3843: double lli; /* Individual log likelihood */
3844: double llt;
3845: int s1, s2;
1.228 brouard 3846: int iv=0, iqv=0, itv=0, iqtv=0 ; /* Index of varying covariate, fixed quantitative cov, time varying covariate, quantitative time varying covariate */
3847:
1.126 brouard 3848: double bbh, survp;
1.187 brouard 3849: double agexact;
1.214 brouard 3850: double agebegin, ageend;
1.126 brouard 3851: /*extern weight */
3852: /* We are differentiating ll according to initial status */
3853: /* for (i=1;i<=npar;i++) printf("%f ", x[i]);*/
3854: /*for(i=1;i<imx;i++)
3855: printf(" %d\n",s[4][i]);
3856: */
3857: cov[1]=1.;
3858:
3859: for(k=1; k<=nlstate; k++) ll[k]=0.;
1.224 brouard 3860: ioffset=0;
3861: for (i=1,ipmx=0, sw=0.; i<=imx; i++){
1.243 brouard 3862: /* ioffset=2+nagesqr+cptcovage; */
3863: ioffset=2+nagesqr;
1.232 brouard 3864: /* Fixed */
1.224 brouard 3865: /* for (k=1; k<=cptcovn;k++) cov[2+nagesqr+k]=covar[Tvar[k]][i]; */
1.232 brouard 3866: /* for (k=1; k<=ncoveff;k++){ /\* Simple and product fixed Dummy covariates without age* products *\/ */
1.311 brouard 3867: 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 3868: 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)*/
3869: /* cov[ioffset+TvarFind[1]]=covar[Tvar[TvarFind[1]]][i]; */
3870: /* cov[2+6]=covar[Tvar[6]][i]; */
3871: /* cov[2+6]=covar[2][i]; V2 */
3872: /* cov[TvarFind[2]]=covar[Tvar[TvarFind[2]]][i]; */
3873: /* cov[2+7]=covar[Tvar[7]][i]; */
3874: /* cov[2+7]=covar[7][i]; V7=V1*V2 */
3875: /* cov[TvarFind[3]]=covar[Tvar[TvarFind[3]]][i]; */
3876: /* cov[2+9]=covar[Tvar[9]][i]; */
3877: /* cov[2+9]=covar[1][i]; V1 */
1.225 brouard 3878: }
1.232 brouard 3879: /* for (k=1; k<=nqfveff;k++){ /\* Simple and product fixed Quantitative covariates without age* products *\/ */
3880: /* 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?)*\/ */
3881: /* } */
1.231 brouard 3882: /* for(iqv=1; iqv <= nqfveff; iqv++){ /\* Quantitative fixed covariates *\/ */
3883: /* cov[++ioffset]=coqvar[Tvar[iqv]][i]; /\* Only V2 k=6 and V1*V2 7 *\/ */
3884: /* } */
1.225 brouard 3885:
1.233 brouard 3886:
3887: for(mi=1; mi<= wav[i]-1; mi++){ /* Varying with waves */
1.232 brouard 3888: /* Wave varying (but not age varying) */
3889: for(k=1; k <= ncovv ; k++){ /* Varying covariates (single and product but no age )*/
1.242 brouard 3890: /* cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]][i]; */
3891: cov[ioffset+TvarVind[k]]=cotvar[mw[mi][i]][Tvar[TvarVind[k]]-ncovcol-nqv][i];
3892: }
1.232 brouard 3893: /* for(itv=1; itv <= ntveff; itv++){ /\* Varying dummy covariates (single??)*\/ */
1.242 brouard 3894: /* iv= Tvar[Tmodelind[ioffset-2-nagesqr-cptcovage+itv]]-ncovcol-nqv; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3895: /* cov[ioffset+iv]=cotvar[mw[mi][i]][iv][i]; */
3896: /* k=ioffset-2-nagesqr-cptcovage+itv; /\* position in simple model *\/ */
3897: /* cov[ioffset+itv]=cotvar[mw[mi][i]][TmodelInvind[itv]][i]; */
3898: /* 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 3899: /* for(iqtv=1; iqtv <= nqtveff; iqtv++){ /\* Varying quantitatives covariates *\/ */
1.242 brouard 3900: /* iv=TmodelInvQind[iqtv]; /\* Counting the # varying covariate from 1 to ntveff *\/ */
3901: /* /\* 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]); *\/ */
3902: /* cov[ioffset+ntveff+iqtv]=cotqvar[mw[mi][i]][TmodelInvQind[iqtv]][i]; */
1.232 brouard 3903: /* } */
1.126 brouard 3904: for (ii=1;ii<=nlstate+ndeath;ii++)
1.242 brouard 3905: for (j=1;j<=nlstate+ndeath;j++){
3906: oldm[ii][j]=(ii==j ? 1.0 : 0.0);
3907: savm[ii][j]=(ii==j ? 1.0 : 0.0);
3908: }
1.214 brouard 3909:
3910: agebegin=agev[mw[mi][i]][i]; /* Age at beginning of effective wave */
3911: ageend=agev[mw[mi][i]][i] + (dh[mi][i])*stepm/YEARM; /* Age at end of effective wave and at the end of transition */
3912: for(d=0; d<dh[mi][i]; d++){ /* Delay between two effective waves */
1.247 brouard 3913: /* for(d=0; d<=0; d++){ /\* Delay between two effective waves Only one matrix to speed up*\/ */
1.242 brouard 3914: /*dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
3915: and mw[mi+1][i]. dh depends on stepm.*/
3916: newm=savm;
1.247 brouard 3917: agexact=agev[mw[mi][i]][i]+d*stepm/YEARM; /* Here d is needed */
1.242 brouard 3918: cov[2]=agexact;
3919: if(nagesqr==1)
3920: cov[3]= agexact*agexact;
3921: for (kk=1; kk<=cptcovage;kk++) {
3922: if(!FixedV[Tvar[Tage[kk]]])
3923: cov[Tage[kk]+2+nagesqr]=covar[Tvar[Tage[kk]]][i]*agexact;
3924: else
3925: cov[Tage[kk]+2+nagesqr]=cotvar[mw[mi][i]][Tvar[Tage[kk]]-ncovcol-nqv][i]*agexact;
3926: }
3927: /* printf("i=%d,mi=%d,d=%d,mw[mi][i]=%d\n",i, mi,d,mw[mi][i]); */
3928: /* savm=pmij(pmmij,cov,ncovmodel,x,nlstate); */
3929: out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath,
3930: 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate));
3931: /* out=matprod2(newm,oldm,1,nlstate+ndeath,1,nlstate+ndeath, */
3932: /* 1,nlstate+ndeath,pmij(pmmij,cov,ncovmodel,x,nlstate)); */
3933: savm=oldm;
3934: oldm=newm;
1.126 brouard 3935: } /* end mult */
3936:
3937: s1=s[mw[mi][i]][i];
3938: s2=s[mw[mi+1][i]][i];
1.217 brouard 3939: /* if(s2==-1){ */
1.268 brouard 3940: /* printf(" ERROR s1=%d, s2=%d i=%d \n", s1, s2, i); */
1.217 brouard 3941: /* /\* exit(1); *\/ */
3942: /* } */
1.126 brouard 3943: bbh=(double)bh[mi][i]/(double)stepm;
3944: /* bias is positive if real duration
3945: * is higher than the multiple of stepm and negative otherwise.
3946: */
3947: if( s2 > nlstate && (mle <5) ){ /* Jackson */
1.242 brouard 3948: lli=log(out[s1][s2] - savm[s1][s2]);
1.216 brouard 3949: } else if ( s2==-1 ) { /* alive */
1.242 brouard 3950: for (j=1,survp=0. ; j<=nlstate; j++)
3951: survp += (1.+bbh)*out[s1][j]- bbh*savm[s1][j];
3952: lli= log(survp);
1.126 brouard 3953: }else if (mle==1){
1.242 brouard 3954: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
1.126 brouard 3955: } else if(mle==2){
1.242 brouard 3956: 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 3957: } else if(mle==3){ /* exponential inter-extrapolation */
1.242 brouard 3958: 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 3959: } else if (mle==4){ /* mle=4 no inter-extrapolation */
1.242 brouard 3960: lli=log(out[s1][s2]); /* Original formula */
1.136 brouard 3961: } else{ /* mle=0 back to 1 */
1.242 brouard 3962: lli= log((1.+bbh)*out[s1][s2]- bbh*savm[s1][s2]); /* linear interpolation */
3963: /*lli=log(out[s1][s2]); */ /* Original formula */
1.126 brouard 3964: } /* End of if */
3965: ipmx +=1;
3966: sw += weight[i];
3967: ll[s[mw[mi][i]][i]] += 2*weight[i]*lli;
1.132 brouard 3968: /*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 3969: if(globpr){
1.246 brouard 3970: fprintf(ficresilk,"%09ld %6.1f %6.1f %6d %2d %2d %2d %2d %3d %15.6f %8.4f %8.3f\
1.126 brouard 3971: %11.6f %11.6f %11.6f ", \
1.242 brouard 3972: 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 3973: 2*weight[i]*lli,(s2==-1? -1: out[s1][s2]),(s2==-1? -1: savm[s1][s2]));
1.242 brouard 3974: for(k=1,llt=0.,l=0.; k<=nlstate; k++){
3975: llt +=ll[k]*gipmx/gsw;
3976: fprintf(ficresilk," %10.6f",-ll[k]*gipmx/gsw);
3977: }
3978: fprintf(ficresilk," %10.6f\n", -llt);
1.126 brouard 3979: }
1.232 brouard 3980: } /* end of wave */
3981: } /* end of individual */
3982: for(k=1,l=0.; k<=nlstate; k++) l += ll[k];
3983: /* printf("l1=%f l2=%f ",ll[1],ll[2]); */
3984: l= l*ipmx/sw; /* To get the same order of magnitude as if weight=1 for every body */
3985: if(globpr==0){ /* First time we count the contributions and weights */
3986: gipmx=ipmx;
3987: gsw=sw;
3988: }
3989: return -l;
1.126 brouard 3990: }
3991:
3992:
3993: /*************** function likelione ***********/
1.292 brouard 3994: void likelione(FILE *ficres,double p[], int npar, int nlstate, int *globpri, long *ipmx, double *sw, double *fretone, double (*func)(double []))
1.126 brouard 3995: {
3996: /* This routine should help understanding what is done with
3997: the selection of individuals/waves and
3998: to check the exact contribution to the likelihood.
3999: Plotting could be done.
4000: */
4001: int k;
4002:
4003: if(*globpri !=0){ /* Just counts and sums, no printings */
1.201 brouard 4004: strcpy(fileresilk,"ILK_");
1.202 brouard 4005: strcat(fileresilk,fileresu);
1.126 brouard 4006: if((ficresilk=fopen(fileresilk,"w"))==NULL) {
4007: printf("Problem with resultfile: %s\n", fileresilk);
4008: fprintf(ficlog,"Problem with resultfile: %s\n", fileresilk);
4009: }
1.214 brouard 4010: 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");
4011: fprintf(ficresilk, "#num_i ageb agend i s1 s2 mi mw dh likeli weight %%weight 2wlli out sav ");
1.126 brouard 4012: /* i,s1,s2,mi,mw[mi][i],dh[mi][i],exp(lli),weight[i],2*weight[i]*lli,out[s1][s2],savm[s1][s2]); */
4013: for(k=1; k<=nlstate; k++)
4014: fprintf(ficresilk," -2*gipw/gsw*weight*ll[%d]++",k);
4015: fprintf(ficresilk," -2*gipw/gsw*weight*ll(total)\n");
4016: }
4017:
1.292 brouard 4018: *fretone=(*func)(p);
1.126 brouard 4019: if(*globpri !=0){
4020: fclose(ficresilk);
1.205 brouard 4021: if (mle ==0)
4022: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with initial parameters and mle = %d.",mle);
4023: else if(mle >=1)
4024: fprintf(fichtm,"\n<br>File of contributions to the likelihood computed with optimized parameters mle = %d.",mle);
4025: 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 4026: fprintf(fichtm,"\n<br>Equation of the model: <b>model=1+age+%s</b><br>\n",model);
1.208 brouard 4027:
4028: for (k=1; k<= nlstate ; k++) {
1.211 brouard 4029: 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 4030: <img src=\"%s-p%dj.png\">",k,k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k,subdirf2(optionfilefiname,"ILK_"),k);
4031: }
1.207 brouard 4032: 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 4033: <img src=\"%s-ori.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4034: fprintf(fichtm,"<br>- and by state of destination <a href=\"%s-dest.png\">%s-dest.png</a><br> \
1.204 brouard 4035: <img src=\"%s-dest.png\">",subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"),subdirf2(optionfilefiname,"ILK_"));
1.207 brouard 4036: fflush(fichtm);
1.205 brouard 4037: }
1.126 brouard 4038: return;
4039: }
4040:
4041:
4042: /*********** Maximum Likelihood Estimation ***************/
4043:
4044: void mlikeli(FILE *ficres,double p[], int npar, int ncovmodel, int nlstate, double ftol, double (*func)(double []))
4045: {
1.165 brouard 4046: int i,j, iter=0;
1.126 brouard 4047: double **xi;
4048: double fret;
4049: double fretone; /* Only one call to likelihood */
4050: /* char filerespow[FILENAMELENGTH];*/
1.162 brouard 4051:
4052: #ifdef NLOPT
4053: int creturn;
4054: nlopt_opt opt;
4055: /* double lb[9] = { -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL, -HUGE_VAL }; /\* lower bounds *\/ */
4056: double *lb;
4057: double minf; /* the minimum objective value, upon return */
4058: double * p1; /* Shifted parameters from 0 instead of 1 */
4059: myfunc_data dinst, *d = &dinst;
4060: #endif
4061:
4062:
1.126 brouard 4063: xi=matrix(1,npar,1,npar);
4064: for (i=1;i<=npar;i++)
4065: for (j=1;j<=npar;j++)
4066: xi[i][j]=(i==j ? 1.0 : 0.0);
4067: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.201 brouard 4068: strcpy(filerespow,"POW_");
1.126 brouard 4069: strcat(filerespow,fileres);
4070: if((ficrespow=fopen(filerespow,"w"))==NULL) {
4071: printf("Problem with resultfile: %s\n", filerespow);
4072: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
4073: }
4074: fprintf(ficrespow,"# Powell\n# iter -2*LL");
4075: for (i=1;i<=nlstate;i++)
4076: for(j=1;j<=nlstate+ndeath;j++)
4077: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
4078: fprintf(ficrespow,"\n");
1.162 brouard 4079: #ifdef POWELL
1.126 brouard 4080: powell(p,xi,npar,ftol,&iter,&fret,func);
1.162 brouard 4081: #endif
1.126 brouard 4082:
1.162 brouard 4083: #ifdef NLOPT
4084: #ifdef NEWUOA
4085: opt = nlopt_create(NLOPT_LN_NEWUOA,npar);
4086: #else
4087: opt = nlopt_create(NLOPT_LN_BOBYQA,npar);
4088: #endif
4089: lb=vector(0,npar-1);
4090: for (i=0;i<npar;i++) lb[i]= -HUGE_VAL;
4091: nlopt_set_lower_bounds(opt, lb);
4092: nlopt_set_initial_step1(opt, 0.1);
4093:
4094: p1= (p+1); /* p *(p+1)@8 and p *(p1)@8 are equal p1[0]=p[1] */
4095: d->function = func;
4096: printf(" Func %.12lf \n",myfunc(npar,p1,NULL,d));
4097: nlopt_set_min_objective(opt, myfunc, d);
4098: nlopt_set_xtol_rel(opt, ftol);
4099: if ((creturn=nlopt_optimize(opt, p1, &minf)) < 0) {
4100: printf("nlopt failed! %d\n",creturn);
4101: }
4102: else {
4103: printf("found minimum after %d evaluations (NLOPT=%d)\n", countcallfunc ,NLOPT);
4104: printf("found minimum at f(%g,%g) = %0.10g\n", p[0], p[1], minf);
4105: iter=1; /* not equal */
4106: }
4107: nlopt_destroy(opt);
4108: #endif
1.126 brouard 4109: free_matrix(xi,1,npar,1,npar);
4110: fclose(ficrespow);
1.203 brouard 4111: printf("\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
4112: fprintf(ficlog,"\n#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.180 brouard 4113: fprintf(ficres,"#Number of iterations & function calls = %d & %d, -2 Log likelihood = %.12f\n",iter, countcallfunc,func(p));
1.126 brouard 4114:
4115: }
4116:
4117: /**** Computes Hessian and covariance matrix ***/
1.203 brouard 4118: void hesscov(double **matcov, double **hess, double p[], int npar, double delti[], double ftolhess, double (*func)(double []))
1.126 brouard 4119: {
4120: double **a,**y,*x,pd;
1.203 brouard 4121: /* double **hess; */
1.164 brouard 4122: int i, j;
1.126 brouard 4123: int *indx;
4124:
4125: double hessii(double p[], double delta, int theta, double delti[],double (*func)(double []),int npar);
1.203 brouard 4126: double hessij(double p[], double **hess, double delti[], int i, int j,double (*func)(double []),int npar);
1.126 brouard 4127: void lubksb(double **a, int npar, int *indx, double b[]) ;
4128: void ludcmp(double **a, int npar, int *indx, double *d) ;
4129: double gompertz(double p[]);
1.203 brouard 4130: /* hess=matrix(1,npar,1,npar); */
1.126 brouard 4131:
4132: printf("\nCalculation of the hessian matrix. Wait...\n");
4133: fprintf(ficlog,"\nCalculation of the hessian matrix. Wait...\n");
4134: for (i=1;i<=npar;i++){
1.203 brouard 4135: printf("%d-",i);fflush(stdout);
4136: fprintf(ficlog,"%d-",i);fflush(ficlog);
1.126 brouard 4137:
4138: hess[i][i]=hessii(p,ftolhess,i,delti,func,npar);
4139:
4140: /* printf(" %f ",p[i]);
4141: printf(" %lf %lf %lf",hess[i][i],ftolhess,delti[i]);*/
4142: }
4143:
4144: for (i=1;i<=npar;i++) {
4145: for (j=1;j<=npar;j++) {
4146: if (j>i) {
1.203 brouard 4147: printf(".%d-%d",i,j);fflush(stdout);
4148: fprintf(ficlog,".%d-%d",i,j);fflush(ficlog);
4149: hess[i][j]=hessij(p,hess, delti,i,j,func,npar);
1.126 brouard 4150:
4151: hess[j][i]=hess[i][j];
4152: /*printf(" %lf ",hess[i][j]);*/
4153: }
4154: }
4155: }
4156: printf("\n");
4157: fprintf(ficlog,"\n");
4158:
4159: printf("\nInverting the hessian to get the covariance matrix. Wait...\n");
4160: fprintf(ficlog,"\nInverting the hessian to get the covariance matrix. Wait...\n");
4161:
4162: a=matrix(1,npar,1,npar);
4163: y=matrix(1,npar,1,npar);
4164: x=vector(1,npar);
4165: indx=ivector(1,npar);
4166: for (i=1;i<=npar;i++)
4167: for (j=1;j<=npar;j++) a[i][j]=hess[i][j];
4168: ludcmp(a,npar,indx,&pd);
4169:
4170: for (j=1;j<=npar;j++) {
4171: for (i=1;i<=npar;i++) x[i]=0;
4172: x[j]=1;
4173: lubksb(a,npar,indx,x);
4174: for (i=1;i<=npar;i++){
4175: matcov[i][j]=x[i];
4176: }
4177: }
4178:
4179: printf("\n#Hessian matrix#\n");
4180: fprintf(ficlog,"\n#Hessian matrix#\n");
4181: for (i=1;i<=npar;i++) {
4182: for (j=1;j<=npar;j++) {
1.203 brouard 4183: printf("%.6e ",hess[i][j]);
4184: fprintf(ficlog,"%.6e ",hess[i][j]);
1.126 brouard 4185: }
4186: printf("\n");
4187: fprintf(ficlog,"\n");
4188: }
4189:
1.203 brouard 4190: /* printf("\n#Covariance matrix#\n"); */
4191: /* fprintf(ficlog,"\n#Covariance matrix#\n"); */
4192: /* for (i=1;i<=npar;i++) { */
4193: /* for (j=1;j<=npar;j++) { */
4194: /* printf("%.6e ",matcov[i][j]); */
4195: /* fprintf(ficlog,"%.6e ",matcov[i][j]); */
4196: /* } */
4197: /* printf("\n"); */
4198: /* fprintf(ficlog,"\n"); */
4199: /* } */
4200:
1.126 brouard 4201: /* Recompute Inverse */
1.203 brouard 4202: /* for (i=1;i<=npar;i++) */
4203: /* for (j=1;j<=npar;j++) a[i][j]=matcov[i][j]; */
4204: /* ludcmp(a,npar,indx,&pd); */
4205:
4206: /* printf("\n#Hessian matrix recomputed#\n"); */
4207:
4208: /* for (j=1;j<=npar;j++) { */
4209: /* for (i=1;i<=npar;i++) x[i]=0; */
4210: /* x[j]=1; */
4211: /* lubksb(a,npar,indx,x); */
4212: /* for (i=1;i<=npar;i++){ */
4213: /* y[i][j]=x[i]; */
4214: /* printf("%.3e ",y[i][j]); */
4215: /* fprintf(ficlog,"%.3e ",y[i][j]); */
4216: /* } */
4217: /* printf("\n"); */
4218: /* fprintf(ficlog,"\n"); */
4219: /* } */
4220:
4221: /* Verifying the inverse matrix */
4222: #ifdef DEBUGHESS
4223: y=matprod2(y,hess,1,npar,1,npar,1,npar,matcov);
1.126 brouard 4224:
1.203 brouard 4225: printf("\n#Verification: multiplying the matrix of covariance by the Hessian matrix, should be unity:#\n");
4226: fprintf(ficlog,"\n#Verification: multiplying the matrix of covariance by the Hessian matrix. Should be unity:#\n");
1.126 brouard 4227:
4228: for (j=1;j<=npar;j++) {
4229: for (i=1;i<=npar;i++){
1.203 brouard 4230: printf("%.2f ",y[i][j]);
4231: fprintf(ficlog,"%.2f ",y[i][j]);
1.126 brouard 4232: }
4233: printf("\n");
4234: fprintf(ficlog,"\n");
4235: }
1.203 brouard 4236: #endif
1.126 brouard 4237:
4238: free_matrix(a,1,npar,1,npar);
4239: free_matrix(y,1,npar,1,npar);
4240: free_vector(x,1,npar);
4241: free_ivector(indx,1,npar);
1.203 brouard 4242: /* free_matrix(hess,1,npar,1,npar); */
1.126 brouard 4243:
4244:
4245: }
4246:
4247: /*************** hessian matrix ****************/
4248: double hessii(double x[], double delta, int theta, double delti[], double (*func)(double []), int npar)
1.203 brouard 4249: { /* Around values of x, computes the function func and returns the scales delti and hessian */
1.126 brouard 4250: int i;
4251: int l=1, lmax=20;
1.203 brouard 4252: double k1,k2, res, fx;
1.132 brouard 4253: double p2[MAXPARM+1]; /* identical to x */
1.126 brouard 4254: double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4;
4255: int k=0,kmax=10;
4256: double l1;
4257:
4258: fx=func(x);
4259: for (i=1;i<=npar;i++) p2[i]=x[i];
1.145 brouard 4260: for(l=0 ; l <=lmax; l++){ /* Enlarging the zone around the Maximum */
1.126 brouard 4261: l1=pow(10,l);
4262: delts=delt;
4263: for(k=1 ; k <kmax; k=k+1){
4264: delt = delta*(l1*k);
4265: p2[theta]=x[theta] +delt;
1.145 brouard 4266: k1=func(p2)-fx; /* Might be negative if too close to the theoretical maximum */
1.126 brouard 4267: p2[theta]=x[theta]-delt;
4268: k2=func(p2)-fx;
4269: /*res= (k1-2.0*fx+k2)/delt/delt; */
1.203 brouard 4270: res= (k1+k2)/delt/delt/2.; /* Divided by 2 because L and not 2*L */
1.126 brouard 4271:
1.203 brouard 4272: #ifdef DEBUGHESSII
1.126 brouard 4273: 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);
4274: 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);
4275: #endif
4276: /*if(fabs(k1-2.0*fx+k2) <1.e-13){ */
4277: if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)){
4278: k=kmax;
4279: }
4280: else if((k1 >khi/nkhif) || (k2 >khi/nkhif)){ /* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. */
1.164 brouard 4281: k=kmax; l=lmax*10;
1.126 brouard 4282: }
4283: else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){
4284: delts=delt;
4285: }
1.203 brouard 4286: } /* End loop k */
1.126 brouard 4287: }
4288: delti[theta]=delts;
4289: return res;
4290:
4291: }
4292:
1.203 brouard 4293: double hessij( double x[], double **hess, double delti[], int thetai,int thetaj,double (*func)(double []),int npar)
1.126 brouard 4294: {
4295: int i;
1.164 brouard 4296: int l=1, lmax=20;
1.126 brouard 4297: double k1,k2,k3,k4,res,fx;
1.132 brouard 4298: double p2[MAXPARM+1];
1.203 brouard 4299: int k, kmax=1;
4300: double v1, v2, cv12, lc1, lc2;
1.208 brouard 4301:
4302: int firstime=0;
1.203 brouard 4303:
1.126 brouard 4304: fx=func(x);
1.203 brouard 4305: for (k=1; k<=kmax; k=k+10) {
1.126 brouard 4306: for (i=1;i<=npar;i++) p2[i]=x[i];
1.203 brouard 4307: p2[thetai]=x[thetai]+delti[thetai]*k;
4308: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4309: k1=func(p2)-fx;
4310:
1.203 brouard 4311: p2[thetai]=x[thetai]+delti[thetai]*k;
4312: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4313: k2=func(p2)-fx;
4314:
1.203 brouard 4315: p2[thetai]=x[thetai]-delti[thetai]*k;
4316: p2[thetaj]=x[thetaj]+delti[thetaj]*k;
1.126 brouard 4317: k3=func(p2)-fx;
4318:
1.203 brouard 4319: p2[thetai]=x[thetai]-delti[thetai]*k;
4320: p2[thetaj]=x[thetaj]-delti[thetaj]*k;
1.126 brouard 4321: k4=func(p2)-fx;
1.203 brouard 4322: res=(k1-k2-k3+k4)/4.0/delti[thetai]/k/delti[thetaj]/k/2.; /* Because of L not 2*L */
4323: if(k1*k2*k3*k4 <0.){
1.208 brouard 4324: firstime=1;
1.203 brouard 4325: kmax=kmax+10;
1.208 brouard 4326: }
4327: if(kmax >=10 || firstime ==1){
1.246 brouard 4328: 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);
4329: 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 4330: 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);
4331: 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);
4332: }
4333: #ifdef DEBUGHESSIJ
4334: v1=hess[thetai][thetai];
4335: v2=hess[thetaj][thetaj];
4336: cv12=res;
4337: /* Computing eigen value of Hessian matrix */
4338: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4339: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
4340: if ((lc2 <0) || (lc1 <0) ){
4341: printf("Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4342: fprintf(ficlog, "Warning: sub Hessian matrix '%d%d' does not have positive eigen values \n",thetai,thetaj);
4343: 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);
4344: 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);
4345: }
1.126 brouard 4346: #endif
4347: }
4348: return res;
4349: }
4350:
1.203 brouard 4351: /* Not done yet: Was supposed to fix if not exactly at the maximum */
4352: /* double hessij( double x[], double delti[], int thetai,int thetaj,double (*func)(double []),int npar) */
4353: /* { */
4354: /* int i; */
4355: /* int l=1, lmax=20; */
4356: /* double k1,k2,k3,k4,res,fx; */
4357: /* double p2[MAXPARM+1]; */
4358: /* double delt=0.0001, delts, nkhi=10.,nkhif=1., khi=1.e-4; */
4359: /* int k=0,kmax=10; */
4360: /* double l1; */
4361:
4362: /* fx=func(x); */
4363: /* for(l=0 ; l <=lmax; l++){ /\* Enlarging the zone around the Maximum *\/ */
4364: /* l1=pow(10,l); */
4365: /* delts=delt; */
4366: /* for(k=1 ; k <kmax; k=k+1){ */
4367: /* delt = delti*(l1*k); */
4368: /* for (i=1;i<=npar;i++) p2[i]=x[i]; */
4369: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4370: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4371: /* k1=func(p2)-fx; */
4372:
4373: /* p2[thetai]=x[thetai]+delti[thetai]/k; */
4374: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4375: /* k2=func(p2)-fx; */
4376:
4377: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4378: /* p2[thetaj]=x[thetaj]+delti[thetaj]/k; */
4379: /* k3=func(p2)-fx; */
4380:
4381: /* p2[thetai]=x[thetai]-delti[thetai]/k; */
4382: /* p2[thetaj]=x[thetaj]-delti[thetaj]/k; */
4383: /* k4=func(p2)-fx; */
4384: /* res=(k1-k2-k3+k4)/4.0/delti[thetai]*k/delti[thetaj]*k/2.; /\* Because of L not 2*L *\/ */
4385: /* #ifdef DEBUGHESSIJ */
4386: /* 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); */
4387: /* 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); */
4388: /* #endif */
4389: /* if((k1 <khi/nkhi/2.) || (k2 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)|| (k4 <khi/nkhi/2.)){ */
4390: /* k=kmax; */
4391: /* } */
4392: /* else if((k1 >khi/nkhif) || (k2 >khi/nkhif) || (k4 >khi/nkhif) || (k4 >khi/nkhif)){ /\* Keeps lastvalue before 3.84/2 KHI2 5% 1d.f. *\/ */
4393: /* k=kmax; l=lmax*10; */
4394: /* } */
4395: /* else if((k1 >khi/nkhi) || (k2 >khi/nkhi)){ */
4396: /* delts=delt; */
4397: /* } */
4398: /* } /\* End loop k *\/ */
4399: /* } */
4400: /* delti[theta]=delts; */
4401: /* return res; */
4402: /* } */
4403:
4404:
1.126 brouard 4405: /************** Inverse of matrix **************/
4406: void ludcmp(double **a, int n, int *indx, double *d)
4407: {
4408: int i,imax,j,k;
4409: double big,dum,sum,temp;
4410: double *vv;
4411:
4412: vv=vector(1,n);
4413: *d=1.0;
4414: for (i=1;i<=n;i++) {
4415: big=0.0;
4416: for (j=1;j<=n;j++)
4417: if ((temp=fabs(a[i][j])) > big) big=temp;
1.256 brouard 4418: if (big == 0.0){
4419: printf(" Singular Hessian matrix at row %d:\n",i);
4420: for (j=1;j<=n;j++) {
4421: printf(" a[%d][%d]=%f,",i,j,a[i][j]);
4422: fprintf(ficlog," a[%d][%d]=%f,",i,j,a[i][j]);
4423: }
4424: fflush(ficlog);
4425: fclose(ficlog);
4426: nrerror("Singular matrix in routine ludcmp");
4427: }
1.126 brouard 4428: vv[i]=1.0/big;
4429: }
4430: for (j=1;j<=n;j++) {
4431: for (i=1;i<j;i++) {
4432: sum=a[i][j];
4433: for (k=1;k<i;k++) sum -= a[i][k]*a[k][j];
4434: a[i][j]=sum;
4435: }
4436: big=0.0;
4437: for (i=j;i<=n;i++) {
4438: sum=a[i][j];
4439: for (k=1;k<j;k++)
4440: sum -= a[i][k]*a[k][j];
4441: a[i][j]=sum;
4442: if ( (dum=vv[i]*fabs(sum)) >= big) {
4443: big=dum;
4444: imax=i;
4445: }
4446: }
4447: if (j != imax) {
4448: for (k=1;k<=n;k++) {
4449: dum=a[imax][k];
4450: a[imax][k]=a[j][k];
4451: a[j][k]=dum;
4452: }
4453: *d = -(*d);
4454: vv[imax]=vv[j];
4455: }
4456: indx[j]=imax;
4457: if (a[j][j] == 0.0) a[j][j]=TINY;
4458: if (j != n) {
4459: dum=1.0/(a[j][j]);
4460: for (i=j+1;i<=n;i++) a[i][j] *= dum;
4461: }
4462: }
4463: free_vector(vv,1,n); /* Doesn't work */
4464: ;
4465: }
4466:
4467: void lubksb(double **a, int n, int *indx, double b[])
4468: {
4469: int i,ii=0,ip,j;
4470: double sum;
4471:
4472: for (i=1;i<=n;i++) {
4473: ip=indx[i];
4474: sum=b[ip];
4475: b[ip]=b[i];
4476: if (ii)
4477: for (j=ii;j<=i-1;j++) sum -= a[i][j]*b[j];
4478: else if (sum) ii=i;
4479: b[i]=sum;
4480: }
4481: for (i=n;i>=1;i--) {
4482: sum=b[i];
4483: for (j=i+1;j<=n;j++) sum -= a[i][j]*b[j];
4484: b[i]=sum/a[i][i];
4485: }
4486: }
4487:
4488: void pstamp(FILE *fichier)
4489: {
1.196 brouard 4490: fprintf(fichier,"# %s.%s\n#IMaCh version %s, %s\n#%s\n# %s", optionfilefiname,optionfilext,version,copyright, fullversion, strstart);
1.126 brouard 4491: }
4492:
1.297 brouard 4493: void date2dmy(double date,double *day, double *month, double *year){
4494: double yp=0., yp1=0., yp2=0.;
4495:
4496: yp1=modf(date,&yp);/* extracts integral of date in yp and
4497: fractional in yp1 */
4498: *year=yp;
4499: yp2=modf((yp1*12),&yp);
4500: *month=yp;
4501: yp1=modf((yp2*30.5),&yp);
4502: *day=yp;
4503: if(*day==0) *day=1;
4504: if(*month==0) *month=1;
4505: }
4506:
1.253 brouard 4507:
4508:
1.126 brouard 4509: /************ Frequencies ********************/
1.251 brouard 4510: void freqsummary(char fileres[], double p[], double pstart[], int iagemin, int iagemax, int **s, double **agev, int nlstate, int imx, \
1.226 brouard 4511: int *Tvaraff, int *invalidvarcomb, int **nbcode, int *ncodemax,double **mint,double **anint, char strstart[], \
4512: int firstpass, int lastpass, int stepm, int weightopt, char model[])
1.250 brouard 4513: { /* Some frequencies as well as proposing some starting values */
1.226 brouard 4514:
1.265 brouard 4515: int i, m, jk, j1, bool, z1,j, nj, nl, k, iv, jj=0, s1=1, s2=1;
1.226 brouard 4516: int iind=0, iage=0;
4517: int mi; /* Effective wave */
4518: int first;
4519: double ***freq; /* Frequencies */
1.268 brouard 4520: 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 */
4521: 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 4522: double *meanq, *stdq, *idq;
1.226 brouard 4523: double **meanqt;
4524: double *pp, **prop, *posprop, *pospropt;
4525: double pos=0., posproptt=0., pospropta=0., k2, dateintsum=0,k2cpt=0;
4526: char fileresp[FILENAMELENGTH], fileresphtm[FILENAMELENGTH], fileresphtmfr[FILENAMELENGTH];
4527: double agebegin, ageend;
4528:
4529: pp=vector(1,nlstate);
1.251 brouard 4530: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4531: posprop=vector(1,nlstate); /* Counting the number of transition starting from a live state per age */
4532: pospropt=vector(1,nlstate); /* Counting the number of transition starting from a live state */
4533: /* prop=matrix(1,nlstate,iagemin,iagemax+3); */
4534: meanq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.284 brouard 4535: stdq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.283 brouard 4536: idq=vector(1,nqfveff); /* Number of Quantitative Fixed Variables Effective */
1.226 brouard 4537: meanqt=matrix(1,lastpass,1,nqtveff);
4538: strcpy(fileresp,"P_");
4539: strcat(fileresp,fileresu);
4540: /*strcat(fileresphtm,fileresu);*/
4541: if((ficresp=fopen(fileresp,"w"))==NULL) {
4542: printf("Problem with prevalence resultfile: %s\n", fileresp);
4543: fprintf(ficlog,"Problem with prevalence resultfile: %s\n", fileresp);
4544: exit(0);
4545: }
1.240 brouard 4546:
1.226 brouard 4547: strcpy(fileresphtm,subdirfext(optionfilefiname,"PHTM_",".htm"));
4548: if((ficresphtm=fopen(fileresphtm,"w"))==NULL) {
4549: printf("Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4550: fprintf(ficlog,"Problem with prevalence HTM resultfile '%s' with errno='%s'\n",fileresphtm,strerror(errno));
4551: fflush(ficlog);
4552: exit(70);
4553: }
4554: else{
4555: fprintf(ficresphtm,"<html><head>\n<title>IMaCh PHTM_ %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4556: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4557: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4558: fileresphtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4559: }
1.237 brouard 4560: fprintf(ficresphtm,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>Frequencies and prevalence by age at begin of transition and dummy covariate value at beginning of transition</h4>\n",fileresphtm, fileresphtm);
1.240 brouard 4561:
1.226 brouard 4562: strcpy(fileresphtmfr,subdirfext(optionfilefiname,"PHTMFR_",".htm"));
4563: if((ficresphtmfr=fopen(fileresphtmfr,"w"))==NULL) {
4564: printf("Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4565: fprintf(ficlog,"Problem with frequency table HTM resultfile '%s' with errno='%s'\n",fileresphtmfr,strerror(errno));
4566: fflush(ficlog);
4567: exit(70);
1.240 brouard 4568: } else{
1.226 brouard 4569: fprintf(ficresphtmfr,"<html><head>\n<title>IMaCh PHTM_Frequency table %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
1.240 brouard 4570: <hr size=\"2\" color=\"#EC5E5E\"> \n \
1.214 brouard 4571: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.226 brouard 4572: fileresphtmfr,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
4573: }
1.240 brouard 4574: fprintf(ficresphtmfr,"Current page is file <a href=\"%s\">%s</a><br>\n\n<h4>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);
4575:
1.253 brouard 4576: y= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
4577: x= vector(iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.251 brouard 4578: freq= ma3x(-5,nlstate+ndeath,-5,nlstate+ndeath,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.226 brouard 4579: j1=0;
1.126 brouard 4580:
1.227 brouard 4581: /* j=ncoveff; /\* Only fixed dummy covariates *\/ */
4582: j=cptcoveff; /* Only dummy covariates of the model */
1.226 brouard 4583: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.240 brouard 4584:
4585:
1.226 brouard 4586: /* Detects if a combination j1 is empty: for a multinomial variable like 3 education levels:
4587: reference=low_education V1=0,V2=0
4588: med_educ V1=1 V2=0,
4589: high_educ V1=0 V2=1
4590: Then V1=1 and V2=1 is a noisy combination that we want to exclude for the list 2**cptcoveff
4591: */
1.249 brouard 4592: dateintsum=0;
4593: k2cpt=0;
4594:
1.253 brouard 4595: if(cptcoveff == 0 )
1.265 brouard 4596: nl=1; /* Constant and age model only */
1.253 brouard 4597: else
4598: nl=2;
1.265 brouard 4599:
4600: /* if a constant only model, one pass to compute frequency tables and to write it on ficresp */
4601: /* Loop on nj=1 or 2 if dummy covariates j!=0
4602: * Loop on j1(1 to 2**cptcoveff) covariate combination
4603: * freq[s1][s2][iage] =0.
4604: * Loop on iind
4605: * ++freq[s1][s2][iage] weighted
4606: * end iind
4607: * if covariate and j!0
4608: * headers Variable on one line
4609: * endif cov j!=0
4610: * header of frequency table by age
4611: * Loop on age
4612: * pp[s1]+=freq[s1][s2][iage] weighted
4613: * pos+=freq[s1][s2][iage] weighted
4614: * Loop on s1 initial state
4615: * fprintf(ficresp
4616: * end s1
4617: * end age
4618: * if j!=0 computes starting values
4619: * end compute starting values
4620: * end j1
4621: * end nl
4622: */
1.253 brouard 4623: for (nj = 1; nj <= nl; nj++){ /* nj= 1 constant model, nl number of loops. */
4624: if(nj==1)
4625: j=0; /* First pass for the constant */
1.265 brouard 4626: else{
1.253 brouard 4627: j=cptcoveff; /* Other passes for the covariate values */
1.265 brouard 4628: }
1.251 brouard 4629: first=1;
1.265 brouard 4630: for (j1 = 1; j1 <= (int) pow(2,j); j1++){ /* Loop on all covariates combination of the model, excluding quantitatives, V4=0, V3=0 for example, fixed or varying covariates */
1.251 brouard 4631: posproptt=0.;
4632: /*printf("cptcoveff=%d Tvaraff=%d", cptcoveff,Tvaraff[1]);
4633: scanf("%d", i);*/
4634: for (i=-5; i<=nlstate+ndeath; i++)
1.265 brouard 4635: for (s2=-5; s2<=nlstate+ndeath; s2++)
1.251 brouard 4636: for(m=iagemin; m <= iagemax+3; m++)
1.265 brouard 4637: freq[i][s2][m]=0;
1.251 brouard 4638:
4639: for (i=1; i<=nlstate; i++) {
1.240 brouard 4640: for(m=iagemin; m <= iagemax+3; m++)
1.251 brouard 4641: prop[i][m]=0;
4642: posprop[i]=0;
4643: pospropt[i]=0;
4644: }
1.283 brouard 4645: for (z1=1; z1<= nqfveff; z1++) { /* zeroing for each combination j1 as well as for the total */
1.284 brouard 4646: idq[z1]=0.;
4647: meanq[z1]=0.;
4648: stdq[z1]=0.;
1.283 brouard 4649: }
4650: /* for (z1=1; z1<= nqtveff; z1++) { */
1.251 brouard 4651: /* for(m=1;m<=lastpass;m++){ */
1.283 brouard 4652: /* meanqt[m][z1]=0.; */
4653: /* } */
4654: /* } */
1.251 brouard 4655: /* dateintsum=0; */
4656: /* k2cpt=0; */
4657:
1.265 brouard 4658: /* For that combination of covariates j1 (V4=1 V3=0 for example), we count and print the frequencies in one pass */
1.251 brouard 4659: for (iind=1; iind<=imx; iind++) { /* For each individual iind */
4660: bool=1;
4661: if(j !=0){
4662: if(anyvaryingduminmodel==0){ /* If All fixed covariates */
4663: if (cptcoveff >0) { /* Filter is here: Must be looked at for model=V1+V2+V3+V4 */
4664: for (z1=1; z1<=cptcoveff; z1++) { /* loops on covariates in the model */
4665: /* if(Tvaraff[z1] ==-20){ */
4666: /* /\* sumnew+=cotvar[mw[mi][iind]][z1][iind]; *\/ */
4667: /* }else if(Tvaraff[z1] ==-10){ */
4668: /* /\* sumnew+=coqvar[z1][iind]; *\/ */
4669: /* }else */
4670: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]){ /* for combination j1 of covariates */
1.265 brouard 4671: /* Tests if the value of the covariate z1 for this individual iind responded to combination j1 (V4=1 V3=0) */
1.251 brouard 4672: bool=0; /* bool should be equal to 1 to be selected, one covariate value failed */
4673: /* 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",
4674: bool,i,z1, z1, Tvaraff[z1],i,covar[Tvaraff[z1]][i],j1,z1,codtabm(j1,z1),
4675: j1,z1,nbcode[Tvaraff[z1]][codtabm(j1,z1)],j1);*/
4676: /* For j1=7 in V1+V2+V3+V4 = 0 1 1 0 and codtabm(7,3)=1 and nbcde[3][?]=1*/
4677: } /* Onlyf fixed */
4678: } /* end z1 */
4679: } /* cptcovn > 0 */
4680: } /* end any */
4681: }/* end j==0 */
1.265 brouard 4682: if (bool==1){ /* We selected an individual iind satisfying combination j1 (V4=1 V3=0) or all fixed covariates */
1.251 brouard 4683: /* for(m=firstpass; m<=lastpass; m++){ */
1.284 brouard 4684: for(mi=1; mi<wav[iind];mi++){ /* For each wave */
1.251 brouard 4685: m=mw[mi][iind];
4686: if(j!=0){
4687: if(anyvaryingduminmodel==1){ /* Some are varying covariates */
4688: for (z1=1; z1<=cptcoveff; z1++) {
4689: if( Fixed[Tmodelind[z1]]==1){
4690: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
4691: if (cotvar[m][iv][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality. If covariate's
4692: value is -1, we don't select. It differs from the
4693: constant and age model which counts them. */
4694: bool=0; /* not selected */
4695: }else if( Fixed[Tmodelind[z1]]== 0) { /* fixed */
4696: if (covar[Tvaraff[z1]][iind]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
4697: bool=0;
4698: }
4699: }
4700: }
4701: }/* Some are varying covariates, we tried to speed up if all fixed covariates in the model, avoiding waves loop */
4702: } /* end j==0 */
4703: /* bool =0 we keep that guy which corresponds to the combination of dummy values */
1.284 brouard 4704: if(bool==1){ /*Selected */
1.251 brouard 4705: /* dh[m][iind] or dh[mw[mi][iind]][iind] is the delay between two effective (mi) waves m=mw[mi][iind]
4706: and mw[mi+1][iind]. dh depends on stepm. */
4707: agebegin=agev[m][iind]; /* Age at beginning of wave before transition*/
4708: ageend=agev[m][iind]+(dh[m][iind])*stepm/YEARM; /* Age at end of wave and transition */
4709: if(m >=firstpass && m <=lastpass){
4710: k2=anint[m][iind]+(mint[m][iind]/12.);
4711: /*if ((k2>=dateprev1) && (k2<=dateprev2)) {*/
4712: if(agev[m][iind]==0) agev[m][iind]=iagemax+1; /* All ages equal to 0 are in iagemax+1 */
4713: if(agev[m][iind]==1) agev[m][iind]=iagemax+2; /* All ages equal to 1 are in iagemax+2 */
4714: if (s[m][iind]>0 && s[m][iind]<=nlstate) /* If status at wave m is known and a live state */
4715: prop[s[m][iind]][(int)agev[m][iind]] += weight[iind]; /* At age of beginning of transition, where status is known */
4716: if (m<lastpass) {
4717: /* if(s[m][iind]==4 && s[m+1][iind]==4) */
4718: /* 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]); */
4719: if(s[m][iind]==-1)
4720: 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.));
4721: 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 4722: for (z1=1; z1<= nqfveff; z1++) { /* Quantitative variables, calculating mean on known values only */
4723: if(!isnan(covar[ncovcol+z1][iind])){
4724: idq[z1]=idq[z1]+weight[iind];
4725: meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /* Computes mean of quantitative with selected filter */
4726: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; *//*error*/
4727: stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]; /* *weight[iind];*/ /* Computes mean of quantitative with selected filter */
4728: }
1.284 brouard 4729: }
1.251 brouard 4730: /* if((int)agev[m][iind] == 55) */
4731: /* printf("j=%d, j1=%d Age %d, iind=%d, num=%09ld m=%d\n",j,j1,(int)agev[m][iind],iind, num[iind],m); */
4732: /* freq[s[m][iind]][s[m+1][iind]][(int)((agebegin+ageend)/2.)] += weight[iind]; */
4733: 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 4734: }
1.251 brouard 4735: } /* end if between passes */
4736: if ((agev[m][iind]>1) && (agev[m][iind]< (iagemax+3)) && (anint[m][iind]!=9999) && (mint[m][iind]!=99) && (j==0)) {
4737: dateintsum=dateintsum+k2; /* on all covariates ?*/
4738: k2cpt++;
4739: /* printf("iind=%ld dateintmean = %lf dateintsum=%lf k2cpt=%lf k2=%lf\n",iind, dateintsum/k2cpt, dateintsum,k2cpt, k2); */
1.234 brouard 4740: }
1.251 brouard 4741: }else{
4742: bool=1;
4743: }/* end bool 2 */
4744: } /* end m */
1.284 brouard 4745: /* for (z1=1; z1<= nqfveff; z1++) { /\* Quantitative variables, calculating mean *\/ */
4746: /* idq[z1]=idq[z1]+weight[iind]; */
4747: /* meanq[z1]+=covar[ncovcol+z1][iind]*weight[iind]; /\* Computes mean of quantitative with selected filter *\/ */
4748: /* stdq[z1]+=covar[ncovcol+z1][iind]*covar[ncovcol+z1][iind]*weight[iind]*weight[iind]; /\* *weight[iind];*\/ /\* Computes mean of quantitative with selected filter *\/ */
4749: /* } */
1.251 brouard 4750: } /* end bool */
4751: } /* end iind = 1 to imx */
4752: /* prop[s][age] is feeded for any initial and valid live state as well as
4753: freq[s1][s2][age] at single age of beginning the transition, for a combination j1 */
4754:
4755:
4756: /* fprintf(ficresp, "#Count between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2);*/
1.265 brouard 4757: if(cptcoveff==0 && nj==1) /* no covariate and first pass */
4758: pstamp(ficresp);
1.251 brouard 4759: if (cptcoveff>0 && j!=0){
1.265 brouard 4760: pstamp(ficresp);
1.251 brouard 4761: printf( "\n#********** Variable ");
4762: fprintf(ficresp, "\n#********** Variable ");
4763: fprintf(ficresphtm, "\n<br/><br/><h3>********** Variable ");
4764: fprintf(ficresphtmfr, "\n<br/><br/><h3>********** Variable ");
4765: fprintf(ficlog, "\n#********** Variable ");
4766: for (z1=1; z1<=cptcoveff; z1++){
4767: if(!FixedV[Tvaraff[z1]]){
4768: printf( "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4769: fprintf(ficresp, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4770: fprintf(ficresphtm, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4771: fprintf(ficresphtmfr, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4772: fprintf(ficlog, "V%d(fixed)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.250 brouard 4773: }else{
1.251 brouard 4774: printf( "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4775: fprintf(ficresp, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4776: fprintf(ficresphtm, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4777: fprintf(ficresphtmfr, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4778: fprintf(ficlog, "V%d(varying)=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4779: }
4780: }
4781: printf( "**********\n#");
4782: fprintf(ficresp, "**********\n#");
4783: fprintf(ficresphtm, "**********</h3>\n");
4784: fprintf(ficresphtmfr, "**********</h3>\n");
4785: fprintf(ficlog, "**********\n");
4786: }
1.284 brouard 4787: /*
4788: Printing means of quantitative variables if any
4789: */
4790: for (z1=1; z1<= nqfveff; z1++) {
1.311 brouard 4791: fprintf(ficlog,"Mean of fixed quantitative variable V%d on %.3g (weighted) individuals sum=%f", ncovcol+z1, idq[z1], meanq[z1]);
1.312 brouard 4792: fprintf(ficlog,", mean=%.3g\n",meanq[z1]/idq[z1]);
1.284 brouard 4793: if(weightopt==1){
4794: printf(" Weighted mean and standard deviation of");
4795: fprintf(ficlog," Weighted mean and standard deviation of");
4796: fprintf(ficresphtmfr," Weighted mean and standard deviation of");
4797: }
1.311 brouard 4798: /* mu = \frac{w x}{\sum w}
4799: var = \frac{\sum w (x-mu)^2}{\sum w} = \frac{w x^2}{\sum w} - mu^2
4800: */
4801: 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]));
4802: 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]));
4803: 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 4804: }
4805: /* for (z1=1; z1<= nqtveff; z1++) { */
4806: /* for(m=1;m<=lastpass;m++){ */
4807: /* fprintf(ficresphtmfr,"V quantitative id %d, pass id=%d, mean=%f<p>\n", z1, m, meanqt[m][z1]); */
4808: /* } */
4809: /* } */
1.283 brouard 4810:
1.251 brouard 4811: fprintf(ficresphtm,"<table style=\"text-align:center; border: 1px solid\">");
1.265 brouard 4812: if((cptcoveff==0 && nj==1)|| nj==2 ) /* no covariate and first pass */
4813: fprintf(ficresp, " Age");
4814: if(nj==2) for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " V%d=%d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.251 brouard 4815: for(i=1; i<=nlstate;i++) {
1.265 brouard 4816: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," Prev(%d) N(%d) N ",i,i);
1.251 brouard 4817: fprintf(ficresphtm, "<th>Age</th><th>Prev(%d)</th><th>N(%d)</th><th>N</th>",i,i);
4818: }
1.265 brouard 4819: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp, "\n");
1.251 brouard 4820: fprintf(ficresphtm, "\n");
4821:
4822: /* Header of frequency table by age */
4823: fprintf(ficresphtmfr,"<table style=\"text-align:center; border: 1px solid\">");
4824: fprintf(ficresphtmfr,"<th>Age</th> ");
1.265 brouard 4825: for(s2=-1; s2 <=nlstate+ndeath; s2++){
1.251 brouard 4826: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4827: if(s2!=0 && m!=0)
4828: fprintf(ficresphtmfr,"<th>%d%d</th> ",s2,m);
1.240 brouard 4829: }
1.226 brouard 4830: }
1.251 brouard 4831: fprintf(ficresphtmfr, "\n");
4832:
4833: /* For each age */
4834: for(iage=iagemin; iage <= iagemax+3; iage++){
4835: fprintf(ficresphtm,"<tr>");
4836: if(iage==iagemax+1){
4837: fprintf(ficlog,"1");
4838: fprintf(ficresphtmfr,"<tr><th>0</th> ");
4839: }else if(iage==iagemax+2){
4840: fprintf(ficlog,"0");
4841: fprintf(ficresphtmfr,"<tr><th>Unknown</th> ");
4842: }else if(iage==iagemax+3){
4843: fprintf(ficlog,"Total");
4844: fprintf(ficresphtmfr,"<tr><th>Total</th> ");
4845: }else{
1.240 brouard 4846: if(first==1){
1.251 brouard 4847: first=0;
4848: printf("See log file for details...\n");
4849: }
4850: fprintf(ficresphtmfr,"<tr><th>%d</th> ",iage);
4851: fprintf(ficlog,"Age %d", iage);
4852: }
1.265 brouard 4853: for(s1=1; s1 <=nlstate ; s1++){
4854: for(m=-1, pp[s1]=0; m <=nlstate+ndeath ; m++)
4855: pp[s1] += freq[s1][m][iage];
1.251 brouard 4856: }
1.265 brouard 4857: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4858: for(m=-1, pos=0; m <=0 ; m++)
1.265 brouard 4859: pos += freq[s1][m][iage];
4860: if(pp[s1]>=1.e-10){
1.251 brouard 4861: if(first==1){
1.265 brouard 4862: printf(" %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4863: }
1.265 brouard 4864: fprintf(ficlog," %d.=%.0f loss[%d]=%.1f%%",s1,pp[s1],s1,100*pos/pp[s1]);
1.251 brouard 4865: }else{
4866: if(first==1)
1.265 brouard 4867: printf(" %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
4868: fprintf(ficlog," %d.=%.0f loss[%d]=NaNQ%%",s1,pp[s1],s1);
1.240 brouard 4869: }
4870: }
4871:
1.265 brouard 4872: for(s1=1; s1 <=nlstate ; s1++){
4873: /* posprop[s1]=0; */
4874: for(m=0, pp[s1]=0; m <=nlstate+ndeath; m++)/* Summing on all ages */
4875: pp[s1] += freq[s1][m][iage];
4876: } /* pp[s1] is the total number of transitions starting from state s1 and any ending status until this age */
4877:
4878: for(s1=1,pos=0, pospropta=0.; s1 <=nlstate ; s1++){
4879: pos += pp[s1]; /* pos is the total number of transitions until this age */
4880: posprop[s1] += prop[s1][iage]; /* prop is the number of transitions from a live state
4881: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4882: pospropta += prop[s1][iage]; /* prop is the number of transitions from a live state
4883: from s1 at age iage prop[s[m][iind]][(int)agev[m][iind]] += weight[iind] */
4884: }
4885:
4886: /* Writing ficresp */
4887: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4888: if( iage <= iagemax){
4889: fprintf(ficresp," %d",iage);
4890: }
4891: }else if( nj==2){
4892: if( iage <= iagemax){
4893: fprintf(ficresp," %d",iage);
4894: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresp, " %d %d",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
4895: }
1.240 brouard 4896: }
1.265 brouard 4897: for(s1=1; s1 <=nlstate ; s1++){
1.240 brouard 4898: if(pos>=1.e-5){
1.251 brouard 4899: if(first==1)
1.265 brouard 4900: printf(" %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
4901: fprintf(ficlog," %d.=%.0f prev[%d]=%.1f%%",s1,pp[s1],s1,100*pp[s1]/pos);
1.251 brouard 4902: }else{
4903: if(first==1)
1.265 brouard 4904: printf(" %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
4905: fprintf(ficlog," %d.=%.0f prev[%d]=NaNQ%%",s1,pp[s1],s1);
1.251 brouard 4906: }
4907: if( iage <= iagemax){
4908: if(pos>=1.e-5){
1.265 brouard 4909: if(cptcoveff==0 && nj==1){ /* no covariate and first pass */
4910: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4911: }else if( nj==2){
4912: fprintf(ficresp," %.5f %.0f %.0f",prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4913: }
4914: fprintf(ficresphtm,"<th>%d</th><td>%.5f</td><td>%.0f</td><td>%.0f</td>",iage,prop[s1][iage]/pospropta, prop[s1][iage],pospropta);
4915: /*probs[iage][s1][j1]= pp[s1]/pos;*/
4916: /*printf("\niage=%d s1=%d j1=%d %.5f %.0f %.0f %f",iage,s1,j1,pp[s1]/pos, pp[s1],pos,probs[iage][s1][j1]);*/
4917: } else{
4918: if((cptcoveff==0 && nj==1)|| nj==2 ) fprintf(ficresp," NaNq %.0f %.0f",prop[s1][iage],pospropta);
4919: fprintf(ficresphtm,"<th>%d</th><td>NaNq</td><td>%.0f</td><td>%.0f</td>",iage, prop[s1][iage],pospropta);
1.251 brouard 4920: }
1.240 brouard 4921: }
1.265 brouard 4922: pospropt[s1] +=posprop[s1];
4923: } /* end loop s1 */
1.251 brouard 4924: /* pospropt=0.; */
1.265 brouard 4925: for(s1=-1; s1 <=nlstate+ndeath; s1++){
1.251 brouard 4926: for(m=-1; m <=nlstate+ndeath; m++){
1.265 brouard 4927: if(freq[s1][m][iage] !=0 ) { /* minimizing output */
1.251 brouard 4928: if(first==1){
1.265 brouard 4929: printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4930: }
1.265 brouard 4931: /* printf(" %d%d=%.0f",s1,m,freq[s1][m][iage]); */
4932: fprintf(ficlog," %d%d=%.0f",s1,m,freq[s1][m][iage]);
1.251 brouard 4933: }
1.265 brouard 4934: if(s1!=0 && m!=0)
4935: fprintf(ficresphtmfr,"<td>%.0f</td> ",freq[s1][m][iage]);
1.240 brouard 4936: }
1.265 brouard 4937: } /* end loop s1 */
1.251 brouard 4938: posproptt=0.;
1.265 brouard 4939: for(s1=1; s1 <=nlstate; s1++){
4940: posproptt += pospropt[s1];
1.251 brouard 4941: }
4942: fprintf(ficresphtmfr,"</tr>\n ");
1.265 brouard 4943: fprintf(ficresphtm,"</tr>\n");
4944: if((cptcoveff==0 && nj==1)|| nj==2 ) {
4945: if(iage <= iagemax)
4946: fprintf(ficresp,"\n");
1.240 brouard 4947: }
1.251 brouard 4948: if(first==1)
4949: printf("Others in log...\n");
4950: fprintf(ficlog,"\n");
4951: } /* end loop age iage */
1.265 brouard 4952:
1.251 brouard 4953: fprintf(ficresphtm,"<tr><th>Tot</th>");
1.265 brouard 4954: for(s1=1; s1 <=nlstate ; s1++){
1.251 brouard 4955: if(posproptt < 1.e-5){
1.265 brouard 4956: fprintf(ficresphtm,"<td>Nanq</td><td>%.0f</td><td>%.0f</td>",pospropt[s1],posproptt);
1.251 brouard 4957: }else{
1.265 brouard 4958: fprintf(ficresphtm,"<td>%.5f</td><td>%.0f</td><td>%.0f</td>",pospropt[s1]/posproptt,pospropt[s1],posproptt);
1.240 brouard 4959: }
1.226 brouard 4960: }
1.251 brouard 4961: fprintf(ficresphtm,"</tr>\n");
4962: fprintf(ficresphtm,"</table>\n");
4963: fprintf(ficresphtmfr,"</table>\n");
1.226 brouard 4964: if(posproptt < 1.e-5){
1.251 brouard 4965: fprintf(ficresphtm,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
4966: fprintf(ficresphtmfr,"\n <p><b> This combination (%d) is not valid and no result will be produced</b></p>",j1);
1.260 brouard 4967: fprintf(ficlog,"# This combination (%d) is not valid and no result will be produced\n",j1);
4968: printf("# This combination (%d) is not valid and no result will be produced\n",j1);
1.251 brouard 4969: invalidvarcomb[j1]=1;
1.226 brouard 4970: }else{
1.251 brouard 4971: fprintf(ficresphtm,"\n <p> This combination (%d) is valid and result will be produced.</p>",j1);
4972: invalidvarcomb[j1]=0;
1.226 brouard 4973: }
1.251 brouard 4974: fprintf(ficresphtmfr,"</table>\n");
4975: fprintf(ficlog,"\n");
4976: if(j!=0){
4977: printf("#Freqsummary: Starting values for combination j1=%d:\n", j1);
1.265 brouard 4978: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 4979: for(k=1; k <=(nlstate+ndeath); k++){
4980: if (k != i) {
1.265 brouard 4981: for(jj=1; jj <=ncovmodel; jj++){ /* For counting s1 */
1.253 brouard 4982: if(jj==1){ /* Constant case (in fact cste + age) */
1.251 brouard 4983: if(j1==1){ /* All dummy covariates to zero */
4984: freq[i][k][iagemax+4]=freq[i][k][iagemax+3]; /* Stores case 0 0 0 */
4985: freq[i][i][iagemax+4]=freq[i][i][iagemax+3]; /* Stores case 0 0 0 */
1.252 brouard 4986: printf("%d%d ",i,k);
4987: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 4988: 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]));
4989: 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]));
4990: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
1.251 brouard 4991: }
1.253 brouard 4992: }else if((j1==1) && (jj==2 || nagesqr==1)){ /* age or age*age parameter without covariate V4*age (to be done later) */
4993: for(iage=iagemin; iage <= iagemax+3; iage++){
4994: x[iage]= (double)iage;
4995: y[iage]= log(freq[i][k][iage]/freq[i][i][iage]);
1.265 brouard 4996: /* 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 4997: }
1.268 brouard 4998: /* Some are not finite, but linreg will ignore these ages */
4999: no=0;
1.253 brouard 5000: linreg(iagemin,iagemax,&no,x,y,&a,&b,&r, &sa, &sb ); /* y= a+b*x with standard errors */
1.265 brouard 5001: pstart[s1]=b;
5002: pstart[s1-1]=a;
1.252 brouard 5003: }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 */
5004: 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]);
5005: 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 5006: 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 5007: printf("%d%d ",i,k);
5008: fprintf(ficlog,"%d%d ",i,k);
1.265 brouard 5009: 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 5010: }else{ /* Other cases, like quantitative fixed or varying covariates */
5011: ;
5012: }
5013: /* printf("%12.7f )", param[i][jj][k]); */
5014: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5015: s1++;
1.251 brouard 5016: } /* end jj */
5017: } /* end k!= i */
5018: } /* end k */
1.265 brouard 5019: } /* end i, s1 */
1.251 brouard 5020: } /* end j !=0 */
5021: } /* end selected combination of covariate j1 */
5022: if(j==0){ /* We can estimate starting values from the occurences in each case */
5023: printf("#Freqsummary: Starting values for the constants:\n");
5024: fprintf(ficlog,"\n");
1.265 brouard 5025: for(i=1,s1=1; i <=nlstate; i++){
1.251 brouard 5026: for(k=1; k <=(nlstate+ndeath); k++){
5027: if (k != i) {
5028: printf("%d%d ",i,k);
5029: fprintf(ficlog,"%d%d ",i,k);
5030: for(jj=1; jj <=ncovmodel; jj++){
1.265 brouard 5031: pstart[s1]=p[s1]; /* Setting pstart to p values by default */
1.253 brouard 5032: if(jj==1){ /* Age has to be done */
1.265 brouard 5033: pstart[s1]= log(freq[i][k][iagemax+3]/freq[i][i][iagemax+3]);
5034: 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]));
5035: 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 5036: }
5037: /* printf("%12.7f )", param[i][jj][k]); */
5038: /* fprintf(ficlog,"%12.7f )", param[i][jj][k]); */
1.265 brouard 5039: s1++;
1.250 brouard 5040: }
1.251 brouard 5041: printf("\n");
5042: fprintf(ficlog,"\n");
1.250 brouard 5043: }
5044: }
1.284 brouard 5045: } /* end of state i */
1.251 brouard 5046: printf("#Freqsummary\n");
5047: fprintf(ficlog,"\n");
1.265 brouard 5048: for(s1=-1; s1 <=nlstate+ndeath; s1++){
5049: for(s2=-1; s2 <=nlstate+ndeath; s2++){
5050: /* param[i]|j][k]= freq[s1][s2][iagemax+3] */
5051: printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5052: fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]);
5053: /* if(freq[s1][s2][iage] !=0 ) { /\* minimizing output *\/ */
5054: /* printf(" %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
5055: /* fprintf(ficlog," %d%d=%.0f",s1,s2,freq[s1][s2][iagemax+3]); */
1.251 brouard 5056: /* } */
5057: }
1.265 brouard 5058: } /* end loop s1 */
1.251 brouard 5059:
5060: printf("\n");
5061: fprintf(ficlog,"\n");
5062: } /* end j=0 */
1.249 brouard 5063: } /* end j */
1.252 brouard 5064:
1.253 brouard 5065: if(mle == -2){ /* We want to use these values as starting values */
1.252 brouard 5066: for(i=1, jk=1; i <=nlstate; i++){
5067: for(j=1; j <=nlstate+ndeath; j++){
5068: if(j!=i){
5069: /*ca[0]= k+'a'-1;ca[1]='\0';*/
5070: printf("%1d%1d",i,j);
5071: fprintf(ficparo,"%1d%1d",i,j);
5072: for(k=1; k<=ncovmodel;k++){
5073: /* printf(" %lf",param[i][j][k]); */
5074: /* fprintf(ficparo," %lf",param[i][j][k]); */
5075: p[jk]=pstart[jk];
5076: printf(" %f ",pstart[jk]);
5077: fprintf(ficparo," %f ",pstart[jk]);
5078: jk++;
5079: }
5080: printf("\n");
5081: fprintf(ficparo,"\n");
5082: }
5083: }
5084: }
5085: } /* end mle=-2 */
1.226 brouard 5086: dateintmean=dateintsum/k2cpt;
1.296 brouard 5087: date2dmy(dateintmean,&jintmean,&mintmean,&aintmean);
1.240 brouard 5088:
1.226 brouard 5089: fclose(ficresp);
5090: fclose(ficresphtm);
5091: fclose(ficresphtmfr);
1.283 brouard 5092: free_vector(idq,1,nqfveff);
1.226 brouard 5093: free_vector(meanq,1,nqfveff);
1.284 brouard 5094: free_vector(stdq,1,nqfveff);
1.226 brouard 5095: free_matrix(meanqt,1,lastpass,1,nqtveff);
1.253 brouard 5096: free_vector(x, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
5097: free_vector(y, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.251 brouard 5098: free_ma3x(freq,-5,nlstate+ndeath,-5,nlstate+ndeath, iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5099: free_vector(pospropt,1,nlstate);
5100: free_vector(posprop,1,nlstate);
1.251 brouard 5101: free_matrix(prop,1,nlstate,iagemin-AGEMARGE, iagemax+4+AGEMARGE);
1.226 brouard 5102: free_vector(pp,1,nlstate);
5103: /* End of freqsummary */
5104: }
1.126 brouard 5105:
1.268 brouard 5106: /* Simple linear regression */
5107: int linreg(int ifi, int ila, int *no, const double x[], const double y[], double* a, double* b, double* r, double* sa, double * sb) {
5108:
5109: /* y=a+bx regression */
5110: double sumx = 0.0; /* sum of x */
5111: double sumx2 = 0.0; /* sum of x**2 */
5112: double sumxy = 0.0; /* sum of x * y */
5113: double sumy = 0.0; /* sum of y */
5114: double sumy2 = 0.0; /* sum of y**2 */
5115: double sume2 = 0.0; /* sum of square or residuals */
5116: double yhat;
5117:
5118: double denom=0;
5119: int i;
5120: int ne=*no;
5121:
5122: for ( i=ifi, ne=0;i<=ila;i++) {
5123: if(!isfinite(x[i]) || !isfinite(y[i])){
5124: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5125: continue;
5126: }
5127: ne=ne+1;
5128: sumx += x[i];
5129: sumx2 += x[i]*x[i];
5130: sumxy += x[i] * y[i];
5131: sumy += y[i];
5132: sumy2 += y[i]*y[i];
5133: denom = (ne * sumx2 - sumx*sumx);
5134: /* 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); */
5135: }
5136:
5137: denom = (ne * sumx2 - sumx*sumx);
5138: if (denom == 0) {
5139: // vertical, slope m is infinity
5140: *b = INFINITY;
5141: *a = 0;
5142: if (r) *r = 0;
5143: return 1;
5144: }
5145:
5146: *b = (ne * sumxy - sumx * sumy) / denom;
5147: *a = (sumy * sumx2 - sumx * sumxy) / denom;
5148: if (r!=NULL) {
5149: *r = (sumxy - sumx * sumy / ne) / /* compute correlation coeff */
5150: sqrt((sumx2 - sumx*sumx/ne) *
5151: (sumy2 - sumy*sumy/ne));
5152: }
5153: *no=ne;
5154: for ( i=ifi, ne=0;i<=ila;i++) {
5155: if(!isfinite(x[i]) || !isfinite(y[i])){
5156: /* printf(" x[%d]=%f, y[%d]=%f\n",i,x[i],i,y[i]); */
5157: continue;
5158: }
5159: ne=ne+1;
5160: yhat = y[i] - *a -*b* x[i];
5161: sume2 += yhat * yhat ;
5162:
5163: denom = (ne * sumx2 - sumx*sumx);
5164: /* 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); */
5165: }
5166: *sb = sqrt(sume2/(double)(ne-2)/(sumx2 - sumx * sumx /(double)ne));
5167: *sa= *sb * sqrt(sumx2/ne);
5168:
5169: return 0;
5170: }
5171:
1.126 brouard 5172: /************ Prevalence ********************/
1.227 brouard 5173: 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)
5174: {
5175: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
5176: in each health status at the date of interview (if between dateprev1 and dateprev2).
5177: We still use firstpass and lastpass as another selection.
5178: */
1.126 brouard 5179:
1.227 brouard 5180: int i, m, jk, j1, bool, z1,j, iv;
5181: int mi; /* Effective wave */
5182: int iage;
5183: double agebegin, ageend;
5184:
5185: double **prop;
5186: double posprop;
5187: double y2; /* in fractional years */
5188: int iagemin, iagemax;
5189: int first; /** to stop verbosity which is redirected to log file */
5190:
5191: iagemin= (int) agemin;
5192: iagemax= (int) agemax;
5193: /*pp=vector(1,nlstate);*/
1.251 brouard 5194: prop=matrix(1,nlstate,iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5195: /* freq=ma3x(-1,nlstate+ndeath,-1,nlstate+ndeath,iagemin,iagemax+3);*/
5196: j1=0;
1.222 brouard 5197:
1.227 brouard 5198: /*j=cptcoveff;*/
5199: if (cptcovn<1) {j=1;ncodemax[1]=1;}
1.222 brouard 5200:
1.288 brouard 5201: first=0;
1.227 brouard 5202: for(j1=1; j1<= (int) pow(2,cptcoveff);j1++){ /* For each combination of covariate */
5203: for (i=1; i<=nlstate; i++)
1.251 brouard 5204: for(iage=iagemin-AGEMARGE; iage <= iagemax+4+AGEMARGE; iage++)
1.227 brouard 5205: prop[i][iage]=0.0;
5206: printf("Prevalence combination of varying and fixed dummies %d\n",j1);
5207: /* fprintf(ficlog," V%d=%d ",Tvaraff[j1],nbcode[Tvaraff[j1]][codtabm(k,j1)]); */
5208: fprintf(ficlog,"Prevalence combination of varying and fixed dummies %d\n",j1);
5209:
5210: for (i=1; i<=imx; i++) { /* Each individual */
5211: bool=1;
5212: /* for(m=firstpass; m<=lastpass; m++){/\* Other selection (we can limit to certain interviews*\/ */
5213: for(mi=1; mi<wav[i];mi++){ /* For this wave too look where individual can be counted V4=0 V3=0 */
5214: m=mw[mi][i];
5215: /* Tmodelind[z1]=k is the position of the varying covariate in the model, but which # within 1 to ntv? */
5216: /* Tvar[Tmodelind[z1]] is the n of Vn; n-ncovcol-nqv is the first time varying covariate or iv */
5217: for (z1=1; z1<=cptcoveff; z1++){
5218: if( Fixed[Tmodelind[z1]]==1){
5219: iv= Tvar[Tmodelind[z1]]-ncovcol-nqv;
5220: if (cotvar[m][iv][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) /* iv=1 to ntv, right modality */
5221: bool=0;
5222: }else if( Fixed[Tmodelind[z1]]== 0) /* fixed */
5223: if (covar[Tvaraff[z1]][i]!= nbcode[Tvaraff[z1]][codtabm(j1,z1)]) {
5224: bool=0;
5225: }
5226: }
5227: if(bool==1){ /* Otherwise we skip that wave/person */
5228: agebegin=agev[m][i]; /* Age at beginning of wave before transition*/
5229: /* ageend=agev[m][i]+(dh[m][i])*stepm/YEARM; /\* Age at end of wave and transition *\/ */
5230: if(m >=firstpass && m <=lastpass){
5231: y2=anint[m][i]+(mint[m][i]/12.); /* Fractional date in year */
5232: if ((y2>=dateprev1) && (y2<=dateprev2)) { /* Here is the main selection (fractional years) */
5233: if(agev[m][i]==0) agev[m][i]=iagemax+1;
5234: if(agev[m][i]==1) agev[m][i]=iagemax+2;
1.251 brouard 5235: if((int)agev[m][i] <iagemin-AGEMARGE || (int)agev[m][i] >iagemax+4+AGEMARGE){
1.227 brouard 5236: 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);
5237: exit(1);
5238: }
5239: if (s[m][i]>0 && s[m][i]<=nlstate) {
5240: /*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]]);*/
5241: prop[s[m][i]][(int)agev[m][i]] += weight[i];/* At age of beginning of transition, where status is known */
5242: prop[s[m][i]][iagemax+3] += weight[i];
5243: } /* end valid statuses */
5244: } /* end selection of dates */
5245: } /* end selection of waves */
5246: } /* end bool */
5247: } /* end wave */
5248: } /* end individual */
5249: for(i=iagemin; i <= iagemax+3; i++){
5250: for(jk=1,posprop=0; jk <=nlstate ; jk++) {
5251: posprop += prop[jk][i];
5252: }
5253:
5254: for(jk=1; jk <=nlstate ; jk++){
5255: if( i <= iagemax){
5256: if(posprop>=1.e-5){
5257: probs[i][jk][j1]= prop[jk][i]/posprop;
5258: } else{
1.288 brouard 5259: if(!first){
5260: first=1;
1.266 brouard 5261: 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]);
5262: }else{
1.288 brouard 5263: 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 5264: }
5265: }
5266: }
5267: }/* end jk */
5268: }/* end i */
1.222 brouard 5269: /*} *//* end i1 */
1.227 brouard 5270: } /* end j1 */
1.222 brouard 5271:
1.227 brouard 5272: /* free_ma3x(freq,-1,nlstate+ndeath,-1,nlstate+ndeath, iagemin, iagemax+3);*/
5273: /*free_vector(pp,1,nlstate);*/
1.251 brouard 5274: free_matrix(prop,1,nlstate, iagemin-AGEMARGE,iagemax+4+AGEMARGE);
1.227 brouard 5275: } /* End of prevalence */
1.126 brouard 5276:
5277: /************* Waves Concatenation ***************/
5278:
5279: 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)
5280: {
1.298 brouard 5281: /* 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 5282: Death is a valid wave (if date is known).
5283: mw[mi][i] is the mi (mi=1 to wav[i]) effective wave of individual i
5284: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
1.298 brouard 5285: and mw[mi+1][i]. dh depends on stepm. s[m][i] exists for any wave from firstpass to lastpass
1.227 brouard 5286: */
1.126 brouard 5287:
1.224 brouard 5288: int i=0, mi=0, m=0, mli=0;
1.126 brouard 5289: /* int j, k=0,jk, ju, jl,jmin=1e+5, jmax=-1;
5290: double sum=0., jmean=0.;*/
1.224 brouard 5291: int first=0, firstwo=0, firsthree=0, firstfour=0, firstfiv=0;
1.126 brouard 5292: int j, k=0,jk, ju, jl;
5293: double sum=0.;
5294: first=0;
1.214 brouard 5295: firstwo=0;
1.217 brouard 5296: firsthree=0;
1.218 brouard 5297: firstfour=0;
1.164 brouard 5298: jmin=100000;
1.126 brouard 5299: jmax=-1;
5300: jmean=0.;
1.224 brouard 5301:
5302: /* Treating live states */
1.214 brouard 5303: for(i=1; i<=imx; i++){ /* For simple cases and if state is death */
1.224 brouard 5304: mi=0; /* First valid wave */
1.227 brouard 5305: mli=0; /* Last valid wave */
1.309 brouard 5306: m=firstpass; /* Loop on waves */
5307: while(s[m][i] <= nlstate){ /* a live state or unknown state */
1.227 brouard 5308: 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 */
5309: mli=m-1;/* mw[++mi][i]=m-1; */
5310: }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 5311: 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 5312: mli=m;
1.224 brouard 5313: } /* else might be a useless wave -1 and mi is not incremented and mw[mi] not updated */
5314: if(m < lastpass){ /* m < lastpass, standard case */
1.227 brouard 5315: m++; /* mi gives the "effective" current wave, m the current wave, go to next wave by incrementing m */
1.216 brouard 5316: }
1.309 brouard 5317: else{ /* m = lastpass, eventual special issue with warning */
1.224 brouard 5318: #ifdef UNKNOWNSTATUSNOTCONTRIBUTING
1.227 brouard 5319: break;
1.224 brouard 5320: #else
1.309 brouard 5321: if(s[m][i]==-1 && (int) andc[i] == 9999 && (int)anint[m][i] != 9999){ /* case -2 (vital status unknown is warned later */
1.227 brouard 5322: if(firsthree == 0){
1.302 brouard 5323: 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 5324: firsthree=1;
5325: }
1.302 brouard 5326: 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);
1.309 brouard 5327: mw[++mi][i]=m; /* Valid transition with unknown status */
1.227 brouard 5328: mli=m;
5329: }
5330: if(s[m][i]==-2){ /* Vital status is really unknown */
5331: nbwarn++;
1.309 brouard 5332: if((int)anint[m][i] == 9999){ /* Has the vital status really been verified?not a transition */
1.227 brouard 5333: 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);
5334: 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);
5335: }
5336: break;
5337: }
5338: break;
1.224 brouard 5339: #endif
1.227 brouard 5340: }/* End m >= lastpass */
1.126 brouard 5341: }/* end while */
1.224 brouard 5342:
1.227 brouard 5343: /* 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 5344: /* After last pass */
1.224 brouard 5345: /* Treating death states */
1.214 brouard 5346: if (s[m][i] > nlstate){ /* In a death state */
1.227 brouard 5347: /* if( mint[m][i]==mdc[m][i] && anint[m][i]==andc[m][i]){ /\* same date of death and date of interview *\/ */
5348: /* } */
1.126 brouard 5349: mi++; /* Death is another wave */
5350: /* if(mi==0) never been interviewed correctly before death */
1.227 brouard 5351: /* Only death is a correct wave */
1.126 brouard 5352: mw[mi][i]=m;
1.257 brouard 5353: } /* else not in a death state */
1.224 brouard 5354: #ifndef DISPATCHINGKNOWNDEATHAFTERLASTWAVE
1.257 brouard 5355: else if ((int) andc[i] != 9999) { /* Date of death is known */
1.218 brouard 5356: if ((int)anint[m][i]!= 9999) { /* date of last interview is known */
1.309 brouard 5357: 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 5358: nbwarn++;
5359: if(firstfiv==0){
1.309 brouard 5360: 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 5361: firstfiv=1;
5362: }else{
1.309 brouard 5363: 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 5364: }
1.309 brouard 5365: s[m][i]=nlstate+1; /* Fixing the status as death. Be careful if multiple death states */
5366: }else{ /* Month of Death occured afer last wave month, potential bias */
1.227 brouard 5367: nberr++;
5368: if(firstwo==0){
1.309 brouard 5369: 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 5370: firstwo=1;
5371: }
1.309 brouard 5372: 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 5373: }
1.257 brouard 5374: }else{ /* if date of interview is unknown */
1.227 brouard 5375: /* death is known but not confirmed by death status at any wave */
5376: if(firstfour==0){
1.309 brouard 5377: 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 5378: firstfour=1;
5379: }
1.309 brouard 5380: 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 5381: }
1.224 brouard 5382: } /* end if date of death is known */
5383: #endif
1.309 brouard 5384: wav[i]=mi; /* mi should be the last effective wave (or mli), */
5385: /* wav[i]=mw[mi][i]; */
1.126 brouard 5386: if(mi==0){
5387: nbwarn++;
5388: if(first==0){
1.227 brouard 5389: printf("Warning! No valid information for individual %ld line=%d (skipped) and may be others, see log file\n",num[i],i);
5390: first=1;
1.126 brouard 5391: }
5392: if(first==1){
1.227 brouard 5393: fprintf(ficlog,"Warning! No valid information for individual %ld line=%d (skipped)\n",num[i],i);
1.126 brouard 5394: }
5395: } /* end mi==0 */
5396: } /* End individuals */
1.214 brouard 5397: /* wav and mw are no more changed */
1.223 brouard 5398:
1.214 brouard 5399:
1.126 brouard 5400: for(i=1; i<=imx; i++){
5401: for(mi=1; mi<wav[i];mi++){
5402: if (stepm <=0)
1.227 brouard 5403: dh[mi][i]=1;
1.126 brouard 5404: else{
1.260 brouard 5405: if (s[mw[mi+1][i]][i] > nlstate) { /* A death, but what if date is unknown? */
1.227 brouard 5406: if (agedc[i] < 2*AGESUP) {
5407: j= rint(agedc[i]*12-agev[mw[mi][i]][i]*12);
5408: if(j==0) j=1; /* Survives at least one month after exam */
5409: else if(j<0){
5410: nberr++;
5411: 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]);
5412: j=1; /* Temporary Dangerous patch */
5413: 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);
5414: 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]);
5415: 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);
5416: }
5417: k=k+1;
5418: if (j >= jmax){
5419: jmax=j;
5420: ijmax=i;
5421: }
5422: if (j <= jmin){
5423: jmin=j;
5424: ijmin=i;
5425: }
5426: sum=sum+j;
5427: /*if (j<0) printf("j=%d num=%d \n",j,i);*/
5428: /* printf("%d %d %d %d\n", s[mw[mi][i]][i] ,s[mw[mi+1][i]][i],j,i);*/
5429: }
5430: }
5431: else{
5432: j= rint( (agev[mw[mi+1][i]][i]*12 - agev[mw[mi][i]][i]*12));
1.126 brouard 5433: /* 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 5434:
1.227 brouard 5435: k=k+1;
5436: if (j >= jmax) {
5437: jmax=j;
5438: ijmax=i;
5439: }
5440: else if (j <= jmin){
5441: jmin=j;
5442: ijmin=i;
5443: }
5444: /* if (j<10) printf("j=%d jmin=%d num=%d ",j,jmin,i); */
5445: /*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]);*/
5446: if(j<0){
5447: nberr++;
5448: 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]);
5449: 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]);
5450: }
5451: sum=sum+j;
5452: }
5453: jk= j/stepm;
5454: jl= j -jk*stepm;
5455: ju= j -(jk+1)*stepm;
5456: if(mle <=1){ /* only if we use a the linear-interpoloation pseudo-likelihood */
5457: if(jl==0){
5458: dh[mi][i]=jk;
5459: bh[mi][i]=0;
5460: }else{ /* We want a negative bias in order to only have interpolation ie
5461: * to avoid the price of an extra matrix product in likelihood */
5462: dh[mi][i]=jk+1;
5463: bh[mi][i]=ju;
5464: }
5465: }else{
5466: if(jl <= -ju){
5467: dh[mi][i]=jk;
5468: bh[mi][i]=jl; /* bias is positive if real duration
5469: * is higher than the multiple of stepm and negative otherwise.
5470: */
5471: }
5472: else{
5473: dh[mi][i]=jk+1;
5474: bh[mi][i]=ju;
5475: }
5476: if(dh[mi][i]==0){
5477: dh[mi][i]=1; /* At least one step */
5478: bh[mi][i]=ju; /* At least one step */
5479: /* 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);*/
5480: }
5481: } /* end if mle */
1.126 brouard 5482: }
5483: } /* end wave */
5484: }
5485: jmean=sum/k;
5486: 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 5487: 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 5488: }
1.126 brouard 5489:
5490: /*********** Tricode ****************************/
1.220 brouard 5491: void tricode(int *cptcov, int *Tvar, int **nbcode, int imx, int *Ndum)
1.242 brouard 5492: {
5493: /**< Uses cptcovn+2*cptcovprod as the number of covariates */
5494: /* Tvar[i]=atoi(stre); find 'n' in Vn and stores in Tvar. If model=V2+V1 Tvar[1]=2 and Tvar[2]=1
5495: * Boring subroutine which should only output nbcode[Tvar[j]][k]
5496: * Tvar[5] in V2+V1+V3*age+V2*V4 is 4 (V4) even it is a time varying or quantitative variable
5497: * nbcode[Tvar[5]][1]= nbcode[4][1]=0, nbcode[4][2]=1 (usually);
5498: */
1.130 brouard 5499:
1.242 brouard 5500: int ij=1, k=0, j=0, i=0, maxncov=NCOVMAX;
5501: int modmaxcovj=0; /* Modality max of covariates j */
5502: int cptcode=0; /* Modality max of covariates j */
5503: int modmincovj=0; /* Modality min of covariates j */
1.145 brouard 5504:
5505:
1.242 brouard 5506: /* cptcoveff=0; */
5507: /* *cptcov=0; */
1.126 brouard 5508:
1.242 brouard 5509: for (k=1; k <= maxncov; k++) ncodemax[k]=0; /* Horrible constant again replaced by NCOVMAX */
1.285 brouard 5510: for (k=1; k <= maxncov; k++)
5511: for(j=1; j<=2; j++)
5512: nbcode[k][j]=0; /* Valgrind */
1.126 brouard 5513:
1.242 brouard 5514: /* Loop on covariates without age and products and no quantitative variable */
5515: for (k=1; k<=cptcovt; k++) { /* From model V1 + V2*age + V3 + V3*V4 keeps V1 + V3 = 2 only */
5516: for (j=-1; (j < maxncov); j++) Ndum[j]=0;
5517: if(Dummy[k]==0 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5518: switch(Fixed[k]) {
5519: case 0: /* Testing on fixed dummy covariate, simple or product of fixed */
1.311 brouard 5520: modmaxcovj=0;
5521: modmincovj=0;
1.242 brouard 5522: 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*/
5523: ij=(int)(covar[Tvar[k]][i]);
5524: /* ij=0 or 1 or -1. Value of the covariate Tvar[j] for individual i
5525: * If product of Vn*Vm, still boolean *:
5526: * If it was coded 1, 2, 3, 4 should be splitted into 3 boolean variables
5527: * 1 => 0 0 0, 2 => 0 0 1, 3 => 0 1 1, 4=1 0 0 */
5528: /* Finds for covariate j, n=Tvar[j] of Vn . ij is the
5529: modality of the nth covariate of individual i. */
5530: if (ij > modmaxcovj)
5531: modmaxcovj=ij;
5532: else if (ij < modmincovj)
5533: modmincovj=ij;
1.287 brouard 5534: if (ij <0 || ij >1 ){
1.311 brouard 5535: printf("ERROR, IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5536: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=-1, individual %d will be skipped.\n",Tvar[k],i);
5537: fflush(ficlog);
5538: exit(1);
1.287 brouard 5539: }
5540: if ((ij < -1) || (ij > NCOVMAX)){
1.242 brouard 5541: printf( "Error: minimal is less than -1 or maximal is bigger than %d. Exiting. \n", NCOVMAX );
5542: exit(1);
5543: }else
5544: Ndum[ij]++; /*counts and stores the occurence of this modality 0, 1, -1*/
5545: /* If coded 1, 2, 3 , counts the number of 1 Ndum[1], number of 2, Ndum[2], etc */
5546: /*printf("i=%d ij=%d Ndum[ij]=%d imx=%d",i,ij,Ndum[ij],imx);*/
5547: /* getting the maximum value of the modality of the covariate
5548: (should be 0 or 1 now) Tvar[j]. If V=sex and male is coded 0 and
5549: female ies 1, then modmaxcovj=1.
5550: */
5551: } /* end for loop on individuals i */
5552: printf(" Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5553: fprintf(ficlog," Minimal and maximal values of %d th (fixed) covariate V%d: min=%d max=%d \n", k, Tvar[k], modmincovj, modmaxcovj);
5554: cptcode=modmaxcovj;
5555: /* Ndum[0] = frequency of 0 for model-covariate j, Ndum[1] frequency of 1 etc. */
5556: /*for (i=0; i<=cptcode; i++) {*/
5557: for (j=modmincovj; j<=modmaxcovj; j++) { /* j=-1 ? 0 and 1*//* For each value j of the modality of model-cov k */
5558: printf("Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5559: fprintf(ficlog, "Frequencies of (fixed) covariate %d ie V%d with value %d: %d\n", k, Tvar[k], j, Ndum[j]);
5560: if( Ndum[j] != 0 ){ /* Counts if nobody answered modality j ie empty modality, we skip it and reorder */
5561: if( j != -1){
5562: ncodemax[k]++; /* ncodemax[k]= Number of modalities of the k th
5563: covariate for which somebody answered excluding
5564: undefined. Usually 2: 0 and 1. */
5565: }
5566: ncodemaxwundef[k]++; /* ncodemax[j]= Number of modalities of the k th
5567: covariate for which somebody answered including
5568: undefined. Usually 3: -1, 0 and 1. */
5569: } /* In fact ncodemax[k]=2 (dichotom. variables only) but it could be more for
5570: * historical reasons: 3 if coded 1, 2, 3 and 4 and Ndum[2]=0 */
5571: } /* Ndum[-1] number of undefined modalities */
1.231 brouard 5572:
1.242 brouard 5573: /* j is a covariate, n=Tvar[j] of Vn; Fills nbcode */
5574: /* For covariate j, modalities could be 1, 2, 3, 4, 5, 6, 7. */
5575: /* If Ndum[1]=0, Ndum[2]=0, Ndum[3]= 635, Ndum[4]=0, Ndum[5]=0, Ndum[6]=27, Ndum[7]=125; */
5576: /* modmincovj=3; modmaxcovj = 7; */
5577: /* There are only 3 modalities non empty 3, 6, 7 (or 2 if 27 is too few) : ncodemax[j]=3; */
5578: /* which will be coded 0, 1, 2 which in binary on 2=3-1 digits are 0=00 1=01, 2=10; */
5579: /* defining two dummy variables: variables V1_1 and V1_2.*/
5580: /* nbcode[Tvar[j]][ij]=k; */
5581: /* nbcode[Tvar[j]][1]=0; */
5582: /* nbcode[Tvar[j]][2]=1; */
5583: /* nbcode[Tvar[j]][3]=2; */
5584: /* To be continued (not working yet). */
5585: ij=0; /* ij is similar to i but can jump over null modalities */
1.287 brouard 5586:
5587: /* 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*/
5588: /* Skipping the case of missing values by reducing nbcode to 0 and 1 and not -1, 0, 1 */
5589: /* model=V1+V2+V3, if V2=-1, 0 or 1, then nbcode[2][1]=0 and nbcode[2][2]=1 instead of
5590: * nbcode[2][1]=-1, nbcode[2][2]=0 and nbcode[2][3]=1 */
5591: /*, could be restored in the future */
5592: 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 5593: if (Ndum[i] == 0) { /* If nobody responded to this modality k */
5594: break;
5595: }
5596: ij++;
1.287 brouard 5597: 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 5598: cptcode = ij; /* New max modality for covar j */
5599: } /* end of loop on modality i=-1 to 1 or more */
5600: break;
5601: case 1: /* Testing on varying covariate, could be simple and
5602: * should look at waves or product of fixed *
5603: * varying. No time to test -1, assuming 0 and 1 only */
5604: ij=0;
5605: for(i=0; i<=1;i++){
5606: nbcode[Tvar[k]][++ij]=i;
5607: }
5608: break;
5609: default:
5610: break;
5611: } /* end switch */
5612: } /* end dummy test */
1.311 brouard 5613: if(Dummy[k]==1 && Typevar[k] !=1){ /* Dummy covariate and not age product */
5614: 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*/
5615: if(isnan(covar[Tvar[k]][i])){
5616: printf("ERROR, IMaCh doesn't treat fixed quantitative covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5617: fprintf(ficlog,"ERROR, currently IMaCh doesn't treat covariate with missing values V%d=., individual %d will be skipped.\n",Tvar[k],i);
5618: fflush(ficlog);
5619: exit(1);
5620: }
5621: }
5622: }
1.287 brouard 5623: } /* 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 5624:
5625: for (k=-1; k< maxncov; k++) Ndum[k]=0;
5626: /* Look at fixed dummy (single or product) covariates to check empty modalities */
5627: for (i=1; i<=ncovmodel-2-nagesqr; i++) { /* -2, cste and age and eventually age*age */
5628: /* Listing of all covariables in statement model to see if some covariates appear twice. For example, V1 appears twice in V1+V1*V2.*/
5629: 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 */
5630: 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 */
5631: /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, {2, 1, 1, 1, 2, 1, 1, 0, 0} */
5632: } /* V4+V3+V5, Ndum[1]@5={0, 0, 1, 1, 1} */
5633:
5634: ij=0;
5635: /* for (i=0; i<= maxncov-1; i++) { /\* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) *\/ */
5636: for (k=1; k<= cptcovt; k++) { /* modmaxcovj is unknown here. Only Ndum[2(V2),3(age*V3), 5(V3*V2) 6(V1*V4) */
5637: /*printf("Ndum[%d]=%d\n",i, Ndum[i]);*/
5638: /* if((Ndum[i]!=0) && (i<=ncovcol)){ /\* Tvar[i] <= ncovmodel ? *\/ */
5639: if(Ndum[Tvar[k]]!=0 && Dummy[k] == 0 && Typevar[k]==0){ /* Only Dummy and non empty in the model */
5640: /* If product not in single variable we don't print results */
5641: /*printf("diff Ndum[%d]=%d\n",i, Ndum[i]);*/
5642: ++ij;/* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, */
5643: 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*/
5644: Tmodelind[ij]=k; /* Tmodelind: index in model of dummies Tmodelind[1]=2 V4: pos=2; V3: pos=3, V1=9 {2, 3, 9, ?, ?,} */
5645: 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 */
5646: if(Fixed[k]!=0)
5647: anyvaryingduminmodel=1;
5648: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv)){ */
5649: /* Tvaraff[++ij]=-10; /\* Dont'n know how to treat quantitative variables yet *\/ */
5650: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv)){ */
5651: /* Tvaraff[++ij]=i; /\*For printing (unclear) *\/ */
5652: /* }else if((Ndum[i]!=0) && (i<=ncovcol+nqv+ntv+nqtv)){ */
5653: /* Tvaraff[++ij]=-20; /\* Dont'n know how to treat quantitative variables yet *\/ */
5654: }
5655: } /* Tvaraff[1]@5 {3, 4, -20, 0, 0} Very strange */
5656: /* ij--; */
5657: /* cptcoveff=ij; /\*Number of total covariates*\/ */
5658: *cptcov=ij; /*Number of total real effective covariates: effective
5659: * because they can be excluded from the model and real
5660: * if in the model but excluded because missing values, but how to get k from ij?*/
5661: for(j=ij+1; j<= cptcovt; j++){
5662: Tvaraff[j]=0;
5663: Tmodelind[j]=0;
5664: }
5665: for(j=ntveff+1; j<= cptcovt; j++){
5666: TmodelInvind[j]=0;
5667: }
5668: /* To be sorted */
5669: ;
5670: }
1.126 brouard 5671:
1.145 brouard 5672:
1.126 brouard 5673: /*********** Health Expectancies ****************/
5674:
1.235 brouard 5675: 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 5676:
5677: {
5678: /* Health expectancies, no variances */
1.164 brouard 5679: int i, j, nhstepm, hstepm, h, nstepm;
1.126 brouard 5680: int nhstepma, nstepma; /* Decreasing with age */
5681: double age, agelim, hf;
5682: double ***p3mat;
5683: double eip;
5684:
1.238 brouard 5685: /* pstamp(ficreseij); */
1.126 brouard 5686: fprintf(ficreseij,"# (a) Life expectancies by health status at initial age and (b) health expectancies by health status at initial age\n");
5687: fprintf(ficreseij,"# Age");
5688: for(i=1; i<=nlstate;i++){
5689: for(j=1; j<=nlstate;j++){
5690: fprintf(ficreseij," e%1d%1d ",i,j);
5691: }
5692: fprintf(ficreseij," e%1d. ",i);
5693: }
5694: fprintf(ficreseij,"\n");
5695:
5696:
5697: if(estepm < stepm){
5698: printf ("Problem %d lower than %d\n",estepm, stepm);
5699: }
5700: else hstepm=estepm;
5701: /* We compute the life expectancy from trapezoids spaced every estepm months
5702: * This is mainly to measure the difference between two models: for example
5703: * if stepm=24 months pijx are given only every 2 years and by summing them
5704: * we are calculating an estimate of the Life Expectancy assuming a linear
5705: * progression in between and thus overestimating or underestimating according
5706: * to the curvature of the survival function. If, for the same date, we
5707: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5708: * to compare the new estimate of Life expectancy with the same linear
5709: * hypothesis. A more precise result, taking into account a more precise
5710: * curvature will be obtained if estepm is as small as stepm. */
5711:
5712: /* For example we decided to compute the life expectancy with the smallest unit */
5713: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5714: nhstepm is the number of hstepm from age to agelim
5715: nstepm is the number of stepm from age to agelin.
1.270 brouard 5716: Look at hpijx to understand the reason which relies in memory size consideration
1.126 brouard 5717: and note for a fixed period like estepm months */
5718: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5719: survival function given by stepm (the optimization length). Unfortunately it
5720: means that if the survival funtion is printed only each two years of age and if
5721: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5722: results. So we changed our mind and took the option of the best precision.
5723: */
5724: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5725:
5726: agelim=AGESUP;
5727: /* If stepm=6 months */
5728: /* Computed by stepm unit matrices, product of hstepm matrices, stored
5729: in an array of nhstepm length: nhstepm=10, hstepm=4, stepm=6 months */
5730:
5731: /* nhstepm age range expressed in number of stepm */
5732: nstepm=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5733: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5734: /* if (stepm >= YEARM) hstepm=1;*/
5735: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5736: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5737:
5738: for (age=bage; age<=fage; age ++){
5739: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5740: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5741: /* if (stepm >= YEARM) hstepm=1;*/
5742: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
5743:
5744: /* If stepm=6 months */
5745: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5746: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5747:
1.235 brouard 5748: hpxij(p3mat,nhstepma,age,hstepm,x,nlstate,stepm,oldm, savm, cij, nres);
1.126 brouard 5749:
5750: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
5751:
5752: printf("%d|",(int)age);fflush(stdout);
5753: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5754:
5755: /* Computing expectancies */
5756: for(i=1; i<=nlstate;i++)
5757: for(j=1; j<=nlstate;j++)
5758: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5759: eij[i][j][(int)age] += (p3mat[i][j][h]+p3mat[i][j][h+1])/2.0*hf;
5760:
5761: /* 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]);*/
5762:
5763: }
5764:
5765: fprintf(ficreseij,"%3.0f",age );
5766: for(i=1; i<=nlstate;i++){
5767: eip=0;
5768: for(j=1; j<=nlstate;j++){
5769: eip +=eij[i][j][(int)age];
5770: fprintf(ficreseij,"%9.4f", eij[i][j][(int)age] );
5771: }
5772: fprintf(ficreseij,"%9.4f", eip );
5773: }
5774: fprintf(ficreseij,"\n");
5775:
5776: }
5777: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5778: printf("\n");
5779: fprintf(ficlog,"\n");
5780:
5781: }
5782:
1.235 brouard 5783: 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 5784:
5785: {
5786: /* Covariances of health expectancies eij and of total life expectancies according
1.222 brouard 5787: to initial status i, ei. .
1.126 brouard 5788: */
5789: int i, j, nhstepm, hstepm, h, nstepm, k, cptj, cptj2, i2, j2, ij, ji;
5790: int nhstepma, nstepma; /* Decreasing with age */
5791: double age, agelim, hf;
5792: double ***p3matp, ***p3matm, ***varhe;
5793: double **dnewm,**doldm;
5794: double *xp, *xm;
5795: double **gp, **gm;
5796: double ***gradg, ***trgradg;
5797: int theta;
5798:
5799: double eip, vip;
5800:
5801: varhe=ma3x(1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int) fage);
5802: xp=vector(1,npar);
5803: xm=vector(1,npar);
5804: dnewm=matrix(1,nlstate*nlstate,1,npar);
5805: doldm=matrix(1,nlstate*nlstate,1,nlstate*nlstate);
5806:
5807: pstamp(ficresstdeij);
5808: fprintf(ficresstdeij,"# Health expectancies with standard errors\n");
5809: fprintf(ficresstdeij,"# Age");
5810: for(i=1; i<=nlstate;i++){
5811: for(j=1; j<=nlstate;j++)
5812: fprintf(ficresstdeij," e%1d%1d (SE)",i,j);
5813: fprintf(ficresstdeij," e%1d. ",i);
5814: }
5815: fprintf(ficresstdeij,"\n");
5816:
5817: pstamp(ficrescveij);
5818: fprintf(ficrescveij,"# Subdiagonal matrix of covariances of health expectancies by age: cov(eij,ekl)\n");
5819: fprintf(ficrescveij,"# Age");
5820: for(i=1; i<=nlstate;i++)
5821: for(j=1; j<=nlstate;j++){
5822: cptj= (j-1)*nlstate+i;
5823: for(i2=1; i2<=nlstate;i2++)
5824: for(j2=1; j2<=nlstate;j2++){
5825: cptj2= (j2-1)*nlstate+i2;
5826: if(cptj2 <= cptj)
5827: fprintf(ficrescveij," %1d%1d,%1d%1d",i,j,i2,j2);
5828: }
5829: }
5830: fprintf(ficrescveij,"\n");
5831:
5832: if(estepm < stepm){
5833: printf ("Problem %d lower than %d\n",estepm, stepm);
5834: }
5835: else hstepm=estepm;
5836: /* We compute the life expectancy from trapezoids spaced every estepm months
5837: * This is mainly to measure the difference between two models: for example
5838: * if stepm=24 months pijx are given only every 2 years and by summing them
5839: * we are calculating an estimate of the Life Expectancy assuming a linear
5840: * progression in between and thus overestimating or underestimating according
5841: * to the curvature of the survival function. If, for the same date, we
5842: * estimate the model with stepm=1 month, we can keep estepm to 24 months
5843: * to compare the new estimate of Life expectancy with the same linear
5844: * hypothesis. A more precise result, taking into account a more precise
5845: * curvature will be obtained if estepm is as small as stepm. */
5846:
5847: /* For example we decided to compute the life expectancy with the smallest unit */
5848: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
5849: nhstepm is the number of hstepm from age to agelim
5850: nstepm is the number of stepm from age to agelin.
5851: Look at hpijx to understand the reason of that which relies in memory size
5852: and note for a fixed period like estepm months */
5853: /* We decided (b) to get a life expectancy respecting the most precise curvature of the
5854: survival function given by stepm (the optimization length). Unfortunately it
5855: means that if the survival funtion is printed only each two years of age and if
5856: you sum them up and add 1 year (area under the trapezoids) you won't get the same
5857: results. So we changed our mind and took the option of the best precision.
5858: */
5859: hstepm=hstepm/stepm; /* Typically in stepm units, if stepm=6 & estepm=24 , = 24/6 months = 4 */
5860:
5861: /* If stepm=6 months */
5862: /* nhstepm age range expressed in number of stepm */
5863: agelim=AGESUP;
5864: nstepm=(int) rint((agelim-bage)*YEARM/stepm);
5865: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5866: /* if (stepm >= YEARM) hstepm=1;*/
5867: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
5868:
5869: p3matp=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5870: p3matm=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5871: gradg=ma3x(0,nhstepm,1,npar,1,nlstate*nlstate);
5872: trgradg =ma3x(0,nhstepm,1,nlstate*nlstate,1,npar);
5873: gp=matrix(0,nhstepm,1,nlstate*nlstate);
5874: gm=matrix(0,nhstepm,1,nlstate*nlstate);
5875:
5876: for (age=bage; age<=fage; age ++){
5877: nstepma=(int) rint((agelim-bage)*YEARM/stepm); /* Biggest nstepm */
5878: /* Typically if 20 years nstepm = 20*12/6=40 stepm */
5879: /* if (stepm >= YEARM) hstepm=1;*/
5880: nhstepma = nstepma/hstepm;/* Expressed in hstepm, typically nhstepma=40/4=10 */
1.218 brouard 5881:
1.126 brouard 5882: /* If stepm=6 months */
5883: /* Computed by stepm unit matrices, product of hstepma matrices, stored
5884: in an array of nhstepma length: nhstepma=10, hstepm=4, stepm=6 months */
5885:
5886: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
1.218 brouard 5887:
1.126 brouard 5888: /* Computing Variances of health expectancies */
5889: /* Gradient is computed with plus gp and minus gm. Code is duplicated in order to
5890: decrease memory allocation */
5891: for(theta=1; theta <=npar; theta++){
5892: for(i=1; i<=npar; i++){
1.222 brouard 5893: xp[i] = x[i] + (i==theta ?delti[theta]:0);
5894: xm[i] = x[i] - (i==theta ?delti[theta]:0);
1.126 brouard 5895: }
1.235 brouard 5896: hpxij(p3matp,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, cij, nres);
5897: hpxij(p3matm,nhstepm,age,hstepm,xm,nlstate,stepm,oldm,savm, cij, nres);
1.218 brouard 5898:
1.126 brouard 5899: for(j=1; j<= nlstate; j++){
1.222 brouard 5900: for(i=1; i<=nlstate; i++){
5901: for(h=0; h<=nhstepm-1; h++){
5902: gp[h][(j-1)*nlstate + i] = (p3matp[i][j][h]+p3matp[i][j][h+1])/2.;
5903: gm[h][(j-1)*nlstate + i] = (p3matm[i][j][h]+p3matm[i][j][h+1])/2.;
5904: }
5905: }
1.126 brouard 5906: }
1.218 brouard 5907:
1.126 brouard 5908: for(ij=1; ij<= nlstate*nlstate; ij++)
1.222 brouard 5909: for(h=0; h<=nhstepm-1; h++){
5910: gradg[h][theta][ij]= (gp[h][ij]-gm[h][ij])/2./delti[theta];
5911: }
1.126 brouard 5912: }/* End theta */
5913:
5914:
5915: for(h=0; h<=nhstepm-1; h++)
5916: for(j=1; j<=nlstate*nlstate;j++)
1.222 brouard 5917: for(theta=1; theta <=npar; theta++)
5918: trgradg[h][j][theta]=gradg[h][theta][j];
1.126 brouard 5919:
1.218 brouard 5920:
1.222 brouard 5921: for(ij=1;ij<=nlstate*nlstate;ij++)
1.126 brouard 5922: for(ji=1;ji<=nlstate*nlstate;ji++)
1.222 brouard 5923: varhe[ij][ji][(int)age] =0.;
1.218 brouard 5924:
1.222 brouard 5925: printf("%d|",(int)age);fflush(stdout);
5926: fprintf(ficlog,"%d|",(int)age);fflush(ficlog);
5927: for(h=0;h<=nhstepm-1;h++){
1.126 brouard 5928: for(k=0;k<=nhstepm-1;k++){
1.222 brouard 5929: matprod2(dnewm,trgradg[h],1,nlstate*nlstate,1,npar,1,npar,matcov);
5930: matprod2(doldm,dnewm,1,nlstate*nlstate,1,npar,1,nlstate*nlstate,gradg[k]);
5931: for(ij=1;ij<=nlstate*nlstate;ij++)
5932: for(ji=1;ji<=nlstate*nlstate;ji++)
5933: varhe[ij][ji][(int)age] += doldm[ij][ji]*hf*hf;
1.126 brouard 5934: }
5935: }
1.218 brouard 5936:
1.126 brouard 5937: /* Computing expectancies */
1.235 brouard 5938: hpxij(p3matm,nhstepm,age,hstepm,x,nlstate,stepm,oldm, savm, cij,nres);
1.126 brouard 5939: for(i=1; i<=nlstate;i++)
5940: for(j=1; j<=nlstate;j++)
1.222 brouard 5941: for (h=0, eij[i][j][(int)age]=0; h<=nhstepm-1; h++){
5942: eij[i][j][(int)age] += (p3matm[i][j][h]+p3matm[i][j][h+1])/2.0*hf;
1.218 brouard 5943:
1.222 brouard 5944: /* 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 5945:
1.222 brouard 5946: }
1.269 brouard 5947:
5948: /* Standard deviation of expectancies ij */
1.126 brouard 5949: fprintf(ficresstdeij,"%3.0f",age );
5950: for(i=1; i<=nlstate;i++){
5951: eip=0.;
5952: vip=0.;
5953: for(j=1; j<=nlstate;j++){
1.222 brouard 5954: eip += eij[i][j][(int)age];
5955: for(k=1; k<=nlstate;k++) /* Sum on j and k of cov(eij,eik) */
5956: vip += varhe[(j-1)*nlstate+i][(k-1)*nlstate+i][(int)age];
5957: 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 5958: }
5959: fprintf(ficresstdeij," %9.4f (%.4f)", eip, sqrt(vip));
5960: }
5961: fprintf(ficresstdeij,"\n");
1.218 brouard 5962:
1.269 brouard 5963: /* Variance of expectancies ij */
1.126 brouard 5964: fprintf(ficrescveij,"%3.0f",age );
5965: for(i=1; i<=nlstate;i++)
5966: for(j=1; j<=nlstate;j++){
1.222 brouard 5967: cptj= (j-1)*nlstate+i;
5968: for(i2=1; i2<=nlstate;i2++)
5969: for(j2=1; j2<=nlstate;j2++){
5970: cptj2= (j2-1)*nlstate+i2;
5971: if(cptj2 <= cptj)
5972: fprintf(ficrescveij," %.4f", varhe[cptj][cptj2][(int)age]);
5973: }
1.126 brouard 5974: }
5975: fprintf(ficrescveij,"\n");
1.218 brouard 5976:
1.126 brouard 5977: }
5978: free_matrix(gm,0,nhstepm,1,nlstate*nlstate);
5979: free_matrix(gp,0,nhstepm,1,nlstate*nlstate);
5980: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate*nlstate);
5981: free_ma3x(trgradg,0,nhstepm,1,nlstate*nlstate,1,npar);
5982: free_ma3x(p3matm,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5983: free_ma3x(p3matp,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
5984: printf("\n");
5985: fprintf(ficlog,"\n");
1.218 brouard 5986:
1.126 brouard 5987: free_vector(xm,1,npar);
5988: free_vector(xp,1,npar);
5989: free_matrix(dnewm,1,nlstate*nlstate,1,npar);
5990: free_matrix(doldm,1,nlstate*nlstate,1,nlstate*nlstate);
5991: free_ma3x(varhe,1,nlstate*nlstate,1,nlstate*nlstate,(int) bage, (int)fage);
5992: }
1.218 brouard 5993:
1.126 brouard 5994: /************ Variance ******************/
1.235 brouard 5995: 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 5996: {
1.279 brouard 5997: /** Variance of health expectancies
5998: * double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double ** savm,double ftolpl);
5999: * double **newm;
6000: * int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav)
6001: */
1.218 brouard 6002:
6003: /* int movingaverage(); */
6004: double **dnewm,**doldm;
6005: double **dnewmp,**doldmp;
6006: int i, j, nhstepm, hstepm, h, nstepm ;
1.288 brouard 6007: int first=0;
1.218 brouard 6008: int k;
6009: double *xp;
1.279 brouard 6010: double **gp, **gm; /**< for var eij */
6011: double ***gradg, ***trgradg; /**< for var eij */
6012: double **gradgp, **trgradgp; /**< for var p point j */
6013: double *gpp, *gmp; /**< for var p point j */
6014: double **varppt; /**< for var p point j nlstate to nlstate+ndeath */
1.218 brouard 6015: double ***p3mat;
6016: double age,agelim, hf;
6017: /* double ***mobaverage; */
6018: int theta;
6019: char digit[4];
6020: char digitp[25];
6021:
6022: char fileresprobmorprev[FILENAMELENGTH];
6023:
6024: if(popbased==1){
6025: if(mobilav!=0)
6026: strcpy(digitp,"-POPULBASED-MOBILAV_");
6027: else strcpy(digitp,"-POPULBASED-NOMOBIL_");
6028: }
6029: else
6030: strcpy(digitp,"-STABLBASED_");
1.126 brouard 6031:
1.218 brouard 6032: /* if (mobilav!=0) { */
6033: /* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6034: /* if (movingaverage(probs, bage, fage, mobaverage,mobilav)!=0){ */
6035: /* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); */
6036: /* printf(" Error in movingaverage mobilav=%d\n",mobilav); */
6037: /* } */
6038: /* } */
6039:
6040: strcpy(fileresprobmorprev,"PRMORPREV-");
6041: sprintf(digit,"%-d",ij);
6042: /*printf("DIGIT=%s, ij=%d ijr=%-d|\n",digit, ij,ij);*/
6043: strcat(fileresprobmorprev,digit); /* Tvar to be done */
6044: strcat(fileresprobmorprev,digitp); /* Popbased or not, mobilav or not */
6045: strcat(fileresprobmorprev,fileresu);
6046: if((ficresprobmorprev=fopen(fileresprobmorprev,"w"))==NULL) {
6047: printf("Problem with resultfile: %s\n", fileresprobmorprev);
6048: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobmorprev);
6049: }
6050: printf("Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6051: fprintf(ficlog,"Computing total mortality p.j=w1*p1j+w2*p2j+..: result on file '%s' \n",fileresprobmorprev);
6052: pstamp(ficresprobmorprev);
6053: 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 6054: fprintf(ficresprobmorprev,"# Selected quantitative variables and dummies");
6055: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
6056: fprintf(ficresprobmorprev," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
6057: }
6058: for(j=1;j<=cptcoveff;j++)
6059: fprintf(ficresprobmorprev,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(ij,j)]);
6060: fprintf(ficresprobmorprev,"\n");
6061:
1.218 brouard 6062: fprintf(ficresprobmorprev,"# Age cov=%-d",ij);
6063: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6064: fprintf(ficresprobmorprev," p.%-d SE",j);
6065: for(i=1; i<=nlstate;i++)
6066: fprintf(ficresprobmorprev," w%1d p%-d%-d",i,i,j);
6067: }
6068: fprintf(ficresprobmorprev,"\n");
6069:
6070: fprintf(ficgp,"\n# Routine varevsij");
6071: fprintf(ficgp,"\nunset title \n");
6072: /* fprintf(fichtm, "#Local time at start: %s", strstart);*/
6073: 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");
6074: fprintf(fichtm,"\n<br>%s <br>\n",digitp);
1.279 brouard 6075:
1.218 brouard 6076: varppt = matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6077: pstamp(ficresvij);
6078: fprintf(ficresvij,"# Variance and covariance of health expectancies e.j \n# (weighted average of eij where weights are ");
6079: if(popbased==1)
6080: 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);
6081: else
6082: fprintf(ficresvij,"the age specific period (stable) prevalences in each health state \n");
6083: fprintf(ficresvij,"# Age");
6084: for(i=1; i<=nlstate;i++)
6085: for(j=1; j<=nlstate;j++)
6086: fprintf(ficresvij," Cov(e.%1d, e.%1d)",i,j);
6087: fprintf(ficresvij,"\n");
6088:
6089: xp=vector(1,npar);
6090: dnewm=matrix(1,nlstate,1,npar);
6091: doldm=matrix(1,nlstate,1,nlstate);
6092: dnewmp= matrix(nlstate+1,nlstate+ndeath,1,npar);
6093: doldmp= matrix(nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6094:
6095: gradgp=matrix(1,npar,nlstate+1,nlstate+ndeath);
6096: gpp=vector(nlstate+1,nlstate+ndeath);
6097: gmp=vector(nlstate+1,nlstate+ndeath);
6098: trgradgp =matrix(nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
1.126 brouard 6099:
1.218 brouard 6100: if(estepm < stepm){
6101: printf ("Problem %d lower than %d\n",estepm, stepm);
6102: }
6103: else hstepm=estepm;
6104: /* For example we decided to compute the life expectancy with the smallest unit */
6105: /* hstepm beeing the number of stepms, if hstepm=1 the length of hstepm is stepm.
6106: nhstepm is the number of hstepm from age to agelim
6107: nstepm is the number of stepm from age to agelim.
6108: Look at function hpijx to understand why because of memory size limitations,
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 every 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: agelim = AGESUP;
6117: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6118: nstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6119: nhstepm = nstepm/hstepm;/* Expressed in hstepm, typically nhstepm=40/4=10 */
6120: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6121: gradg=ma3x(0,nhstepm,1,npar,1,nlstate);
6122: gp=matrix(0,nhstepm,1,nlstate);
6123: gm=matrix(0,nhstepm,1,nlstate);
6124:
6125:
6126: for(theta=1; theta <=npar; theta++){
6127: for(i=1; i<=npar; i++){ /* Computes gradient x + delta*/
6128: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6129: }
1.279 brouard 6130: /**< Computes the prevalence limit with parameter theta shifted of delta up to ftolpl precision and
6131: * returns into prlim .
1.288 brouard 6132: */
1.242 brouard 6133: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.279 brouard 6134:
6135: /* If popbased = 1 we use crossection prevalences. Previous step is useless but prlim is created */
1.218 brouard 6136: if (popbased==1) {
6137: if(mobilav ==0){
6138: for(i=1; i<=nlstate;i++)
6139: prlim[i][i]=probs[(int)age][i][ij];
6140: }else{ /* mobilav */
6141: for(i=1; i<=nlstate;i++)
6142: prlim[i][i]=mobaverage[(int)age][i][ij];
6143: }
6144: }
1.295 brouard 6145: /**< Computes the shifted transition matrix \f$ {}{h}_p^{ij}x\f$ at horizon h.
1.279 brouard 6146: */
6147: 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 6148: /**< 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 6149: * at horizon h in state j including mortality.
6150: */
1.218 brouard 6151: for(j=1; j<= nlstate; j++){
6152: for(h=0; h<=nhstepm; h++){
6153: for(i=1, gp[h][j]=0.;i<=nlstate;i++)
6154: gp[h][j] += prlim[i][i]*p3mat[i][j][h];
6155: }
6156: }
1.279 brouard 6157: /* Next for computing shifted+ probability of death (h=1 means
1.218 brouard 6158: computed over hstepm matrices product = hstepm*stepm months)
1.279 brouard 6159: as a weighted average of prlim(i) * p(i,j) p.3=w1*p13 + w2*p23 .
1.218 brouard 6160: */
6161: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6162: for(i=1,gpp[j]=0.; i<= nlstate; i++)
6163: gpp[j] += prlim[i][i]*p3mat[i][j][1];
1.279 brouard 6164: }
6165:
6166: /* Again with minus shift */
1.218 brouard 6167:
6168: for(i=1; i<=npar; i++) /* Computes gradient x - delta */
6169: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6170:
1.242 brouard 6171: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp, ij, nres);
1.218 brouard 6172:
6173: if (popbased==1) {
6174: if(mobilav ==0){
6175: for(i=1; i<=nlstate;i++)
6176: prlim[i][i]=probs[(int)age][i][ij];
6177: }else{ /* mobilav */
6178: for(i=1; i<=nlstate;i++)
6179: prlim[i][i]=mobaverage[(int)age][i][ij];
6180: }
6181: }
6182:
1.235 brouard 6183: hpxij(p3mat,nhstepm,age,hstepm,xp,nlstate,stepm,oldm,savm, ij,nres);
1.218 brouard 6184:
6185: for(j=1; j<= nlstate; j++){ /* Sum of wi * eij = e.j */
6186: for(h=0; h<=nhstepm; h++){
6187: for(i=1, gm[h][j]=0.;i<=nlstate;i++)
6188: gm[h][j] += prlim[i][i]*p3mat[i][j][h];
6189: }
6190: }
6191: /* This for computing probability of death (h=1 means
6192: computed over hstepm matrices product = hstepm*stepm months)
6193: as a weighted average of prlim.
6194: */
6195: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6196: for(i=1,gmp[j]=0.; i<= nlstate; i++)
6197: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6198: }
1.279 brouard 6199: /* end shifting computations */
6200:
6201: /**< Computing gradient matrix at horizon h
6202: */
1.218 brouard 6203: for(j=1; j<= nlstate; j++) /* vareij */
6204: for(h=0; h<=nhstepm; h++){
6205: gradg[h][theta][j]= (gp[h][j]-gm[h][j])/2./delti[theta];
6206: }
1.279 brouard 6207: /**< Gradient of overall mortality p.3 (or p.j)
6208: */
6209: for(j=nlstate+1; j<= nlstate+ndeath; j++){ /* var mu mortality from j */
1.218 brouard 6210: gradgp[theta][j]= (gpp[j]-gmp[j])/2./delti[theta];
6211: }
6212:
6213: } /* End theta */
1.279 brouard 6214:
6215: /* We got the gradient matrix for each theta and state j */
1.218 brouard 6216: trgradg =ma3x(0,nhstepm,1,nlstate,1,npar); /* veij */
6217:
6218: for(h=0; h<=nhstepm; h++) /* veij */
6219: for(j=1; j<=nlstate;j++)
6220: for(theta=1; theta <=npar; theta++)
6221: trgradg[h][j][theta]=gradg[h][theta][j];
6222:
6223: for(j=nlstate+1; j<=nlstate+ndeath;j++) /* mu */
6224: for(theta=1; theta <=npar; theta++)
6225: trgradgp[j][theta]=gradgp[theta][j];
1.279 brouard 6226: /**< as well as its transposed matrix
6227: */
1.218 brouard 6228:
6229: hf=hstepm*stepm/YEARM; /* Duration of hstepm expressed in year unit. */
6230: for(i=1;i<=nlstate;i++)
6231: for(j=1;j<=nlstate;j++)
6232: vareij[i][j][(int)age] =0.;
1.279 brouard 6233:
6234: /* Computing trgradg by matcov by gradg at age and summing over h
6235: * and k (nhstepm) formula 15 of article
6236: * Lievre-Brouard-Heathcote
6237: */
6238:
1.218 brouard 6239: for(h=0;h<=nhstepm;h++){
6240: for(k=0;k<=nhstepm;k++){
6241: matprod2(dnewm,trgradg[h],1,nlstate,1,npar,1,npar,matcov);
6242: matprod2(doldm,dnewm,1,nlstate,1,npar,1,nlstate,gradg[k]);
6243: for(i=1;i<=nlstate;i++)
6244: for(j=1;j<=nlstate;j++)
6245: vareij[i][j][(int)age] += doldm[i][j]*hf*hf;
6246: }
6247: }
6248:
1.279 brouard 6249: /* pptj is p.3 or p.j = trgradgp by cov by gradgp, variance of
6250: * p.j overall mortality formula 49 but computed directly because
6251: * we compute the grad (wix pijx) instead of grad (pijx),even if
6252: * wix is independent of theta.
6253: */
1.218 brouard 6254: matprod2(dnewmp,trgradgp,nlstate+1,nlstate+ndeath,1,npar,1,npar,matcov);
6255: matprod2(doldmp,dnewmp,nlstate+1,nlstate+ndeath,1,npar,nlstate+1,nlstate+ndeath,gradgp);
6256: for(j=nlstate+1;j<=nlstate+ndeath;j++)
6257: for(i=nlstate+1;i<=nlstate+ndeath;i++)
6258: varppt[j][i]=doldmp[j][i];
6259: /* end ppptj */
6260: /* x centered again */
6261:
1.242 brouard 6262: prevalim(prlim,nlstate,x,age,oldm,savm,ftolpl,ncvyearp,ij, nres);
1.218 brouard 6263:
6264: if (popbased==1) {
6265: if(mobilav ==0){
6266: for(i=1; i<=nlstate;i++)
6267: prlim[i][i]=probs[(int)age][i][ij];
6268: }else{ /* mobilav */
6269: for(i=1; i<=nlstate;i++)
6270: prlim[i][i]=mobaverage[(int)age][i][ij];
6271: }
6272: }
6273:
6274: /* This for computing probability of death (h=1 means
6275: computed over hstepm (estepm) matrices product = hstepm*stepm months)
6276: as a weighted average of prlim.
6277: */
1.235 brouard 6278: hpxij(p3mat,nhstepm,age,hstepm,x,nlstate,stepm,oldm,savm, ij, nres);
1.218 brouard 6279: for(j=nlstate+1;j<=nlstate+ndeath;j++){
6280: for(i=1,gmp[j]=0.;i<= nlstate; i++)
6281: gmp[j] += prlim[i][i]*p3mat[i][j][1];
6282: }
6283: /* end probability of death */
6284:
6285: fprintf(ficresprobmorprev,"%3d %d ",(int) age, ij);
6286: for(j=nlstate+1; j<=(nlstate+ndeath);j++){
6287: fprintf(ficresprobmorprev," %11.3e %11.3e",gmp[j], sqrt(varppt[j][j]));
6288: for(i=1; i<=nlstate;i++){
6289: fprintf(ficresprobmorprev," %11.3e %11.3e ",prlim[i][i],p3mat[i][j][1]);
6290: }
6291: }
6292: fprintf(ficresprobmorprev,"\n");
6293:
6294: fprintf(ficresvij,"%.0f ",age );
6295: for(i=1; i<=nlstate;i++)
6296: for(j=1; j<=nlstate;j++){
6297: fprintf(ficresvij," %.4f", vareij[i][j][(int)age]);
6298: }
6299: fprintf(ficresvij,"\n");
6300: free_matrix(gp,0,nhstepm,1,nlstate);
6301: free_matrix(gm,0,nhstepm,1,nlstate);
6302: free_ma3x(gradg,0,nhstepm,1,npar,1,nlstate);
6303: free_ma3x(trgradg,0,nhstepm,1,nlstate,1,npar);
6304: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
6305: } /* End age */
6306: free_vector(gpp,nlstate+1,nlstate+ndeath);
6307: free_vector(gmp,nlstate+1,nlstate+ndeath);
6308: free_matrix(gradgp,1,npar,nlstate+1,nlstate+ndeath);
6309: free_matrix(trgradgp,nlstate+1,nlstate+ndeath,1,npar); /* mu or p point j*/
6310: /* fprintf(ficgp,"\nunset parametric;unset label; set ter png small size 320, 240"); */
6311: fprintf(ficgp,"\nunset parametric;unset label; set ter svg size 640, 480");
6312: /* for(j=nlstate+1; j<= nlstate+ndeath; j++){ *//* Only the first actually */
6313: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Force of mortality (year-1)\";");
6314: fprintf(ficgp,"\nset out \"%s%s.svg\";",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
6315: /* fprintf(ficgp,"\n plot \"%s\" u 1:($3*%6.3f) not w l 1 ",fileresprobmorprev,YEARM/estepm); */
6316: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)*%6.3f) t \"95\%% interval\" w l 2 ",fileresprobmorprev,YEARM/estepm); */
6317: /* fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)*%6.3f) not w l 2 ",fileresprobmorprev,YEARM/estepm); */
6318: fprintf(ficgp,"\n plot \"%s\" u 1:($3) not w l lt 1 ",subdirf(fileresprobmorprev));
6319: fprintf(ficgp,"\n replot \"%s\" u 1:(($3+1.96*$4)) t \"95%% interval\" w l lt 2 ",subdirf(fileresprobmorprev));
6320: fprintf(ficgp,"\n replot \"%s\" u 1:(($3-1.96*$4)) not w l lt 2 ",subdirf(fileresprobmorprev));
6321: fprintf(fichtm,"\n<br> File (multiple files are possible if covariates are present): <A href=\"%s\">%s</a>\n",subdirf(fileresprobmorprev),subdirf(fileresprobmorprev));
6322: 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);
6323: /* 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 6324: */
1.218 brouard 6325: /* fprintf(ficgp,"\nset out \"varmuptjgr%s%s%s.svg\";replot;",digitp,optionfilefiname,digit); */
6326: fprintf(ficgp,"\nset out;\nset out \"%s%s.svg\";replot;set out;\n",subdirf3(optionfilefiname,"VARMUPTJGR-",digitp),digit);
1.126 brouard 6327:
1.218 brouard 6328: free_vector(xp,1,npar);
6329: free_matrix(doldm,1,nlstate,1,nlstate);
6330: free_matrix(dnewm,1,nlstate,1,npar);
6331: free_matrix(doldmp,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6332: free_matrix(dnewmp,nlstate+1,nlstate+ndeath,1,npar);
6333: free_matrix(varppt,nlstate+1,nlstate+ndeath,nlstate+1,nlstate+ndeath);
6334: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
6335: fclose(ficresprobmorprev);
6336: fflush(ficgp);
6337: fflush(fichtm);
6338: } /* end varevsij */
1.126 brouard 6339:
6340: /************ Variance of prevlim ******************/
1.269 brouard 6341: 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 6342: {
1.205 brouard 6343: /* Variance of prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
1.126 brouard 6344: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
1.164 brouard 6345:
1.268 brouard 6346: double **dnewmpar,**doldm;
1.126 brouard 6347: int i, j, nhstepm, hstepm;
6348: double *xp;
6349: double *gp, *gm;
6350: double **gradg, **trgradg;
1.208 brouard 6351: double **mgm, **mgp;
1.126 brouard 6352: double age,agelim;
6353: int theta;
6354:
6355: pstamp(ficresvpl);
1.288 brouard 6356: fprintf(ficresvpl,"# Standard deviation of period (forward stable) prevalences \n");
1.241 brouard 6357: fprintf(ficresvpl,"# Age ");
6358: if(nresult >=1)
6359: fprintf(ficresvpl," Result# ");
1.126 brouard 6360: for(i=1; i<=nlstate;i++)
6361: fprintf(ficresvpl," %1d-%1d",i,i);
6362: fprintf(ficresvpl,"\n");
6363:
6364: xp=vector(1,npar);
1.268 brouard 6365: dnewmpar=matrix(1,nlstate,1,npar);
1.126 brouard 6366: doldm=matrix(1,nlstate,1,nlstate);
6367:
6368: hstepm=1*YEARM; /* Every year of age */
6369: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6370: agelim = AGESUP;
6371: for (age=bage; age<=fage; age ++){ /* If stepm=6 months */
6372: nhstepm=(int) rint((agelim-age)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6373: if (stepm >= YEARM) hstepm=1;
6374: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6375: gradg=matrix(1,npar,1,nlstate);
1.208 brouard 6376: mgp=matrix(1,npar,1,nlstate);
6377: mgm=matrix(1,npar,1,nlstate);
1.126 brouard 6378: gp=vector(1,nlstate);
6379: gm=vector(1,nlstate);
6380:
6381: for(theta=1; theta <=npar; theta++){
6382: for(i=1; i<=npar; i++){ /* Computes gradient */
6383: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6384: }
1.288 brouard 6385: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6386: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6387: /* else */
6388: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6389: for(i=1;i<=nlstate;i++){
1.126 brouard 6390: gp[i] = prlim[i][i];
1.208 brouard 6391: mgp[theta][i] = prlim[i][i];
6392: }
1.126 brouard 6393: for(i=1; i<=npar; i++) /* Computes gradient */
6394: xp[i] = x[i] - (i==theta ?delti[theta]:0);
1.288 brouard 6395: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ) */
6396: /* prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres); */
6397: /* else */
6398: prevalim(prlim,nlstate,xp,age,oldm,savm,ftolpl,ncvyearp,ij,nres);
1.208 brouard 6399: for(i=1;i<=nlstate;i++){
1.126 brouard 6400: gm[i] = prlim[i][i];
1.208 brouard 6401: mgm[theta][i] = prlim[i][i];
6402: }
1.126 brouard 6403: for(i=1;i<=nlstate;i++)
6404: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
1.209 brouard 6405: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
1.126 brouard 6406: } /* End theta */
6407:
6408: trgradg =matrix(1,nlstate,1,npar);
6409:
6410: for(j=1; j<=nlstate;j++)
6411: for(theta=1; theta <=npar; theta++)
6412: trgradg[j][theta]=gradg[theta][j];
1.209 brouard 6413: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6414: /* printf("\nmgm mgp %d ",(int)age); */
6415: /* for(j=1; j<=nlstate;j++){ */
6416: /* printf(" %d ",j); */
6417: /* for(theta=1; theta <=npar; theta++) */
6418: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6419: /* printf("\n "); */
6420: /* } */
6421: /* } */
6422: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6423: /* printf("\n gradg %d ",(int)age); */
6424: /* for(j=1; j<=nlstate;j++){ */
6425: /* printf("%d ",j); */
6426: /* for(theta=1; theta <=npar; theta++) */
6427: /* printf("%d %lf ",theta,gradg[theta][j]); */
6428: /* printf("\n "); */
6429: /* } */
6430: /* } */
1.126 brouard 6431:
6432: for(i=1;i<=nlstate;i++)
6433: varpl[i][(int)age] =0.;
1.209 brouard 6434: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
1.268 brouard 6435: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6436: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6437: }else{
1.268 brouard 6438: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6439: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
1.205 brouard 6440: }
1.126 brouard 6441: for(i=1;i<=nlstate;i++)
6442: varpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6443:
6444: fprintf(ficresvpl,"%.0f ",age );
1.241 brouard 6445: if(nresult >=1)
6446: fprintf(ficresvpl,"%d ",nres );
1.288 brouard 6447: for(i=1; i<=nlstate;i++){
1.126 brouard 6448: fprintf(ficresvpl," %.5f (%.5f)",prlim[i][i],sqrt(varpl[i][(int)age]));
1.288 brouard 6449: /* for(j=1;j<=nlstate;j++) */
6450: /* fprintf(ficresvpl," %d %.5f ",j,prlim[j][i]); */
6451: }
1.126 brouard 6452: fprintf(ficresvpl,"\n");
6453: free_vector(gp,1,nlstate);
6454: free_vector(gm,1,nlstate);
1.208 brouard 6455: free_matrix(mgm,1,npar,1,nlstate);
6456: free_matrix(mgp,1,npar,1,nlstate);
1.126 brouard 6457: free_matrix(gradg,1,npar,1,nlstate);
6458: free_matrix(trgradg,1,nlstate,1,npar);
6459: } /* End age */
6460:
6461: free_vector(xp,1,npar);
6462: free_matrix(doldm,1,nlstate,1,npar);
1.268 brouard 6463: free_matrix(dnewmpar,1,nlstate,1,nlstate);
6464:
6465: }
6466:
6467:
6468: /************ Variance of backprevalence limit ******************/
1.269 brouard 6469: 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 6470: {
6471: /* Variance of backward prevalence limit for each state ij using current parameters x[] and estimates of neighbourhood give by delti*/
6472: /* double **prevalim(double **prlim, int nlstate, double *xp, double age, double **oldm, double **savm,double ftolpl);*/
6473:
6474: double **dnewmpar,**doldm;
6475: int i, j, nhstepm, hstepm;
6476: double *xp;
6477: double *gp, *gm;
6478: double **gradg, **trgradg;
6479: double **mgm, **mgp;
6480: double age,agelim;
6481: int theta;
6482:
6483: pstamp(ficresvbl);
6484: fprintf(ficresvbl,"# Standard deviation of back (stable) prevalences \n");
6485: fprintf(ficresvbl,"# Age ");
6486: if(nresult >=1)
6487: fprintf(ficresvbl," Result# ");
6488: for(i=1; i<=nlstate;i++)
6489: fprintf(ficresvbl," %1d-%1d",i,i);
6490: fprintf(ficresvbl,"\n");
6491:
6492: xp=vector(1,npar);
6493: dnewmpar=matrix(1,nlstate,1,npar);
6494: doldm=matrix(1,nlstate,1,nlstate);
6495:
6496: hstepm=1*YEARM; /* Every year of age */
6497: hstepm=hstepm/stepm; /* Typically in stepm units, if j= 2 years, = 2/6 months = 4 */
6498: agelim = AGEINF;
6499: for (age=fage; age>=bage; age --){ /* If stepm=6 months */
6500: nhstepm=(int) rint((age-agelim)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
6501: if (stepm >= YEARM) hstepm=1;
6502: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
6503: gradg=matrix(1,npar,1,nlstate);
6504: mgp=matrix(1,npar,1,nlstate);
6505: mgm=matrix(1,npar,1,nlstate);
6506: gp=vector(1,nlstate);
6507: gm=vector(1,nlstate);
6508:
6509: for(theta=1; theta <=npar; theta++){
6510: for(i=1; i<=npar; i++){ /* Computes gradient */
6511: xp[i] = x[i] + (i==theta ?delti[theta]:0);
6512: }
6513: if(mobilavproj > 0 )
6514: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6515: else
6516: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6517: for(i=1;i<=nlstate;i++){
6518: gp[i] = bprlim[i][i];
6519: mgp[theta][i] = bprlim[i][i];
6520: }
6521: for(i=1; i<=npar; i++) /* Computes gradient */
6522: xp[i] = x[i] - (i==theta ?delti[theta]:0);
6523: if(mobilavproj > 0 )
6524: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6525: else
6526: bprevalim(bprlim, mobaverage,nlstate,xp,age,ftolpl,ncvyearp,ij,nres);
6527: for(i=1;i<=nlstate;i++){
6528: gm[i] = bprlim[i][i];
6529: mgm[theta][i] = bprlim[i][i];
6530: }
6531: for(i=1;i<=nlstate;i++)
6532: gradg[theta][i]= (gp[i]-gm[i])/2./delti[theta];
6533: /* gradg[theta][2]= -gradg[theta][1]; */ /* For testing if nlstate=2 */
6534: } /* End theta */
6535:
6536: trgradg =matrix(1,nlstate,1,npar);
6537:
6538: for(j=1; j<=nlstate;j++)
6539: for(theta=1; theta <=npar; theta++)
6540: trgradg[j][theta]=gradg[theta][j];
6541: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6542: /* printf("\nmgm mgp %d ",(int)age); */
6543: /* for(j=1; j<=nlstate;j++){ */
6544: /* printf(" %d ",j); */
6545: /* for(theta=1; theta <=npar; theta++) */
6546: /* printf(" %d %lf %lf",theta,mgm[theta][j],mgp[theta][j]); */
6547: /* printf("\n "); */
6548: /* } */
6549: /* } */
6550: /* if((int)age==79 ||(int)age== 80 ||(int)age== 81 ){ */
6551: /* printf("\n gradg %d ",(int)age); */
6552: /* for(j=1; j<=nlstate;j++){ */
6553: /* printf("%d ",j); */
6554: /* for(theta=1; theta <=npar; theta++) */
6555: /* printf("%d %lf ",theta,gradg[theta][j]); */
6556: /* printf("\n "); */
6557: /* } */
6558: /* } */
6559:
6560: for(i=1;i<=nlstate;i++)
6561: varbpl[i][(int)age] =0.;
6562: if((int)age==79 ||(int)age== 80 ||(int)age== 81){
6563: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6564: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6565: }else{
6566: matprod2(dnewmpar,trgradg,1,nlstate,1,npar,1,npar,matcov);
6567: matprod2(doldm,dnewmpar,1,nlstate,1,npar,1,nlstate,gradg);
6568: }
6569: for(i=1;i<=nlstate;i++)
6570: varbpl[i][(int)age] = doldm[i][i]; /* Covariances are useless */
6571:
6572: fprintf(ficresvbl,"%.0f ",age );
6573: if(nresult >=1)
6574: fprintf(ficresvbl,"%d ",nres );
6575: for(i=1; i<=nlstate;i++)
6576: fprintf(ficresvbl," %.5f (%.5f)",bprlim[i][i],sqrt(varbpl[i][(int)age]));
6577: fprintf(ficresvbl,"\n");
6578: free_vector(gp,1,nlstate);
6579: free_vector(gm,1,nlstate);
6580: free_matrix(mgm,1,npar,1,nlstate);
6581: free_matrix(mgp,1,npar,1,nlstate);
6582: free_matrix(gradg,1,npar,1,nlstate);
6583: free_matrix(trgradg,1,nlstate,1,npar);
6584: } /* End age */
6585:
6586: free_vector(xp,1,npar);
6587: free_matrix(doldm,1,nlstate,1,npar);
6588: free_matrix(dnewmpar,1,nlstate,1,nlstate);
1.126 brouard 6589:
6590: }
6591:
6592: /************ Variance of one-step probabilities ******************/
6593: 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 6594: {
6595: int i, j=0, k1, l1, tj;
6596: int k2, l2, j1, z1;
6597: int k=0, l;
6598: int first=1, first1, first2;
6599: double cv12, mu1, mu2, lc1, lc2, v12, v21, v11, v22,v1,v2, c12, tnalp;
6600: double **dnewm,**doldm;
6601: double *xp;
6602: double *gp, *gm;
6603: double **gradg, **trgradg;
6604: double **mu;
6605: double age, cov[NCOVMAX+1];
6606: double std=2.0; /* Number of standard deviation wide of confidence ellipsoids */
6607: int theta;
6608: char fileresprob[FILENAMELENGTH];
6609: char fileresprobcov[FILENAMELENGTH];
6610: char fileresprobcor[FILENAMELENGTH];
6611: double ***varpij;
6612:
6613: strcpy(fileresprob,"PROB_");
6614: strcat(fileresprob,fileres);
6615: if((ficresprob=fopen(fileresprob,"w"))==NULL) {
6616: printf("Problem with resultfile: %s\n", fileresprob);
6617: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprob);
6618: }
6619: strcpy(fileresprobcov,"PROBCOV_");
6620: strcat(fileresprobcov,fileresu);
6621: if((ficresprobcov=fopen(fileresprobcov,"w"))==NULL) {
6622: printf("Problem with resultfile: %s\n", fileresprobcov);
6623: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcov);
6624: }
6625: strcpy(fileresprobcor,"PROBCOR_");
6626: strcat(fileresprobcor,fileresu);
6627: if((ficresprobcor=fopen(fileresprobcor,"w"))==NULL) {
6628: printf("Problem with resultfile: %s\n", fileresprobcor);
6629: fprintf(ficlog,"Problem with resultfile: %s\n", fileresprobcor);
6630: }
6631: printf("Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6632: fprintf(ficlog,"Computing standard deviation of one-step probabilities: result on file '%s' \n",fileresprob);
6633: printf("Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6634: fprintf(ficlog,"Computing matrix of variance covariance of one-step probabilities: result on file '%s' \n",fileresprobcov);
6635: printf("and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6636: fprintf(ficlog,"and correlation matrix of one-step probabilities: result on file '%s' \n",fileresprobcor);
6637: pstamp(ficresprob);
6638: fprintf(ficresprob,"#One-step probabilities and stand. devi in ()\n");
6639: fprintf(ficresprob,"# Age");
6640: pstamp(ficresprobcov);
6641: fprintf(ficresprobcov,"#One-step probabilities and covariance matrix\n");
6642: fprintf(ficresprobcov,"# Age");
6643: pstamp(ficresprobcor);
6644: fprintf(ficresprobcor,"#One-step probabilities and correlation matrix\n");
6645: fprintf(ficresprobcor,"# Age");
1.126 brouard 6646:
6647:
1.222 brouard 6648: for(i=1; i<=nlstate;i++)
6649: for(j=1; j<=(nlstate+ndeath);j++){
6650: fprintf(ficresprob," p%1d-%1d (SE)",i,j);
6651: fprintf(ficresprobcov," p%1d-%1d ",i,j);
6652: fprintf(ficresprobcor," p%1d-%1d ",i,j);
6653: }
6654: /* fprintf(ficresprob,"\n");
6655: fprintf(ficresprobcov,"\n");
6656: fprintf(ficresprobcor,"\n");
6657: */
6658: xp=vector(1,npar);
6659: dnewm=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6660: doldm=matrix(1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6661: mu=matrix(1,(nlstate)*(nlstate+ndeath), (int) bage, (int)fage);
6662: varpij=ma3x(1,nlstate*(nlstate+ndeath),1,nlstate*(nlstate+ndeath),(int) bage, (int) fage);
6663: first=1;
6664: fprintf(ficgp,"\n# Routine varprob");
6665: fprintf(fichtm,"\n<li><h4> Computing and drawing one step probabilities with their confidence intervals</h4></li>\n");
6666: fprintf(fichtm,"\n");
6667:
1.288 brouard 6668: 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 6669: 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);
6670: fprintf(fichtmcov,"\nEllipsoids of confidence centered on point (p<inf>ij</inf>, p<inf>kl</inf>) are estimated \
1.126 brouard 6671: and drawn. It helps understanding how is the covariance between two incidences.\
6672: They are expressed in year<sup>-1</sup> in order to be less dependent of stepm.<br>\n");
1.222 brouard 6673: 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 6674: It can be understood this way: if pij and pkl where uncorrelated the (2x2) matrix of covariance \
6675: would have been (1/(var pij), 0 , 0, 1/(var pkl)), and the confidence interval would be 2 \
6676: standard deviations wide on each axis. <br>\
6677: Now, if both incidences are correlated (usual case) we diagonalised the inverse of the covariance matrix\
6678: and made the appropriate rotation to look at the uncorrelated principal directions.<br>\
6679: To be simple, these graphs help to understand the significativity of each parameter in relation to a second other one.<br> \n");
6680:
1.222 brouard 6681: cov[1]=1;
6682: /* tj=cptcoveff; */
1.225 brouard 6683: tj = (int) pow(2,cptcoveff);
1.222 brouard 6684: if (cptcovn<1) {tj=1;ncodemax[1]=1;}
6685: j1=0;
1.224 brouard 6686: for(j1=1; j1<=tj;j1++){ /* For each valid combination of covariates or only once*/
1.222 brouard 6687: if (cptcovn>0) {
6688: fprintf(ficresprob, "\n#********** Variable ");
1.225 brouard 6689: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprob, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6690: fprintf(ficresprob, "**********\n#\n");
6691: fprintf(ficresprobcov, "\n#********** Variable ");
1.225 brouard 6692: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcov, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6693: fprintf(ficresprobcov, "**********\n#\n");
1.220 brouard 6694:
1.222 brouard 6695: fprintf(ficgp, "\n#********** Variable ");
1.225 brouard 6696: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficgp, " V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6697: fprintf(ficgp, "**********\n#\n");
1.220 brouard 6698:
6699:
1.222 brouard 6700: fprintf(fichtmcov, "\n<hr size=\"2\" color=\"#EC5E5E\">********** Variable ");
1.225 brouard 6701: for (z1=1; z1<=cptcoveff; z1++) fprintf(fichtm, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6702: fprintf(fichtmcov, "**********\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 6703:
1.222 brouard 6704: fprintf(ficresprobcor, "\n#********** Variable ");
1.225 brouard 6705: for (z1=1; z1<=cptcoveff; z1++) fprintf(ficresprobcor, "V%d=%d ",Tvaraff[z1],nbcode[Tvaraff[z1]][codtabm(j1,z1)]);
1.222 brouard 6706: fprintf(ficresprobcor, "**********\n#");
6707: if(invalidvarcomb[j1]){
6708: fprintf(ficgp,"\n#Combination (%d) ignored because no cases \n",j1);
6709: fprintf(fichtmcov,"\n<h3>Combination (%d) ignored because no cases </h3>\n",j1);
6710: continue;
6711: }
6712: }
6713: gradg=matrix(1,npar,1,(nlstate)*(nlstate+ndeath));
6714: trgradg=matrix(1,(nlstate)*(nlstate+ndeath),1,npar);
6715: gp=vector(1,(nlstate)*(nlstate+ndeath));
6716: gm=vector(1,(nlstate)*(nlstate+ndeath));
6717: for (age=bage; age<=fage; age ++){
6718: cov[2]=age;
6719: if(nagesqr==1)
6720: cov[3]= age*age;
6721: for (k=1; k<=cptcovn;k++) {
6722: cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,k)];
6723: /*cov[2+nagesqr+k]=nbcode[Tvar[k]][codtabm(j1,Tvar[k])];*//* j1 1 2 3 4
6724: * 1 1 1 1 1
6725: * 2 2 1 1 1
6726: * 3 1 2 1 1
6727: */
6728: /* nbcode[1][1]=0 nbcode[1][2]=1;*/
6729: }
6730: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=cov[2+Tage[k]]*cov[2]; */
6731: for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,k)]*cov[2];
6732: for (k=1; k<=cptcovprod;k++)
6733: cov[2+nagesqr+Tprod[k]]=nbcode[Tvard[k][1]][codtabm(ij,k)]*nbcode[Tvard[k][2]][codtabm(ij,k)];
1.220 brouard 6734:
6735:
1.222 brouard 6736: for(theta=1; theta <=npar; theta++){
6737: for(i=1; i<=npar; i++)
6738: xp[i] = x[i] + (i==theta ?delti[theta]:(double)0);
1.220 brouard 6739:
1.222 brouard 6740: pmij(pmmij,cov,ncovmodel,xp,nlstate);
1.220 brouard 6741:
1.222 brouard 6742: k=0;
6743: for(i=1; i<= (nlstate); i++){
6744: for(j=1; j<=(nlstate+ndeath);j++){
6745: k=k+1;
6746: gp[k]=pmmij[i][j];
6747: }
6748: }
1.220 brouard 6749:
1.222 brouard 6750: for(i=1; i<=npar; i++)
6751: xp[i] = x[i] - (i==theta ?delti[theta]:(double)0);
1.220 brouard 6752:
1.222 brouard 6753: pmij(pmmij,cov,ncovmodel,xp,nlstate);
6754: k=0;
6755: for(i=1; i<=(nlstate); i++){
6756: for(j=1; j<=(nlstate+ndeath);j++){
6757: k=k+1;
6758: gm[k]=pmmij[i][j];
6759: }
6760: }
1.220 brouard 6761:
1.222 brouard 6762: for(i=1; i<= (nlstate)*(nlstate+ndeath); i++)
6763: gradg[theta][i]=(gp[i]-gm[i])/(double)2./delti[theta];
6764: }
1.126 brouard 6765:
1.222 brouard 6766: for(j=1; j<=(nlstate)*(nlstate+ndeath);j++)
6767: for(theta=1; theta <=npar; theta++)
6768: trgradg[j][theta]=gradg[theta][j];
1.220 brouard 6769:
1.222 brouard 6770: matprod2(dnewm,trgradg,1,(nlstate)*(nlstate+ndeath),1,npar,1,npar,matcov);
6771: matprod2(doldm,dnewm,1,(nlstate)*(nlstate+ndeath),1,npar,1,(nlstate)*(nlstate+ndeath),gradg);
1.220 brouard 6772:
1.222 brouard 6773: pmij(pmmij,cov,ncovmodel,x,nlstate);
1.220 brouard 6774:
1.222 brouard 6775: k=0;
6776: for(i=1; i<=(nlstate); i++){
6777: for(j=1; j<=(nlstate+ndeath);j++){
6778: k=k+1;
6779: mu[k][(int) age]=pmmij[i][j];
6780: }
6781: }
6782: for(i=1;i<=(nlstate)*(nlstate+ndeath);i++)
6783: for(j=1;j<=(nlstate)*(nlstate+ndeath);j++)
6784: varpij[i][j][(int)age] = doldm[i][j];
1.220 brouard 6785:
1.222 brouard 6786: /*printf("\n%d ",(int)age);
6787: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6788: printf("%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6789: fprintf(ficlog,"%e [%e ;%e] ",gm[i],gm[i]-2*sqrt(doldm[i][i]),gm[i]+2*sqrt(doldm[i][i]));
6790: }*/
1.220 brouard 6791:
1.222 brouard 6792: fprintf(ficresprob,"\n%d ",(int)age);
6793: fprintf(ficresprobcov,"\n%d ",(int)age);
6794: fprintf(ficresprobcor,"\n%d ",(int)age);
1.220 brouard 6795:
1.222 brouard 6796: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++)
6797: fprintf(ficresprob,"%11.3e (%11.3e) ",mu[i][(int) age],sqrt(varpij[i][i][(int)age]));
6798: for (i=1; i<=(nlstate)*(nlstate+ndeath);i++){
6799: fprintf(ficresprobcov,"%11.3e ",mu[i][(int) age]);
6800: fprintf(ficresprobcor,"%11.3e ",mu[i][(int) age]);
6801: }
6802: i=0;
6803: for (k=1; k<=(nlstate);k++){
6804: for (l=1; l<=(nlstate+ndeath);l++){
6805: i++;
6806: fprintf(ficresprobcov,"\n%d %d-%d",(int)age,k,l);
6807: fprintf(ficresprobcor,"\n%d %d-%d",(int)age,k,l);
6808: for (j=1; j<=i;j++){
6809: /* printf(" k=%d l=%d i=%d j=%d\n",k,l,i,j);fflush(stdout); */
6810: fprintf(ficresprobcov," %11.3e",varpij[i][j][(int)age]);
6811: fprintf(ficresprobcor," %11.3e",varpij[i][j][(int) age]/sqrt(varpij[i][i][(int) age])/sqrt(varpij[j][j][(int)age]));
6812: }
6813: }
6814: }/* end of loop for state */
6815: } /* end of loop for age */
6816: free_vector(gp,1,(nlstate+ndeath)*(nlstate+ndeath));
6817: free_vector(gm,1,(nlstate+ndeath)*(nlstate+ndeath));
6818: free_matrix(trgradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6819: free_matrix(gradg,1,(nlstate+ndeath)*(nlstate+ndeath),1,npar);
6820:
6821: /* Confidence intervalle of pij */
6822: /*
6823: fprintf(ficgp,"\nunset parametric;unset label");
6824: fprintf(ficgp,"\nset log y;unset log x; set xlabel \"Age\";set ylabel \"probability (year-1)\"");
6825: fprintf(ficgp,"\nset ter png small\nset size 0.65,0.65");
6826: 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);
6827: fprintf(fichtm,"\n<br><img src=\"pijgr%s.png\"> ",optionfilefiname);
6828: fprintf(ficgp,"\nset out \"pijgr%s.png\"",optionfilefiname);
6829: fprintf(ficgp,"\nplot \"%s\" every :::%d::%d u 1:2 \"\%%lf",k1,k2,xfilevarprob);
6830: */
6831:
6832: /* Drawing ellipsoids of confidence of two variables p(k1-l1,k2-l2)*/
6833: first1=1;first2=2;
6834: for (k2=1; k2<=(nlstate);k2++){
6835: for (l2=1; l2<=(nlstate+ndeath);l2++){
6836: if(l2==k2) continue;
6837: j=(k2-1)*(nlstate+ndeath)+l2;
6838: for (k1=1; k1<=(nlstate);k1++){
6839: for (l1=1; l1<=(nlstate+ndeath);l1++){
6840: if(l1==k1) continue;
6841: i=(k1-1)*(nlstate+ndeath)+l1;
6842: if(i<=j) continue;
6843: for (age=bage; age<=fage; age ++){
6844: if ((int)age %5==0){
6845: v1=varpij[i][i][(int)age]/stepm*YEARM/stepm*YEARM;
6846: v2=varpij[j][j][(int)age]/stepm*YEARM/stepm*YEARM;
6847: cv12=varpij[i][j][(int)age]/stepm*YEARM/stepm*YEARM;
6848: mu1=mu[i][(int) age]/stepm*YEARM ;
6849: mu2=mu[j][(int) age]/stepm*YEARM;
6850: c12=cv12/sqrt(v1*v2);
6851: /* Computing eigen value of matrix of covariance */
6852: lc1=((v1+v2)+sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6853: lc2=((v1+v2)-sqrt((v1+v2)*(v1+v2) - 4*(v1*v2-cv12*cv12)))/2.;
6854: if ((lc2 <0) || (lc1 <0) ){
6855: if(first2==1){
6856: first1=0;
6857: 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);
6858: }
6859: 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);
6860: /* lc1=fabs(lc1); */ /* If we want to have them positive */
6861: /* lc2=fabs(lc2); */
6862: }
1.220 brouard 6863:
1.222 brouard 6864: /* Eigen vectors */
1.280 brouard 6865: if(1+(v1-lc1)*(v1-lc1)/cv12/cv12 <1.e-5){
6866: printf(" Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6867: fprintf(ficlog," Error sqrt of a negative number: %lf\n",1+(v1-lc1)*(v1-lc1)/cv12/cv12);
6868: v11=(1./sqrt(fabs(1+(v1-lc1)*(v1-lc1)/cv12/cv12)));
6869: }else
6870: v11=(1./sqrt(1+(v1-lc1)*(v1-lc1)/cv12/cv12));
1.222 brouard 6871: /*v21=sqrt(1.-v11*v11); *//* error */
6872: v21=(lc1-v1)/cv12*v11;
6873: v12=-v21;
6874: v22=v11;
6875: tnalp=v21/v11;
6876: if(first1==1){
6877: first1=0;
6878: 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);
6879: }
6880: 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);
6881: /*printf(fignu*/
6882: /* mu1+ v11*lc1*cost + v12*lc2*sin(t) */
6883: /* mu2+ v21*lc1*cost + v22*lc2*sin(t) */
6884: if(first==1){
6885: first=0;
6886: fprintf(ficgp,"\n# Ellipsoids of confidence\n#\n");
6887: fprintf(ficgp,"\nset parametric;unset label");
6888: 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);
6889: fprintf(ficgp,"\nset ter svg size 640, 480");
1.266 brouard 6890: fprintf(fichtmcov,"\n<p><br>Ellipsoids of confidence cov(p%1d%1d,p%1d%1d) expressed in year<sup>-1</sup>\
1.220 brouard 6891: :<a href=\"%s_%d%1d%1d-%1d%1d.svg\"> \
1.201 brouard 6892: %s_%d%1d%1d-%1d%1d.svg</A>, ",k1,l1,k2,l2,\
1.222 brouard 6893: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2, \
6894: subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6895: fprintf(fichtmcov,"\n<br><img src=\"%s_%d%1d%1d-%1d%1d.svg\"> ",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6896: fprintf(fichtmcov,"\n<br> Correlation at age %d (%.3f),",(int) age, c12);
6897: fprintf(ficgp,"\nset out \"%s_%d%1d%1d-%1d%1d.svg\"",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6898: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6899: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6900: 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 6901: mu1,std,v11,sqrt(fabs(lc1)),v12,sqrt(fabs(lc2)), \
6902: mu2,std,v21,sqrt(fabs(lc1)),v22,sqrt(fabs(lc2))); /* For gnuplot only */
1.222 brouard 6903: }else{
6904: first=0;
6905: fprintf(fichtmcov," %d (%.3f),",(int) age, c12);
6906: fprintf(ficgp,"\n# Age %d, p%1d%1d - p%1d%1d",(int) age, k1,l1,k2,l2);
6907: fprintf(ficgp,"\nset label \"%d\" at %11.3e,%11.3e center",(int) age, mu1,mu2);
6908: 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 6909: mu1,std,v11,sqrt(lc1),v12,sqrt(fabs(lc2)), \
6910: mu2,std,v21,sqrt(lc1),v22,sqrt(fabs(lc2)));
1.222 brouard 6911: }/* if first */
6912: } /* age mod 5 */
6913: } /* end loop age */
6914: fprintf(ficgp,"\nset out;\nset out \"%s_%d%1d%1d-%1d%1d.svg\";replot;set out;",subdirf2(optionfilefiname,"VARPIJGR_"), j1,k1,l1,k2,l2);
6915: first=1;
6916: } /*l12 */
6917: } /* k12 */
6918: } /*l1 */
6919: }/* k1 */
6920: } /* loop on combination of covariates j1 */
6921: free_ma3x(varpij,1,nlstate,1,nlstate+ndeath,(int) bage, (int)fage);
6922: free_matrix(mu,1,(nlstate+ndeath)*(nlstate+ndeath),(int) bage, (int)fage);
6923: free_matrix(doldm,1,(nlstate)*(nlstate+ndeath),1,(nlstate)*(nlstate+ndeath));
6924: free_matrix(dnewm,1,(nlstate)*(nlstate+ndeath),1,npar);
6925: free_vector(xp,1,npar);
6926: fclose(ficresprob);
6927: fclose(ficresprobcov);
6928: fclose(ficresprobcor);
6929: fflush(ficgp);
6930: fflush(fichtmcov);
6931: }
1.126 brouard 6932:
6933:
6934: /******************* Printing html file ***********/
1.201 brouard 6935: void printinghtml(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 6936: int lastpass, int stepm, int weightopt, char model[],\
6937: int imx,int jmin, int jmax, double jmeanint,char rfileres[],\
1.296 brouard 6938: int popforecast, int mobilav, int prevfcast, int mobilavproj, int prevbcast, int estepm , \
6939: double jprev1, double mprev1,double anprev1, double dateprev1, double dateprojd, double dateback1, \
6940: double jprev2, double mprev2,double anprev2, double dateprev2, double dateprojf, double dateback2){
1.237 brouard 6941: int jj1, k1, i1, cpt, k4, nres;
1.126 brouard 6942:
6943: fprintf(fichtm,"<ul><li><a href='#firstorder'>Result files (first order: no variance)</a>\n \
6944: <li><a href='#secondorder'>Result files (second order (variance)</a>\n \
6945: </ul>");
1.237 brouard 6946: fprintf(fichtm,"<ul><li> model=1+age+%s\n \
6947: </ul>", model);
1.214 brouard 6948: fprintf(fichtm,"<ul><li><h4><a name='firstorder'>Result files (first order: no variance)</a></h4>\n");
6949: 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",
6950: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTMFR_",".htm"),subdirfext3(optionfilefiname,"PHTMFR_",".htm"));
6951: fprintf(fichtm,"<li> - Observed prevalence in each state (during the period defined between %.lf/%.lf/%.lf and %.lf/%.lf/%.lf): <a href=\"%s\">%s</a> (html file) ",
1.213 brouard 6952: jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,subdirfext3(optionfilefiname,"PHTM_",".htm"),subdirfext3(optionfilefiname,"PHTM_",".htm"));
6953: fprintf(fichtm,", <a href=\"%s\">%s</a> (text file) <br>\n",subdirf2(fileresu,"P_"),subdirf2(fileresu,"P_"));
1.126 brouard 6954: fprintf(fichtm,"\
6955: - Estimated transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
1.201 brouard 6956: stepm,subdirf2(fileresu,"PIJ_"),subdirf2(fileresu,"PIJ_"));
1.126 brouard 6957: fprintf(fichtm,"\
1.217 brouard 6958: - Estimated back transition probabilities over %d (stepm) months: <a href=\"%s\">%s</a><br>\n ",
6959: stepm,subdirf2(fileresu,"PIJB_"),subdirf2(fileresu,"PIJB_"));
6960: fprintf(fichtm,"\
1.288 brouard 6961: - Period (forward) prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6962: subdirf2(fileresu,"PL_"),subdirf2(fileresu,"PL_"));
1.126 brouard 6963: fprintf(fichtm,"\
1.288 brouard 6964: - Backward prevalence in each health state: <a href=\"%s\">%s</a> <br>\n",
1.217 brouard 6965: subdirf2(fileresu,"PLB_"),subdirf2(fileresu,"PLB_"));
6966: fprintf(fichtm,"\
1.211 brouard 6967: - (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 6968: <a href=\"%s\">%s</a> <br>\n",
1.201 brouard 6969: estepm,subdirf2(fileresu,"E_"),subdirf2(fileresu,"E_"));
1.211 brouard 6970: if(prevfcast==1){
6971: fprintf(fichtm,"\
6972: - Prevalence projections by age and states: \
1.201 brouard 6973: <a href=\"%s\">%s</a> <br>\n</li>", subdirf2(fileresu,"F_"),subdirf2(fileresu,"F_"));
1.211 brouard 6974: }
1.126 brouard 6975:
6976:
1.225 brouard 6977: m=pow(2,cptcoveff);
1.222 brouard 6978: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 6979:
1.264 brouard 6980: fprintf(fichtm," \n<ul><li><b>Graphs</b></li><p>");
6981:
6982: jj1=0;
6983:
6984: fprintf(fichtm," \n<ul>");
6985: for(nres=1; nres <= nresult; nres++) /* For each resultline */
6986: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
6987: if(m != 1 && TKresult[nres]!= k1)
6988: continue;
6989: jj1++;
6990: if (cptcovn > 0) {
6991: fprintf(fichtm,"\n<li><a size=\"1\" color=\"#EC5E5E\" href=\"#rescov");
6992: for (cpt=1; cpt<=cptcoveff;cpt++){
6993: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
6994: }
6995: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
6996: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
6997: }
6998: fprintf(fichtm,"\">");
6999:
7000: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
7001: fprintf(fichtm,"************ Results for covariates");
7002: for (cpt=1; cpt<=cptcoveff;cpt++){
7003: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7004: }
7005: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7006: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7007: }
7008: if(invalidvarcomb[k1]){
7009: fprintf(fichtm," Warning Combination (%d) ignored because no cases ",k1);
7010: continue;
7011: }
7012: fprintf(fichtm,"</a></li>");
7013: } /* cptcovn >0 */
7014: }
7015: fprintf(fichtm," \n</ul>");
7016:
1.222 brouard 7017: jj1=0;
1.237 brouard 7018:
7019: for(nres=1; nres <= nresult; nres++) /* For each resultline */
1.241 brouard 7020: for(k1=1; k1<=m;k1++){ /* For each combination of covariate */
1.253 brouard 7021: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7022: continue;
1.220 brouard 7023:
1.222 brouard 7024: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7025: jj1++;
7026: if (cptcovn > 0) {
1.264 brouard 7027: fprintf(fichtm,"\n<p><a name=\"rescov");
7028: for (cpt=1; cpt<=cptcoveff;cpt++){
7029: fprintf(fichtm,"_V%d=%d_",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7030: }
7031: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7032: fprintf(fichtm,"_V%d=%f_",Tvqresult[nres][k4],Tqresult[nres][k4]);
7033: }
7034: fprintf(fichtm,"\"</a>");
7035:
1.222 brouard 7036: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7037: for (cpt=1; cpt<=cptcoveff;cpt++){
1.237 brouard 7038: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],(int)Tresult[nres][cpt]);
7039: printf(" V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);fflush(stdout);
7040: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7041: /* printf(" V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]);fflush(stdout); */
1.222 brouard 7042: }
1.237 brouard 7043: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7044: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7045: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);fflush(stdout);
7046: }
7047:
1.230 brouard 7048: /* if(nqfveff+nqtveff 0) */ /* Test to be done */
1.222 brouard 7049: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
7050: if(invalidvarcomb[k1]){
7051: fprintf(fichtm,"\n<h3>Combination (%d) ignored because no cases </h3>\n",k1);
7052: printf("\nCombination (%d) ignored because no cases \n",k1);
7053: continue;
7054: }
7055: }
7056: /* aij, bij */
1.259 brouard 7057: 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 7058: <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 7059: /* Pij */
1.241 brouard 7060: 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> \
7061: <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 7062: /* Quasi-incidences */
7063: 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 7064: before but expressed in per year i.e. quasi incidences if stepm is small and probabilities too, \
1.211 brouard 7065: 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 7066: 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> \
7067: <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 7068: /* Survival functions (period) in state j */
7069: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7070: fprintf(fichtm,"<br>\n- Survival functions in state %d. And probability to be observed in state %d being in state (1 to %d) at different ages. <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> \
1.241 brouard 7071: <img src=\"%s_%d-%d-%d.svg\">", cpt, cpt, nlstate, subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.222 brouard 7072: }
7073: /* State specific survival functions (period) */
7074: for(cpt=1; cpt<=nlstate;cpt++){
1.292 brouard 7075: fprintf(fichtm,"<br>\n- Survival functions in state %d and in any other live state (total).\
7076: And probability to be observed in various states (up to %d) being in state %d at different ages. \
1.283 brouard 7077: <a href=\"%s_%d-%d-%d.svg\">%s_%d-%d-%d.svg</a><br> <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres,subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.222 brouard 7078: }
1.288 brouard 7079: /* Period (forward stable) prevalence in each health state */
1.222 brouard 7080: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7081: 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> \
7082: <img src=\"%s_%d-%d-%d.svg\">", cpt, nlstate, cpt, subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres,subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.222 brouard 7083: }
1.296 brouard 7084: if(prevbcast==1){
1.288 brouard 7085: /* Backward prevalence in each health state */
1.222 brouard 7086: for(cpt=1; cpt<=nlstate;cpt++){
1.264 brouard 7087: 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 7088: <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 7089: }
1.217 brouard 7090: }
1.222 brouard 7091: if(prevfcast==1){
1.288 brouard 7092: /* Projection of prevalence up to period (forward stable) prevalence in each health state */
1.222 brouard 7093: for(cpt=1; cpt<=nlstate;cpt++){
1.314 ! brouard 7094: 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);
! 7095: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"F_"),subdirf2(optionfilefiname,"F_"));
! 7096: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",
! 7097: subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.222 brouard 7098: }
7099: }
1.296 brouard 7100: if(prevbcast==1){
1.268 brouard 7101: /* Back projection of prevalence up to stable (mixed) back-prevalence in each health state */
7102: for(cpt=1; cpt<=nlstate;cpt++){
1.273 brouard 7103: fprintf(fichtm,"<br>\n- Back projection of cross-sectional prevalence (estimated with cases observed from %.1f to %.1f and mobil_average=%d), \
7104: 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 \
7105: 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 7106: 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);
! 7107: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>)\n<br>",subdirf2(optionfilefiname,"FB_"),subdirf2(optionfilefiname,"FB_"));
! 7108: fprintf(fichtm," <img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
1.268 brouard 7109: }
7110: }
1.220 brouard 7111:
1.222 brouard 7112: for(cpt=1; cpt<=nlstate;cpt++) {
1.314 ! brouard 7113: 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);
! 7114: fprintf(fichtm," (data from text file <a href=\"%s.txt\"> %s.txt</a>)\n<br>",subdirf2(optionfilefiname,"E_"),subdirf2(optionfilefiname,"E_"));
! 7115: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">", subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres );
1.222 brouard 7116: }
7117: /* } /\* end i1 *\/ */
7118: }/* End k1 */
7119: fprintf(fichtm,"</ul>");
1.126 brouard 7120:
1.222 brouard 7121: fprintf(fichtm,"\
1.126 brouard 7122: \n<br><li><h4> <a name='secondorder'>Result files (second order: variances)</a></h4>\n\
1.193 brouard 7123: - Parameter file with estimated parameters and covariance matrix: <a href=\"%s\">%s</a> <br> \
1.203 brouard 7124: - 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 7125: But because parameters are usually highly correlated (a higher incidence of disability \
7126: and a higher incidence of recovery can give very close observed transition) it might \
7127: be very useful to look not only at linear confidence intervals estimated from the \
7128: variances but at the covariance matrix. And instead of looking at the estimated coefficients \
7129: (parameters) of the logistic regression, it might be more meaningful to visualize the \
7130: covariance matrix of the one-step probabilities. \
7131: See page 'Matrix of variance-covariance of one-step probabilities' below. \n", rfileres,rfileres);
1.126 brouard 7132:
1.222 brouard 7133: fprintf(fichtm," - Standard deviation of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
7134: subdirf2(fileresu,"PROB_"),subdirf2(fileresu,"PROB_"));
7135: fprintf(fichtm,"\
1.126 brouard 7136: - Variance-covariance of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7137: subdirf2(fileresu,"PROBCOV_"),subdirf2(fileresu,"PROBCOV_"));
1.126 brouard 7138:
1.222 brouard 7139: fprintf(fichtm,"\
1.126 brouard 7140: - Correlation matrix of one-step probabilities: <a href=\"%s\">%s</a> <br>\n",
1.222 brouard 7141: subdirf2(fileresu,"PROBCOR_"),subdirf2(fileresu,"PROBCOR_"));
7142: fprintf(fichtm,"\
1.126 brouard 7143: - 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): \
7144: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7145: estepm,subdirf2(fileresu,"CVE_"),subdirf2(fileresu,"CVE_"));
1.222 brouard 7146: fprintf(fichtm,"\
1.126 brouard 7147: - (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): \
7148: <a href=\"%s\">%s</a> <br>\n</li>",
1.201 brouard 7149: estepm,subdirf2(fileresu,"STDE_"),subdirf2(fileresu,"STDE_"));
1.222 brouard 7150: fprintf(fichtm,"\
1.288 brouard 7151: - 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 7152: estepm, subdirf2(fileresu,"V_"),subdirf2(fileresu,"V_"));
7153: fprintf(fichtm,"\
1.128 brouard 7154: - 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 7155: estepm, subdirf2(fileresu,"T_"),subdirf2(fileresu,"T_"));
7156: fprintf(fichtm,"\
1.288 brouard 7157: - Standard deviation of forward (period) prevalences: <a href=\"%s\">%s</a> <br>\n",\
1.222 brouard 7158: subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
1.126 brouard 7159:
7160: /* if(popforecast==1) fprintf(fichtm,"\n */
7161: /* - Prevalences forecasting: <a href=\"f%s\">f%s</a> <br>\n */
7162: /* - Population forecasting (if popforecast=1): <a href=\"pop%s\">pop%s</a> <br>\n */
7163: /* <br>",fileres,fileres,fileres,fileres); */
7164: /* else */
7165: /* 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 7166: fflush(fichtm);
7167: fprintf(fichtm," <ul><li><b>Graphs</b></li><p>");
1.126 brouard 7168:
1.225 brouard 7169: m=pow(2,cptcoveff);
1.222 brouard 7170: if (cptcovn < 1) {m=1;ncodemax[1]=1;}
1.126 brouard 7171:
1.222 brouard 7172: jj1=0;
1.237 brouard 7173:
1.241 brouard 7174: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.222 brouard 7175: for(k1=1; k1<=m;k1++){
1.253 brouard 7176: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7177: continue;
1.222 brouard 7178: /* for(i1=1; i1<=ncodemax[k1];i1++){ */
7179: jj1++;
1.126 brouard 7180: if (cptcovn > 0) {
7181: fprintf(fichtm,"<hr size=\"2\" color=\"#EC5E5E\">************ Results for covariates");
1.225 brouard 7182: for (cpt=1; cpt<=cptcoveff;cpt++) /**< cptcoveff number of variables */
1.237 brouard 7183: fprintf(fichtm," V%d=%d ",Tvresult[nres][cpt],Tresult[nres][cpt]);
7184: /* fprintf(fichtm," V%d=%d ",Tvaraff[cpt],nbcode[Tvaraff[cpt]][codtabm(jj1,cpt)]); */
7185: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7186: fprintf(fichtm," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7187: }
7188:
1.126 brouard 7189: fprintf(fichtm," ************\n<hr size=\"2\" color=\"#EC5E5E\">");
1.220 brouard 7190:
1.222 brouard 7191: if(invalidvarcomb[k1]){
7192: fprintf(fichtm,"\n<h4>Combination (%d) ignored because no cases </h4>\n",k1);
7193: continue;
7194: }
1.126 brouard 7195: }
7196: for(cpt=1; cpt<=nlstate;cpt++) {
1.258 brouard 7197: fprintf(fichtm,"\n<br>- Observed (cross-sectional with mov_average=%d) and period (incidence based) \
1.314 ! brouard 7198: 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);
! 7199: fprintf(fichtm," (data from text file <a href=\"%s\">%s</a>)\n <br>",subdirf2(fileresu,"VPL_"),subdirf2(fileresu,"VPL_"));
! 7200: fprintf(fichtm,"<img src=\"%s_%d-%d-%d.svg\">",subdirf2(optionfilefiname,"V_"), cpt,k1,nres);
1.126 brouard 7201: }
7202: fprintf(fichtm,"\n<br>- Total life expectancy by age and \
1.314 ! brouard 7203: health expectancies in each live states (1 to %d). If popbased=1 the smooth (due to the model) \
1.128 brouard 7204: true period expectancies (those weighted with period prevalences are also\
7205: drawn in addition to the population based expectancies computed using\
1.314 ! brouard 7206: 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);
! 7207: fprintf(fichtm," (data from text file <a href=\"%s.txt\">%s.txt</a>) \n<br>",subdirf2(optionfilefiname,"T_"),subdirf2(optionfilefiname,"T_"));
! 7208: fprintf(fichtm,"<img src=\"%s_%d-%d.svg\">",subdirf2(optionfilefiname,"E_"),k1,nres);
1.222 brouard 7209: /* } /\* end i1 *\/ */
7210: }/* End k1 */
1.241 brouard 7211: }/* End nres */
1.222 brouard 7212: fprintf(fichtm,"</ul>");
7213: fflush(fichtm);
1.126 brouard 7214: }
7215:
7216: /******************* Gnuplot file **************/
1.296 brouard 7217: 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 7218:
7219: char dirfileres[132],optfileres[132];
1.264 brouard 7220: char gplotcondition[132], gplotlabel[132];
1.237 brouard 7221: 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 7222: int lv=0, vlv=0, kl=0;
1.130 brouard 7223: int ng=0;
1.201 brouard 7224: int vpopbased;
1.223 brouard 7225: int ioffset; /* variable offset for columns */
1.270 brouard 7226: int iyearc=1; /* variable column for year of projection */
7227: int iagec=1; /* variable column for age of projection */
1.235 brouard 7228: int nres=0; /* Index of resultline */
1.266 brouard 7229: int istart=1; /* For starting graphs in projections */
1.219 brouard 7230:
1.126 brouard 7231: /* if((ficgp=fopen(optionfilegnuplot,"a"))==NULL) { */
7232: /* printf("Problem with file %s",optionfilegnuplot); */
7233: /* fprintf(ficlog,"Problem with file %s",optionfilegnuplot); */
7234: /* } */
7235:
7236: /*#ifdef windows */
7237: fprintf(ficgp,"cd \"%s\" \n",pathc);
1.223 brouard 7238: /*#endif */
1.225 brouard 7239: m=pow(2,cptcoveff);
1.126 brouard 7240:
1.274 brouard 7241: /* diagram of the model */
7242: fprintf(ficgp,"\n#Diagram of the model \n");
7243: fprintf(ficgp,"\ndelta=0.03;delta2=0.07;unset arrow;\n");
7244: fprintf(ficgp,"yoff=(%d > 2? 0:1);\n",nlstate);
7245: 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);
7246:
7247: 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);
7248: fprintf(ficgp,"\n#show arrow\nunset label\n");
7249: 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);
7250: fprintf(ficgp,"\nset label %d+1 sprintf(\"State %%d\",%d+1) center at 0.,0. font \"helvetica, 16\" tc rgbcolor \"red\"\n",nlstate,nlstate);
7251: fprintf(ficgp,"\n#show label\nunset border;unset xtics; unset ytics;\n");
7252: fprintf(ficgp,"\n\nset ter svg size 640, 480;set out \"%s_.svg\" \n",subdirf2(optionfilefiname,"D_"));
7253: fprintf(ficgp,"unset log y; plot [-1.2:1.2][yoff-1.2:1.2] 1/0 not; set out;reset;\n");
7254:
1.202 brouard 7255: /* Contribution to likelihood */
7256: /* Plot the probability implied in the likelihood */
1.223 brouard 7257: fprintf(ficgp,"\n# Contributions to the Likelihood, mle >=1. For mle=4 no interpolation, pure matrix products.\n#\n");
7258: fprintf(ficgp,"\n set log y; unset log x;set xlabel \"Age\"; set ylabel \"Likelihood (-2Log(L))\";");
7259: /* fprintf(ficgp,"\nset ter svg size 640, 480"); */ /* Too big for svg */
7260: fprintf(ficgp,"\nset ter pngcairo size 640, 480");
1.204 brouard 7261: /* nice for mle=4 plot by number of matrix products.
1.202 brouard 7262: replot "rrtest1/toto.txt" u 2:($4 == 1 && $5==2 ? $9 : 1/0):5 t "p12" with point lc 1 */
7263: /* replot exp(p1+p2*x)/(1+exp(p1+p2*x)+exp(p3+p4*x)+exp(p5+p6*x)) t "p12(x)" */
1.223 brouard 7264: /* fprintf(ficgp,"\nset out \"%s.svg\";",subdirf2(optionfilefiname,"ILK_")); */
7265: fprintf(ficgp,"\nset out \"%s-dest.png\";",subdirf2(optionfilefiname,"ILK_"));
7266: 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));
7267: fprintf(ficgp,"\nset out \"%s-ori.png\";",subdirf2(optionfilefiname,"ILK_"));
7268: 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));
7269: for (i=1; i<= nlstate ; i ++) {
7270: fprintf(ficgp,"\nset out \"%s-p%dj.png\";set ylabel \"Probability for each individual/wave\";",subdirf2(optionfilefiname,"ILK_"),i);
7271: fprintf(ficgp,"unset log;\n# plot weighted, mean weight should have point size of 0.5\n plot \"%s\"",subdirf(fileresilk));
7272: 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);
7273: for (j=2; j<= nlstate+ndeath ; j ++) {
7274: 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);
7275: }
7276: fprintf(ficgp,";\nset out; unset ylabel;\n");
7277: }
7278: /* 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 */
7279: /* fprintf(ficgp,"\nset log y;plot \"%s\" u 2:(-$11):3 t \"All sample, all transitions\" with dots lc variable",subdirf(fileresilk)); */
7280: /* fprintf(ficgp,"\nreplot \"%s\" u 2:($3 <= 3 ? -$11 : 1/0):3 t \"First 3 individuals\" with line lc variable", subdirf(fileresilk)); */
7281: fprintf(ficgp,"\nset out;unset log\n");
7282: /* fprintf(ficgp,"\nset out \"%s.svg\"; replot; set out; # bug gnuplot",subdirf2(optionfilefiname,"ILK_")); */
1.202 brouard 7283:
1.126 brouard 7284: strcpy(dirfileres,optionfilefiname);
7285: strcpy(optfileres,"vpl");
1.223 brouard 7286: /* 1eme*/
1.238 brouard 7287: for (cpt=1; cpt<= nlstate ; cpt ++){ /* For each live state */
7288: for (k1=1; k1<= m ; k1 ++){ /* For each valid combination of covariate */
1.236 brouard 7289: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.238 brouard 7290: /* plot [100000000000000000000:-100000000000000000000] "mysbiaspar/vplrmysbiaspar.txt to check */
1.253 brouard 7291: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7292: continue;
7293: /* We are interested in selected combination by the resultline */
1.246 brouard 7294: /* printf("\n# 1st: Period (stable) prevalence with CI: 'VPL_' files and live state =%d ", cpt); */
1.288 brouard 7295: fprintf(ficgp,"\n# 1st: Forward (stable period) prevalence with CI: 'VPL_' files and live state =%d ", cpt);
1.264 brouard 7296: strcpy(gplotlabel,"(");
1.238 brouard 7297: for (k=1; k<=cptcoveff; k++){ /* For each covariate k get corresponding value lv for combination k1 */
7298: lv= decodtabm(k1,k,cptcoveff); /* Should be the value of the covariate corresponding to k1 combination */
7299: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7300: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7301: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7302: vlv= nbcode[Tvaraff[k]][lv]; /* vlv is the value of the covariate lv, 0 or 1 */
7303: /* For each combination of covariate k1 (V1=1, V3=0), we printed the current covariate k and its value vlv */
1.246 brouard 7304: /* printf(" V%d=%d ",Tvaraff[k],vlv); */
1.238 brouard 7305: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7306: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7307: }
7308: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.246 brouard 7309: /* printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]); */
1.238 brouard 7310: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7311: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7312: }
7313: strcpy(gplotlabel+strlen(gplotlabel),")");
1.246 brouard 7314: /* printf("\n#\n"); */
1.238 brouard 7315: fprintf(ficgp,"\n#\n");
7316: if(invalidvarcomb[k1]){
1.260 brouard 7317: /*k1=k1-1;*/ /* To be checked */
1.238 brouard 7318: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7319: continue;
7320: }
1.235 brouard 7321:
1.241 brouard 7322: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"V_"),cpt,k1,nres);
7323: fprintf(ficgp,"\n#set out \"V_%s_%d-%d-%d.svg\" \n",optionfilefiname,cpt,k1,nres);
1.276 brouard 7324: /* fprintf(ficgp,"set label \"Alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel); */
7325: fprintf(ficgp,"set title \"Alive state %d %s\" font \"Helvetica,12\"\n",cpt,gplotlabel);
1.260 brouard 7326: 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);
7327: /* 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); */
7328: /* k1-1 error should be nres-1*/
1.238 brouard 7329: for (i=1; i<= nlstate ; i ++) {
7330: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7331: else fprintf(ficgp," %%*lf (%%*lf)");
7332: }
1.288 brouard 7333: 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 7334: for (i=1; i<= nlstate ; i ++) {
7335: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7336: else fprintf(ficgp," %%*lf (%%*lf)");
7337: }
1.260 brouard 7338: fprintf(ficgp,"\" t\"95%% CI\" w l lt 1,\"%s\" every :::%d::%d u 1:($2==%d ? $3-1.96*$4 : 1/0) \"%%lf %%lf",subdirf2(fileresu,"VPL_"),nres-1,nres-1,nres);
1.238 brouard 7339: for (i=1; i<= nlstate ; i ++) {
7340: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7341: else fprintf(ficgp," %%*lf (%%*lf)");
7342: }
1.265 brouard 7343: /* fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" every :::%d::%d u 1:($%d) t\"Observed prevalence\" w l lt 2",subdirf2(fileresu,"P_"),k1-1,k1-1,2+4*(cpt-1)); */
7344:
7345: fprintf(ficgp,"\" t\"\" w l lt 1,\"%s\" u 1:((",subdirf2(fileresu,"P_"));
7346: if(cptcoveff ==0){
1.271 brouard 7347: fprintf(ficgp,"$%d)) t 'Observed prevalence in state %d' with line lt 3", 2+3*(cpt-1), cpt );
1.265 brouard 7348: }else{
7349: kl=0;
7350: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7351: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7352: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7353: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7354: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7355: vlv= nbcode[Tvaraff[k]][lv];
7356: kl++;
7357: /* 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 *\/ */
7358: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7359: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7360: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7361: if(k==cptcoveff){
7362: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Observed prevalence in state %d' w l lt 2",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
7363: 2+cptcoveff*2+3*(cpt-1), cpt ); /* 4 or 6 ?*/
7364: }else{
7365: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7366: kl++;
7367: }
7368: } /* end covariate */
7369: } /* end if no covariate */
7370:
1.296 brouard 7371: if(prevbcast==1){ /* We need to get the corresponding values of the covariates involved in this combination k1 */
1.238 brouard 7372: /* fprintf(ficgp,",\"%s\" every :::%d::%d u 1:($%d) t\"Backward stable prevalence\" w l lt 3",subdirf2(fileresu,"PLB_"),k1-1,k1-1,1+cpt); */
1.242 brouard 7373: fprintf(ficgp,",\"%s\" u 1:((",subdirf2(fileresu,"PLB_")); /* Age is in 1, nres in 2 to be fixed */
1.238 brouard 7374: if(cptcoveff ==0){
1.245 brouard 7375: fprintf(ficgp,"$%d)) t 'Backward prevalence in state %d' with line lt 3", 2+(cpt-1), cpt );
1.238 brouard 7376: }else{
7377: kl=0;
7378: for (k=1; k<=cptcoveff; k++){ /* For each combination of covariate */
7379: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7380: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7381: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7382: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7383: vlv= nbcode[Tvaraff[k]][lv];
1.223 brouard 7384: kl++;
1.238 brouard 7385: /* 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 *\/ */
7386: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7387: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7388: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0)? $9/(1.-$15) : 1/0):($5==2000? 3:2) t 'p.1' with line lc variable*/
7389: if(k==cptcoveff){
1.245 brouard 7390: fprintf(ficgp,"$%d==%d && $%d==%d)? $%d : 1/0) t 'Backward prevalence in state %d' w l lt 3",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv], \
1.242 brouard 7391: 2+cptcoveff*2+(cpt-1), cpt ); /* 4 or 6 ?*/
1.238 brouard 7392: }else{
7393: fprintf(ficgp,"$%d==%d && $%d==%d && ",kl+1, Tvaraff[k],kl+1+1,nbcode[Tvaraff[k]][lv]);
7394: kl++;
7395: }
7396: } /* end covariate */
7397: } /* end if no covariate */
1.296 brouard 7398: if(prevbcast == 1){
1.268 brouard 7399: fprintf(ficgp,", \"%s\" every :::%d::%d u 1:($2==%d ? $3:1/0) \"%%lf %%lf",subdirf2(fileresu,"VBL_"),nres-1,nres-1,nres);
7400: /* k1-1 error should be nres-1*/
7401: for (i=1; i<= nlstate ; i ++) {
7402: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7403: else fprintf(ficgp," %%*lf (%%*lf)");
7404: }
1.271 brouard 7405: 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 7406: for (i=1; i<= nlstate ; i ++) {
7407: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7408: else fprintf(ficgp," %%*lf (%%*lf)");
7409: }
1.276 brouard 7410: 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 7411: for (i=1; i<= nlstate ; i ++) {
7412: if (i==cpt) fprintf(ficgp," %%lf (%%lf)");
7413: else fprintf(ficgp," %%*lf (%%*lf)");
7414: }
1.274 brouard 7415: fprintf(ficgp,"\" t\"\" w l lt 4");
1.268 brouard 7416: } /* end if backprojcast */
1.296 brouard 7417: } /* end if prevbcast */
1.276 brouard 7418: /* fprintf(ficgp,"\nset out ;unset label;\n"); */
7419: fprintf(ficgp,"\nset out ;unset title;\n");
1.238 brouard 7420: } /* nres */
1.201 brouard 7421: } /* k1 */
7422: } /* cpt */
1.235 brouard 7423:
7424:
1.126 brouard 7425: /*2 eme*/
1.238 brouard 7426: for (k1=1; k1<= m ; k1 ++){
7427: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7428: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7429: continue;
7430: fprintf(ficgp,"\n# 2nd: Total life expectancy with CI: 't' files ");
1.264 brouard 7431: strcpy(gplotlabel,"(");
1.238 brouard 7432: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.225 brouard 7433: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
1.223 brouard 7434: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7435: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7436: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7437: vlv= nbcode[Tvaraff[k]][lv];
7438: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7439: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7440: }
1.237 brouard 7441: /* for(k=1; k <= ncovds; k++){ */
1.236 brouard 7442: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 7443: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.236 brouard 7444: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7445: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7446: }
1.264 brouard 7447: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7448: fprintf(ficgp,"\n#\n");
1.223 brouard 7449: if(invalidvarcomb[k1]){
7450: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7451: continue;
7452: }
1.219 brouard 7453:
1.241 brouard 7454: fprintf(ficgp,"\nset out \"%s_%d-%d.svg\" \n",subdirf2(optionfilefiname,"E_"),k1,nres);
1.238 brouard 7455: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.264 brouard 7456: fprintf(ficgp,"\nset label \"popbased %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",vpopbased,gplotlabel);
7457: if(vpopbased==0){
1.238 brouard 7458: fprintf(ficgp,"set ylabel \"Years\" \nset ter svg size 640, 480\nplot [%.f:%.f] ",ageminpar,fage);
1.264 brouard 7459: }else
1.238 brouard 7460: fprintf(ficgp,"\nreplot ");
7461: for (i=1; i<= nlstate+1 ; i ++) {
7462: k=2*i;
1.261 brouard 7463: 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 7464: for (j=1; j<= nlstate+1 ; j ++) {
7465: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7466: else fprintf(ficgp," %%*lf (%%*lf)");
7467: }
7468: if (i== 1) fprintf(ficgp,"\" t\"TLE\" w l lt %d, \\\n",i);
7469: else fprintf(ficgp,"\" t\"LE in state (%d)\" w l lt %d, \\\n",i-1,i+1);
1.261 brouard 7470: 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 7471: for (j=1; j<= nlstate+1 ; j ++) {
7472: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7473: else fprintf(ficgp," %%*lf (%%*lf)");
7474: }
7475: fprintf(ficgp,"\" t\"\" w l lt 0,");
1.261 brouard 7476: 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 7477: for (j=1; j<= nlstate+1 ; j ++) {
7478: if (j==i) fprintf(ficgp," %%lf (%%lf)");
7479: else fprintf(ficgp," %%*lf (%%*lf)");
7480: }
7481: if (i== (nlstate+1)) fprintf(ficgp,"\" t\"\" w l lt 0");
7482: else fprintf(ficgp,"\" t\"\" w l lt 0,\\\n");
7483: } /* state */
7484: } /* vpopbased */
1.264 brouard 7485: 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 7486: } /* end nres */
7487: } /* k1 end 2 eme*/
7488:
7489:
7490: /*3eme*/
7491: for (k1=1; k1<= m ; k1 ++){
7492: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7493: if(m != 1 && TKresult[nres]!= k1)
1.238 brouard 7494: continue;
7495:
7496: for (cpt=1; cpt<= nlstate ; cpt ++) {
1.261 brouard 7497: fprintf(ficgp,"\n\n# 3d: Life expectancy with EXP_ files: combination=%d state=%d",k1, cpt);
1.264 brouard 7498: strcpy(gplotlabel,"(");
1.238 brouard 7499: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7500: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7501: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7502: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7503: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7504: vlv= nbcode[Tvaraff[k]][lv];
7505: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7506: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7507: }
7508: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7509: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7510: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7511: }
1.264 brouard 7512: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7513: fprintf(ficgp,"\n#\n");
7514: if(invalidvarcomb[k1]){
7515: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7516: continue;
7517: }
7518:
7519: /* k=2+nlstate*(2*cpt-2); */
7520: k=2+(nlstate+1)*(cpt-1);
1.241 brouard 7521: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"EXP_"),cpt,k1,nres);
1.264 brouard 7522: fprintf(ficgp,"set label \"%s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel);
1.238 brouard 7523: fprintf(ficgp,"set ter svg size 640, 480\n\
1.261 brouard 7524: 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 7525: /*fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d-2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7526: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7527: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
7528: fprintf(ficgp,",\"e%s\" every :::%d::%d u 1:($%d+2*$%d) \"\%%lf ",fileres,k1-1,k1-1,k,k+1);
7529: for (i=1; i<= nlstate*2 ; i ++) fprintf(ficgp,"\%%lf (\%%lf) ");
7530: fprintf(ficgp,"\" t \"e%d1\" w l",cpt);
1.219 brouard 7531:
1.238 brouard 7532: */
7533: for (i=1; i< nlstate ; i ++) {
1.261 brouard 7534: 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 7535: /* 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 7536:
1.238 brouard 7537: }
1.261 brouard 7538: 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 7539: }
1.264 brouard 7540: fprintf(ficgp,"\nunset label;\n");
1.238 brouard 7541: } /* end nres */
7542: } /* end kl 3eme */
1.126 brouard 7543:
1.223 brouard 7544: /* 4eme */
1.201 brouard 7545: /* Survival functions (period) from state i in state j by initial state i */
1.238 brouard 7546: for (k1=1; k1<=m; k1++){ /* For each covariate and each value */
7547: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7548: if(m != 1 && TKresult[nres]!= k1)
1.223 brouard 7549: continue;
1.238 brouard 7550: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state cpt*/
1.264 brouard 7551: strcpy(gplotlabel,"(");
1.238 brouard 7552: fprintf(ficgp,"\n#\n#\n# Survival functions in state j : 'LIJ_' files, cov=%d state=%d",k1, cpt);
7553: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7554: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7555: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7556: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7557: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7558: vlv= nbcode[Tvaraff[k]][lv];
7559: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7560: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7561: }
7562: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7563: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7564: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7565: }
1.264 brouard 7566: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7567: fprintf(ficgp,"\n#\n");
7568: if(invalidvarcomb[k1]){
7569: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7570: continue;
1.223 brouard 7571: }
1.238 brouard 7572:
1.241 brouard 7573: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJ_"),cpt,k1,nres);
1.264 brouard 7574: 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 7575: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7576: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7577: k=3;
7578: for (i=1; i<= nlstate ; i ++){
7579: if(i==1){
7580: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7581: }else{
7582: fprintf(ficgp,", '' ");
7583: }
7584: l=(nlstate+ndeath)*(i-1)+1;
7585: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7586: for (j=2; j<= nlstate+ndeath ; j ++)
7587: fprintf(ficgp,"+$%d",k+l+j-1);
7588: fprintf(ficgp,")) t \"l(%d,%d)\" w l",i,cpt);
7589: } /* nlstate */
1.264 brouard 7590: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7591: } /* end cpt state*/
7592: } /* end nres */
7593: } /* end covariate k1 */
7594:
1.220 brouard 7595: /* 5eme */
1.201 brouard 7596: /* Survival functions (period) from state i in state j by final state j */
1.238 brouard 7597: for (k1=1; k1<= m ; k1++){ /* For each covariate combination if any */
7598: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7599: if(m != 1 && TKresult[nres]!= k1)
1.227 brouard 7600: continue;
1.238 brouard 7601: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each inital state */
1.264 brouard 7602: strcpy(gplotlabel,"(");
1.238 brouard 7603: 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);
7604: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7605: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7606: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7607: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7608: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7609: vlv= nbcode[Tvaraff[k]][lv];
7610: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7611: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.238 brouard 7612: }
7613: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7614: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7615: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.238 brouard 7616: }
1.264 brouard 7617: strcpy(gplotlabel+strlen(gplotlabel),")");
1.238 brouard 7618: fprintf(ficgp,"\n#\n");
7619: if(invalidvarcomb[k1]){
7620: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7621: continue;
7622: }
1.227 brouard 7623:
1.241 brouard 7624: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"LIJT_"),cpt,k1,nres);
1.264 brouard 7625: 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 7626: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability to be alive\" \n\
7627: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7628: k=3;
7629: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7630: if(j==1)
7631: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7632: else
7633: fprintf(ficgp,", '' ");
7634: l=(nlstate+ndeath)*(cpt-1) +j;
7635: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):($%d",k1,k+l);
7636: /* for (i=2; i<= nlstate+ndeath ; i ++) */
7637: /* fprintf(ficgp,"+$%d",k+l+i-1); */
7638: fprintf(ficgp,") t \"l(%d,%d)\" w l",cpt,j);
7639: } /* nlstate */
7640: fprintf(ficgp,", '' ");
7641: fprintf(ficgp," u (($1==%d && (floor($2)%%5 == 0)) ? ($3):1/0):(",k1);
7642: for (j=1; j<= nlstate ; j ++){ /* Lived in state j */
7643: l=(nlstate+ndeath)*(cpt-1) +j;
7644: if(j < nlstate)
7645: fprintf(ficgp,"$%d +",k+l);
7646: else
7647: fprintf(ficgp,"$%d) t\"l(%d,.)\" w l",k+l,cpt);
7648: }
1.264 brouard 7649: fprintf(ficgp,"\nset out; unset label;\n");
1.238 brouard 7650: } /* end cpt state*/
7651: } /* end covariate */
7652: } /* end nres */
1.227 brouard 7653:
1.220 brouard 7654: /* 6eme */
1.202 brouard 7655: /* CV preval stable (period) for each covariate */
1.237 brouard 7656: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7657: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7658: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7659: continue;
1.255 brouard 7660: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state of arrival */
1.264 brouard 7661: strcpy(gplotlabel,"(");
1.288 brouard 7662: fprintf(ficgp,"\n#\n#\n#CV preval stable (forward): 'pij' files, covariatecombination#=%d state=%d",k1, cpt);
1.225 brouard 7663: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
1.227 brouard 7664: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7665: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7666: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7667: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7668: vlv= nbcode[Tvaraff[k]][lv];
7669: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7670: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.211 brouard 7671: }
1.237 brouard 7672: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7673: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7674: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7675: }
1.264 brouard 7676: strcpy(gplotlabel+strlen(gplotlabel),")");
1.211 brouard 7677: fprintf(ficgp,"\n#\n");
1.223 brouard 7678: if(invalidvarcomb[k1]){
1.227 brouard 7679: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7680: continue;
1.223 brouard 7681: }
1.227 brouard 7682:
1.241 brouard 7683: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"P_"),cpt,k1,nres);
1.264 brouard 7684: 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 7685: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7686: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.211 brouard 7687: k=3; /* Offset */
1.255 brouard 7688: for (i=1; i<= nlstate ; i ++){ /* State of origin */
1.227 brouard 7689: if(i==1)
7690: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJ_"));
7691: else
7692: fprintf(ficgp,", '' ");
1.255 brouard 7693: l=(nlstate+ndeath)*(i-1)+1; /* 1, 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7694: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l);
7695: for (j=2; j<= nlstate ; j ++)
7696: fprintf(ficgp,"+$%d",k+l+j-1);
7697: fprintf(ficgp,")) t \"prev(%d,%d)\" w l",i,cpt);
1.153 brouard 7698: } /* nlstate */
1.264 brouard 7699: fprintf(ficgp,"\nset out; unset label;\n");
1.153 brouard 7700: } /* end cpt state*/
7701: } /* end covariate */
1.227 brouard 7702:
7703:
1.220 brouard 7704: /* 7eme */
1.296 brouard 7705: if(prevbcast == 1){
1.288 brouard 7706: /* CV backward prevalence for each covariate */
1.237 brouard 7707: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7708: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7709: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7710: continue;
1.268 brouard 7711: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life origin state */
1.264 brouard 7712: strcpy(gplotlabel,"(");
1.288 brouard 7713: fprintf(ficgp,"\n#\n#\n#CV Backward stable prevalence: 'pijb' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7714: for (k=1; k<=cptcoveff; k++){ /* For each covariate and each value */
7715: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate number corresponding to k1 combination */
7716: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7717: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
1.223 brouard 7718: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
1.227 brouard 7719: vlv= nbcode[Tvaraff[k]][lv];
7720: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7721: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7722: }
1.237 brouard 7723: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7724: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7725: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7726: }
1.264 brouard 7727: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7728: fprintf(ficgp,"\n#\n");
7729: if(invalidvarcomb[k1]){
7730: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7731: continue;
7732: }
7733:
1.241 brouard 7734: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PB_"),cpt,k1,nres);
1.268 brouard 7735: 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 7736: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Probability\" \n\
1.238 brouard 7737: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.227 brouard 7738: k=3; /* Offset */
1.268 brouard 7739: for (i=1; i<= nlstate ; i ++){ /* State of arrival */
1.227 brouard 7740: if(i==1)
7741: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"PIJB_"));
7742: else
7743: fprintf(ficgp,", '' ");
7744: /* l=(nlstate+ndeath)*(i-1)+1; */
1.255 brouard 7745: l=(nlstate+ndeath)*(cpt-1)+1; /* fixed for i; cpt=1 1, cpt=2 1+ nlstate+ndeath, 1+2*(nlstate+ndeath) */
1.227 brouard 7746: /* fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d/($%d",k1,k+l+(cpt-1),k+l); /\* a vérifier *\/ */
7747: /* 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 7748: fprintf(ficgp," u ($1==%d ? ($3):1/0):($%d",k1,k+l+i-1); /* To be verified */
1.227 brouard 7749: /* for (j=2; j<= nlstate ; j ++) */
7750: /* fprintf(ficgp,"+$%d",k+l+j-1); */
7751: /* /\* fprintf(ficgp,"+$%d",k+l+j-1); *\/ */
1.268 brouard 7752: fprintf(ficgp,") t \"bprev(%d,%d)\" w l",cpt,i);
1.227 brouard 7753: } /* nlstate */
1.264 brouard 7754: fprintf(ficgp,"\nset out; unset label;\n");
1.218 brouard 7755: } /* end cpt state*/
7756: } /* end covariate */
1.296 brouard 7757: } /* End if prevbcast */
1.218 brouard 7758:
1.223 brouard 7759: /* 8eme */
1.218 brouard 7760: if(prevfcast==1){
1.288 brouard 7761: /* Projection from cross-sectional to forward stable (period) prevalence for each covariate */
1.218 brouard 7762:
1.237 brouard 7763: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7764: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 7765: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 7766: continue;
1.211 brouard 7767: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
1.264 brouard 7768: strcpy(gplotlabel,"(");
1.288 brouard 7769: fprintf(ficgp,"\n#\n#\n#Projection of prevalence to forward stable prevalence (period): 'PROJ_' files, covariatecombination#=%d state=%d",k1, cpt);
1.227 brouard 7770: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7771: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7772: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7773: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7774: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7775: vlv= nbcode[Tvaraff[k]][lv];
7776: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
1.264 brouard 7777: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
1.227 brouard 7778: }
1.237 brouard 7779: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7780: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 7781: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 7782: }
1.264 brouard 7783: strcpy(gplotlabel+strlen(gplotlabel),")");
1.227 brouard 7784: fprintf(ficgp,"\n#\n");
7785: if(invalidvarcomb[k1]){
7786: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7787: continue;
7788: }
7789:
7790: fprintf(ficgp,"# hpijx=probability over h years, hp.jx is weighted by observed prev\n ");
1.241 brouard 7791: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJ_"),cpt,k1,nres);
1.264 brouard 7792: 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 7793: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
1.238 brouard 7794: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
1.266 brouard 7795:
7796: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7797: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7798: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7799: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
1.227 brouard 7800: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7801: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7802: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7803: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
1.266 brouard 7804: if(i==istart){
1.227 brouard 7805: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"F_"));
7806: }else{
7807: fprintf(ficgp,",\\\n '' ");
7808: }
7809: if(cptcoveff ==0){ /* No covariate */
7810: ioffset=2; /* Age is in 2 */
7811: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7812: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7813: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7814: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7815: fprintf(ficgp," u %d:(", ioffset);
1.266 brouard 7816: if(i==nlstate+1){
1.270 brouard 7817: fprintf(ficgp," $%d/(1.-$%d)):1 t 'pw.%d' with line lc variable ", \
1.266 brouard 7818: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7819: fprintf(ficgp,",\\\n '' ");
7820: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7821: fprintf(ficgp," (($1-$2) == %d ) ? $%d/(1.-$%d) : 1/0):1 with labels center not ", \
1.266 brouard 7822: offyear, \
1.268 brouard 7823: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate );
1.266 brouard 7824: }else
1.227 brouard 7825: fprintf(ficgp," $%d/(1.-$%d)) t 'p%d%d' with line ", \
7826: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7827: }else{ /* more than 2 covariates */
1.270 brouard 7828: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7829: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7830: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7831: iyearc=ioffset-1;
7832: iagec=ioffset;
1.227 brouard 7833: fprintf(ficgp," u %d:(",ioffset);
7834: kl=0;
7835: strcpy(gplotcondition,"(");
7836: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7837: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7838: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7839: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7840: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7841: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7842: kl++;
7843: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7844: kl++;
7845: if(k <cptcoveff && cptcoveff>1)
7846: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7847: }
7848: strcpy(gplotcondition+strlen(gplotcondition),")");
7849: /* 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 *\/ */
7850: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7851: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7852: /* '' 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*/
7853: if(i==nlstate+1){
1.270 brouard 7854: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0):%d t 'p.%d' with line lc variable", gplotcondition, \
7855: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,iyearc, cpt );
1.266 brouard 7856: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7857: fprintf(ficgp," u %d:(",iagec);
7858: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d/(1.-$%d) : 1/0):%d with labels center not ", gplotcondition, \
7859: iyearc, iagec, offyear, \
7860: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate, iyearc );
1.266 brouard 7861: /* '' 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 7862: }else{
7863: fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \
7864: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset +1+(i-1)+(nlstate+1)*nlstate,i,cpt );
7865: }
7866: } /* end if covariate */
7867: } /* nlstate */
1.264 brouard 7868: fprintf(ficgp,"\nset out; unset label;\n");
1.223 brouard 7869: } /* end cpt state*/
7870: } /* end covariate */
7871: } /* End if prevfcast */
1.227 brouard 7872:
1.296 brouard 7873: if(prevbcast==1){
1.268 brouard 7874: /* Back projection from cross-sectional to stable (mixed) for each covariate */
7875:
7876: for (k1=1; k1<= m ; k1 ++) /* For each covariate combination if any */
7877: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
7878: if(m != 1 && TKresult[nres]!= k1)
7879: continue;
7880: for (cpt=1; cpt<=nlstate ; cpt ++) { /* For each life state */
7881: strcpy(gplotlabel,"(");
7882: fprintf(ficgp,"\n#\n#\n#Back projection of prevalence to stable (mixed) back prevalence: 'BPROJ_' files, covariatecombination#=%d originstate=%d",k1, cpt);
7883: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
7884: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
7885: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7886: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7887: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7888: vlv= nbcode[Tvaraff[k]][lv];
7889: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
7890: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
7891: }
7892: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
7893: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7894: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
7895: }
7896: strcpy(gplotlabel+strlen(gplotlabel),")");
7897: fprintf(ficgp,"\n#\n");
7898: if(invalidvarcomb[k1]){
7899: fprintf(ficgp,"#Combination (%d) ignored because no cases \n",k1);
7900: continue;
7901: }
7902:
7903: fprintf(ficgp,"# hbijx=backprobability over h years, hb.jx is weighted by observed prev at destination state\n ");
7904: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" \n",subdirf2(optionfilefiname,"PROJB_"),cpt,k1,nres);
7905: fprintf(ficgp,"set label \"Origin alive state %d %s\" at graph 0.98,0.5 center rotate font \"Helvetica,12\"\n",cpt,gplotlabel);
7906: fprintf(ficgp,"set xlabel \"Age\" \nset ylabel \"Prevalence\" \n\
7907: set ter svg size 640, 480\nunset log y\nplot [%.f:%.f] ", ageminpar, agemaxpar);
7908:
7909: /* for (i=1; i<= nlstate+1 ; i ++){ /\* nlstate +1 p11 p21 p.1 *\/ */
7910: istart=nlstate+1; /* Could be one if by state, but nlstate+1 is w.i projection only */
7911: /*istart=1;*/ /* Could be one if by state, but nlstate+1 is w.i projection only */
7912: for (i=istart; i<= nlstate+1 ; i ++){ /* nlstate +1 p11 p21 p.1 */
7913: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7914: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7915: /*# yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7916: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7917: if(i==istart){
7918: fprintf(ficgp,"\"%s\"",subdirf2(fileresu,"FB_"));
7919: }else{
7920: fprintf(ficgp,",\\\n '' ");
7921: }
7922: if(cptcoveff ==0){ /* No covariate */
7923: ioffset=2; /* Age is in 2 */
7924: /*# yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7925: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7926: /*# V1 = 1 yearproj age p11 p21 p31 p.1 p12 p22 p32 p.2 p13 p23 p33 p.3 p14 p24 p34 p.4*/
7927: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 */
7928: fprintf(ficgp," u %d:(", ioffset);
7929: if(i==nlstate+1){
1.270 brouard 7930: fprintf(ficgp," $%d/(1.-$%d)):1 t 'bw%d' with line lc variable ", \
1.268 brouard 7931: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt );
7932: fprintf(ficgp,",\\\n '' ");
7933: fprintf(ficgp," u %d:(",ioffset);
1.270 brouard 7934: fprintf(ficgp," (($1-$2) == %d ) ? $%d : 1/0):1 with labels center not ", \
1.268 brouard 7935: offbyear, \
7936: ioffset+(cpt-1)*(nlstate+1)+1+(i-1) );
7937: }else
7938: fprintf(ficgp," $%d/(1.-$%d)) t 'b%d%d' with line ", \
7939: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), ioffset+1+(i-1)+(nlstate+1)*nlstate,cpt,i );
7940: }else{ /* more than 2 covariates */
1.270 brouard 7941: ioffset=2*cptcoveff+2; /* Age is in 4 or 6 or etc.*/
7942: /*# V1 = 1 V2 = 0 yearproj age p11 p21 p.1 p12 p22 p.2 p13 p23 p.3*/
7943: /*# 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 */
7944: iyearc=ioffset-1;
7945: iagec=ioffset;
1.268 brouard 7946: fprintf(ficgp," u %d:(",ioffset);
7947: kl=0;
7948: strcpy(gplotcondition,"(");
7949: for (k=1; k<=cptcoveff; k++){ /* For each covariate writing the chain of conditions */
7950: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to combination k1 and covariate k */
7951: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
7952: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
7953: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
7954: vlv= nbcode[Tvaraff[k]][lv]; /* Value of the modality of Tvaraff[k] */
7955: kl++;
7956: sprintf(gplotcondition+strlen(gplotcondition),"$%d==%d && $%d==%d " ,kl,Tvaraff[k], kl+1, nbcode[Tvaraff[k]][lv]);
7957: kl++;
7958: if(k <cptcoveff && cptcoveff>1)
7959: sprintf(gplotcondition+strlen(gplotcondition)," && ");
7960: }
7961: strcpy(gplotcondition+strlen(gplotcondition),")");
7962: /* 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 *\/ */
7963: /*6+(cpt-1)*(nlstate+1)+1+(i-1)+(nlstate+1)*nlstate; 6+(1-1)*(2+1)+1+(1-1) +(2+1)*2=13 */
7964: /*6+1+(i-1)+(nlstate+1)*nlstate; 6+1+(1-1) +(2+1)*2=13 */
7965: /* '' 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*/
7966: if(i==nlstate+1){
1.270 brouard 7967: fprintf(ficgp,"%s ? $%d : 1/0):%d t 'bw%d' with line lc variable", gplotcondition, \
7968: ioffset+(cpt-1)*(nlstate+1)+1+(i-1),iyearc,cpt );
1.268 brouard 7969: fprintf(ficgp,",\\\n '' ");
1.270 brouard 7970: fprintf(ficgp," u %d:(",iagec);
1.268 brouard 7971: /* fprintf(ficgp,"%s && (($5-$6) == %d ) ? $%d/(1.-$%d) : 1/0):5 with labels center not ", gplotcondition, \ */
1.270 brouard 7972: fprintf(ficgp,"%s && (($%d-$%d) == %d ) ? $%d : 1/0):%d with labels center not ", gplotcondition, \
7973: iyearc,iagec,offbyear, \
7974: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), iyearc );
1.268 brouard 7975: /* '' u 6:(($1==1 && $2==0 && $3==2 && $4==0) && (($5-$6) == 1947) ? $10/(1.-$22) : 1/0):5 with labels center boxed not*/
7976: }else{
7977: /* fprintf(ficgp,"%s ? $%d/(1.-$%d) : 1/0) t 'p%d%d' with line ", gplotcondition, \ */
7978: fprintf(ficgp,"%s ? $%d : 1/0) t 'b%d%d' with line ", gplotcondition, \
7979: ioffset+(cpt-1)*(nlstate+1)+1+(i-1), cpt,i );
7980: }
7981: } /* end if covariate */
7982: } /* nlstate */
7983: fprintf(ficgp,"\nset out; unset label;\n");
7984: } /* end cpt state*/
7985: } /* end covariate */
1.296 brouard 7986: } /* End if prevbcast */
1.268 brouard 7987:
1.227 brouard 7988:
1.238 brouard 7989: /* 9eme writing MLE parameters */
7990: fprintf(ficgp,"\n##############\n#9eme MLE estimated parameters\n#############\n");
1.126 brouard 7991: for(i=1,jk=1; i <=nlstate; i++){
1.187 brouard 7992: fprintf(ficgp,"# initial state %d\n",i);
1.126 brouard 7993: for(k=1; k <=(nlstate+ndeath); k++){
7994: if (k != i) {
1.227 brouard 7995: fprintf(ficgp,"# current state %d\n",k);
7996: for(j=1; j <=ncovmodel; j++){
7997: fprintf(ficgp,"p%d=%f; ",jk,p[jk]);
7998: jk++;
7999: }
8000: fprintf(ficgp,"\n");
1.126 brouard 8001: }
8002: }
1.223 brouard 8003: }
1.187 brouard 8004: fprintf(ficgp,"##############\n#\n");
1.227 brouard 8005:
1.145 brouard 8006: /*goto avoid;*/
1.238 brouard 8007: /* 10eme Graphics of probabilities or incidences using written MLE parameters */
8008: fprintf(ficgp,"\n##############\n#10eme Graphics of probabilities or incidences\n#############\n");
1.187 brouard 8009: fprintf(ficgp,"# logi(p12/p11)=a12+b12*age+c12age*age+d12*V1+e12*V1*age\n");
8010: fprintf(ficgp,"# logi(p12/p11)=p1 +p2*age +p3*age*age+ p4*V1+ p5*V1*age\n");
8011: fprintf(ficgp,"# logi(p13/p11)=a13+b13*age+c13age*age+d13*V1+e13*V1*age\n");
8012: fprintf(ficgp,"# logi(p13/p11)=p6 +p7*age +p8*age*age+ p9*V1+ p10*V1*age\n");
8013: fprintf(ficgp,"# p12+p13+p14+p11=1=p11(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8014: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8015: fprintf(ficgp,"# p11=1/(1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8016: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age)+...)\n");
8017: fprintf(ficgp,"# p12=exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)/\n");
8018: fprintf(ficgp,"# (1+exp(a12+b12*age+c12age*age+d12*V1+e12*V1*age)\n");
8019: fprintf(ficgp,"# +exp(a13+b13*age+c13age*age+d13*V1+e13*V1*age))\n");
8020: fprintf(ficgp,"# +exp(a14+b14*age+c14age*age+d14*V1+e14*V1*age)+...)\n");
8021: fprintf(ficgp,"#\n");
1.223 brouard 8022: for(ng=1; ng<=3;ng++){ /* Number of graphics: first is logit, 2nd is probabilities, third is incidences per year*/
1.238 brouard 8023: fprintf(ficgp,"#Number of graphics: first is logit, 2nd is probabilities, third is incidences per year\n");
1.237 brouard 8024: fprintf(ficgp,"#model=%s \n",model);
1.238 brouard 8025: fprintf(ficgp,"# Type of graphic ng=%d\n",ng);
1.264 brouard 8026: fprintf(ficgp,"# k1=1 to 2^%d=%d\n",cptcoveff,m);/* to be checked */
8027: for(k1=1; k1 <=m; k1++) /* For each combination of covariate */
1.237 brouard 8028: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.264 brouard 8029: if(m != 1 && TKresult[nres]!= k1)
1.237 brouard 8030: continue;
1.264 brouard 8031: fprintf(ficgp,"\n\n# Combination of dummy k1=%d which is ",k1);
8032: strcpy(gplotlabel,"(");
1.276 brouard 8033: /*sprintf(gplotlabel+strlen(gplotlabel)," Dummy combination %d ",k1);*/
1.264 brouard 8034: for (k=1; k<=cptcoveff; k++){ /* For each correspondig covariate value */
8035: lv= decodtabm(k1,k,cptcoveff); /* Should be the covariate value corresponding to k1 combination and kth covariate */
8036: /* decodtabm(1,1,4) = 1 because h=1 k= (1) 1 1 1 */
8037: /* decodtabm(1,2,4) = 1 because h=1 k= 1 (1) 1 1 */
8038: /* decodtabm(13,3,4)= 2 because h=13 k= 1 1 (2) 2 */
8039: vlv= nbcode[Tvaraff[k]][lv];
8040: fprintf(ficgp," V%d=%d ",Tvaraff[k],vlv);
8041: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%d ",Tvaraff[k],vlv);
8042: }
1.237 brouard 8043: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8044: fprintf(ficgp," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.264 brouard 8045: sprintf(gplotlabel+strlen(gplotlabel)," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.237 brouard 8046: }
1.264 brouard 8047: strcpy(gplotlabel+strlen(gplotlabel),")");
1.237 brouard 8048: fprintf(ficgp,"\n#\n");
1.264 brouard 8049: fprintf(ficgp,"\nset out \"%s_%d-%d-%d.svg\" ",subdirf2(optionfilefiname,"PE_"),k1,ng,nres);
1.276 brouard 8050: fprintf(ficgp,"\nset key outside ");
8051: /* fprintf(ficgp,"\nset label \"%s\" at graph 1.2,0.5 center rotate font \"Helvetica,12\"\n",gplotlabel); */
8052: fprintf(ficgp,"\nset title \"%s\" font \"Helvetica,12\"\n",gplotlabel);
1.223 brouard 8053: fprintf(ficgp,"\nset ter svg size 640, 480 ");
8054: if (ng==1){
8055: fprintf(ficgp,"\nset ylabel \"Value of the logit of the model\"\n"); /* exp(a12+b12*x) could be nice */
8056: fprintf(ficgp,"\nunset log y");
8057: }else if (ng==2){
8058: fprintf(ficgp,"\nset ylabel \"Probability\"\n");
8059: fprintf(ficgp,"\nset log y");
8060: }else if (ng==3){
8061: fprintf(ficgp,"\nset ylabel \"Quasi-incidence per year\"\n");
8062: fprintf(ficgp,"\nset log y");
8063: }else
8064: fprintf(ficgp,"\nunset title ");
8065: fprintf(ficgp,"\nplot [%.f:%.f] ",ageminpar,agemaxpar);
8066: i=1;
8067: for(k2=1; k2<=nlstate; k2++) {
8068: k3=i;
8069: for(k=1; k<=(nlstate+ndeath); k++) {
8070: if (k != k2){
8071: switch( ng) {
8072: case 1:
8073: if(nagesqr==0)
8074: fprintf(ficgp," p%d+p%d*x",i,i+1);
8075: else /* nagesqr =1 */
8076: fprintf(ficgp," p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8077: break;
8078: case 2: /* ng=2 */
8079: if(nagesqr==0)
8080: fprintf(ficgp," exp(p%d+p%d*x",i,i+1);
8081: else /* nagesqr =1 */
8082: fprintf(ficgp," exp(p%d+p%d*x+p%d*x*x",i,i+1,i+1+nagesqr);
8083: break;
8084: case 3:
8085: if(nagesqr==0)
8086: fprintf(ficgp," %f*exp(p%d+p%d*x",YEARM/stepm,i,i+1);
8087: else /* nagesqr =1 */
8088: fprintf(ficgp," %f*exp(p%d+p%d*x+p%d*x*x",YEARM/stepm,i,i+1,i+1+nagesqr);
8089: break;
8090: }
8091: ij=1;/* To be checked else nbcode[0][0] wrong */
1.237 brouard 8092: ijp=1; /* product no age */
8093: /* for(j=3; j <=ncovmodel-nagesqr; j++) { */
8094: for(j=1; j <=cptcovt; j++) { /* For each covariate of the simplified model */
1.223 brouard 8095: /* printf("Tage[%d]=%d, j=%d\n", ij, Tage[ij], j); */
1.268 brouard 8096: if(cptcovage >0){ /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8097: if(j==Tage[ij]) { /* Product by age To be looked at!!*/
8098: if(ij <=cptcovage) { /* V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1, 2 V5 and V1 */
8099: if(DummyV[j]==0){
8100: fprintf(ficgp,"+p%d*%d*x",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]);;
8101: }else{ /* quantitative */
8102: fprintf(ficgp,"+p%d*%f*x",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* Tqinvresult in decoderesult */
8103: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8104: }
8105: ij++;
1.237 brouard 8106: }
1.268 brouard 8107: }
8108: }else if(cptcovprod >0){
8109: if(j==Tprod[ijp]) { /* */
8110: /* printf("Tprod[%d]=%d, j=%d\n", ij, Tprod[ijp], j); */
8111: if(ijp <=cptcovprod) { /* Product */
8112: if(DummyV[Tvard[ijp][1]]==0){/* Vn is dummy */
8113: if(DummyV[Tvard[ijp][2]]==0){/* Vn and Vm are dummy */
8114: /* 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)]); */
8115: fprintf(ficgp,"+p%d*%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tinvresult[nres][Tvard[ijp][2]]);
8116: }else{ /* Vn is dummy and Vm is quanti */
8117: /* fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,nbcode[Tvard[ijp][1]][codtabm(k1,j)],Tqinvresult[nres][Tvard[ijp][2]]); */
8118: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8119: }
8120: }else{ /* Vn*Vm Vn is quanti */
8121: if(DummyV[Tvard[ijp][2]]==0){
8122: fprintf(ficgp,"+p%d*%d*%f",i+j+2+nagesqr-1,Tinvresult[nres][Tvard[ijp][2]],Tqinvresult[nres][Tvard[ijp][1]]);
8123: }else{ /* Both quanti */
8124: fprintf(ficgp,"+p%d*%f*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvard[ijp][1]],Tqinvresult[nres][Tvard[ijp][2]]);
8125: }
1.237 brouard 8126: }
1.268 brouard 8127: ijp++;
1.237 brouard 8128: }
1.268 brouard 8129: } /* end Tprod */
1.237 brouard 8130: } else{ /* simple covariate */
1.264 brouard 8131: /* fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,nbcode[Tvar[j]][codtabm(k1,j)]); /\* Valgrind bug nbcode *\/ */
1.237 brouard 8132: if(Dummy[j]==0){
8133: fprintf(ficgp,"+p%d*%d",i+j+2+nagesqr-1,Tinvresult[nres][Tvar[j]]); /* */
8134: }else{ /* quantitative */
8135: fprintf(ficgp,"+p%d*%f",i+j+2+nagesqr-1,Tqinvresult[nres][Tvar[j]]); /* */
1.264 brouard 8136: /* fprintf(ficgp,"+p%d*%d*x",i+j+nagesqr-1,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
1.223 brouard 8137: }
1.237 brouard 8138: } /* end simple */
8139: } /* end j */
1.223 brouard 8140: }else{
8141: i=i-ncovmodel;
8142: if(ng !=1 ) /* For logit formula of log p11 is more difficult to get */
8143: fprintf(ficgp," (1.");
8144: }
1.227 brouard 8145:
1.223 brouard 8146: if(ng != 1){
8147: fprintf(ficgp,")/(1");
1.227 brouard 8148:
1.264 brouard 8149: for(cpt=1; cpt <=nlstate; cpt++){
1.223 brouard 8150: if(nagesqr==0)
1.264 brouard 8151: fprintf(ficgp,"+exp(p%d+p%d*x",k3+(cpt-1)*ncovmodel,k3+(cpt-1)*ncovmodel+1);
1.223 brouard 8152: else /* nagesqr =1 */
1.264 brouard 8153: 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 8154:
1.223 brouard 8155: ij=1;
8156: for(j=3; j <=ncovmodel-nagesqr; j++){
1.268 brouard 8157: if(cptcovage >0){
8158: if((j-2)==Tage[ij]) { /* Bug valgrind */
8159: if(ij <=cptcovage) { /* Bug valgrind */
8160: fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);
8161: /* fprintf(ficgp,"+p%d*%d*x",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,Tvar[j-2])]); */
8162: ij++;
8163: }
8164: }
8165: }else
8166: fprintf(ficgp,"+p%d*%d",k3+(cpt-1)*ncovmodel+1+j-2+nagesqr,nbcode[Tvar[j-2]][codtabm(k1,j-2)]);/* Valgrind bug nbcode */
1.223 brouard 8167: }
8168: fprintf(ficgp,")");
8169: }
8170: fprintf(ficgp,")");
8171: if(ng ==2)
1.276 brouard 8172: 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 8173: else /* ng= 3 */
1.276 brouard 8174: fprintf(ficgp," w l lw 2 lt (%d*%d+%d)%%%d+1 dt %d t \"i%d%d\" ", nlstate+ndeath, k2, k, nlstate+ndeath, k2, k2,k);
1.223 brouard 8175: }else{ /* end ng <> 1 */
8176: if( k !=k2) /* logit p11 is hard to draw */
1.276 brouard 8177: 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 8178: }
8179: if ((k+k2)!= (nlstate*2+ndeath) && ng != 1)
8180: fprintf(ficgp,",");
8181: if (ng == 1 && k!=k2 && (k+k2)!= (nlstate*2+ndeath))
8182: fprintf(ficgp,",");
8183: i=i+ncovmodel;
8184: } /* end k */
8185: } /* end k2 */
1.276 brouard 8186: /* fprintf(ficgp,"\n set out; unset label;set key default;\n"); */
8187: fprintf(ficgp,"\n set out; unset title;set key default;\n");
1.264 brouard 8188: } /* end k1 */
1.223 brouard 8189: } /* end ng */
8190: /* avoid: */
8191: fflush(ficgp);
1.126 brouard 8192: } /* end gnuplot */
8193:
8194:
8195: /*************** Moving average **************/
1.219 brouard 8196: /* int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav, double bageout, double fageout){ */
1.222 brouard 8197: int movingaverage(double ***probs, double bage, double fage, double ***mobaverage, int mobilav){
1.218 brouard 8198:
1.222 brouard 8199: int i, cpt, cptcod;
8200: int modcovmax =1;
8201: int mobilavrange, mob;
8202: int iage=0;
1.288 brouard 8203: int firstA1=0, firstA2=0;
1.222 brouard 8204:
1.266 brouard 8205: double sum=0., sumr=0.;
1.222 brouard 8206: double age;
1.266 brouard 8207: double *sumnewp, *sumnewm, *sumnewmr;
8208: double *agemingood, *agemaxgood;
8209: double *agemingoodr, *agemaxgoodr;
1.222 brouard 8210:
8211:
1.278 brouard 8212: /* modcovmax=2*cptcoveff; Max number of modalities. We suppose */
8213: /* a covariate has 2 modalities, should be equal to ncovcombmax */
1.222 brouard 8214:
8215: sumnewp = vector(1,ncovcombmax);
8216: sumnewm = vector(1,ncovcombmax);
1.266 brouard 8217: sumnewmr = vector(1,ncovcombmax);
1.222 brouard 8218: agemingood = vector(1,ncovcombmax);
1.266 brouard 8219: agemingoodr = vector(1,ncovcombmax);
1.222 brouard 8220: agemaxgood = vector(1,ncovcombmax);
1.266 brouard 8221: agemaxgoodr = vector(1,ncovcombmax);
1.222 brouard 8222:
8223: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
1.266 brouard 8224: sumnewm[cptcod]=0.; sumnewmr[cptcod]=0.;
1.222 brouard 8225: sumnewp[cptcod]=0.;
1.266 brouard 8226: agemingood[cptcod]=0, agemingoodr[cptcod]=0;
8227: agemaxgood[cptcod]=0, agemaxgoodr[cptcod]=0;
1.222 brouard 8228: }
8229: if (cptcovn<1) ncovcombmax=1; /* At least 1 pass */
8230:
1.266 brouard 8231: if(mobilav==-1 || mobilav==1||mobilav ==3 ||mobilav==5 ||mobilav== 7){
8232: if(mobilav==1 || mobilav==-1) mobilavrange=5; /* default */
1.222 brouard 8233: else mobilavrange=mobilav;
8234: for (age=bage; age<=fage; age++)
8235: for (i=1; i<=nlstate;i++)
8236: for (cptcod=1;cptcod<=ncovcombmax;cptcod++)
8237: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8238: /* We keep the original values on the extreme ages bage, fage and for
8239: fage+1 and bage-1 we use a 3 terms moving average; for fage+2 bage+2
8240: we use a 5 terms etc. until the borders are no more concerned.
8241: */
8242: for (mob=3;mob <=mobilavrange;mob=mob+2){
8243: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){
1.266 brouard 8244: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){
8245: sumnewm[cptcod]=0.;
8246: for (i=1; i<=nlstate;i++){
1.222 brouard 8247: mobaverage[(int)age][i][cptcod] =probs[(int)age][i][cptcod];
8248: for (cpt=1;cpt<=(mob-1)/2;cpt++){
8249: mobaverage[(int)age][i][cptcod] +=probs[(int)age-cpt][i][cptcod];
8250: mobaverage[(int)age][i][cptcod] +=probs[(int)age+cpt][i][cptcod];
8251: }
8252: mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/mob;
1.266 brouard 8253: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8254: } /* end i */
8255: if(sumnewm[cptcod] >1.e-3) mobaverage[(int)age][i][cptcod]=mobaverage[(int)age][i][cptcod]/sumnewm[cptcod]; /* Rescaling to sum one */
8256: } /* end cptcod */
1.222 brouard 8257: }/* end age */
8258: }/* end mob */
1.266 brouard 8259: }else{
8260: printf("Error internal in movingaverage, mobilav=%d.\n",mobilav);
1.222 brouard 8261: return -1;
1.266 brouard 8262: }
8263:
8264: for (cptcod=1;cptcod<=ncovcombmax;cptcod++){ /* for each combination */
1.222 brouard 8265: /* for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ */
8266: if(invalidvarcomb[cptcod]){
8267: printf("\nCombination (%d) ignored because no cases \n",cptcod);
8268: continue;
8269: }
1.219 brouard 8270:
1.266 brouard 8271: for (age=fage-(mob-1)/2; age>=bage+(mob-1)/2; age--){ /*looking for the youngest and oldest good age */
8272: sumnewm[cptcod]=0.;
8273: sumnewmr[cptcod]=0.;
8274: for (i=1; i<=nlstate;i++){
8275: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8276: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8277: }
8278: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8279: agemingoodr[cptcod]=age;
8280: }
8281: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8282: agemingood[cptcod]=age;
8283: }
8284: } /* age */
8285: for (age=bage+(mob-1)/2; age<=fage-(mob-1)/2; age++){ /*looking for the youngest and oldest good age */
1.222 brouard 8286: sumnewm[cptcod]=0.;
1.266 brouard 8287: sumnewmr[cptcod]=0.;
1.222 brouard 8288: for (i=1; i<=nlstate;i++){
8289: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8290: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8291: }
8292: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8293: agemaxgoodr[cptcod]=age;
1.222 brouard 8294: }
8295: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
1.266 brouard 8296: agemaxgood[cptcod]=age;
8297: }
8298: } /* age */
8299: /* Thus we have agemingood and agemaxgood as well as goodr for raw (preobs) */
8300: /* but they will change */
1.288 brouard 8301: firstA1=0;firstA2=0;
1.266 brouard 8302: for (age=fage-(mob-1)/2; age>=bage; age--){/* From oldest to youngest, filling up to the youngest */
8303: sumnewm[cptcod]=0.;
8304: sumnewmr[cptcod]=0.;
8305: for (i=1; i<=nlstate;i++){
8306: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8307: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8308: }
8309: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8310: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good without smoothing */
8311: agemaxgoodr[cptcod]=age; /* age min */
8312: for (i=1; i<=nlstate;i++)
8313: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8314: }else{ /* bad we change the value with the values of good ages */
8315: for (i=1; i<=nlstate;i++){
8316: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgoodr[cptcod]][i][cptcod];
8317: } /* i */
8318: } /* end bad */
8319: }else{
8320: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8321: agemaxgood[cptcod]=age;
8322: }else{ /* bad we change the value with the values of good ages */
8323: for (i=1; i<=nlstate;i++){
8324: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod];
8325: } /* i */
8326: } /* end bad */
8327: }/* end else */
8328: sum=0.;sumr=0.;
8329: for (i=1; i<=nlstate;i++){
8330: sum+=mobaverage[(int)age][i][cptcod];
8331: sumr+=probs[(int)age][i][cptcod];
8332: }
8333: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.288 brouard 8334: if(!firstA1){
8335: firstA1=1;
8336: 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);
8337: }
8338: 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 8339: } /* end bad */
8340: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8341: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.288 brouard 8342: if(!firstA2){
8343: firstA2=1;
8344: 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);
8345: }
8346: 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 8347: } /* end bad */
8348: }/* age */
1.266 brouard 8349:
8350: for (age=bage+(mob-1)/2; age<=fage; age++){/* From youngest, finding the oldest wrong */
1.222 brouard 8351: sumnewm[cptcod]=0.;
1.266 brouard 8352: sumnewmr[cptcod]=0.;
1.222 brouard 8353: for (i=1; i<=nlstate;i++){
8354: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
1.266 brouard 8355: sumnewmr[cptcod]+=probs[(int)age][i][cptcod];
8356: }
8357: if(mobilav==-1){ /* Forcing raw ages if good else agemingood */
8358: if(fabs(sumnewmr[cptcod] - 1.) <= 1.e-3) { /* good */
8359: agemingoodr[cptcod]=age;
8360: for (i=1; i<=nlstate;i++)
8361: mobaverage[(int)age][i][cptcod]=probs[(int)age][i][cptcod];
8362: }else{ /* bad we change the value with the values of good ages */
8363: for (i=1; i<=nlstate;i++){
8364: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingoodr[cptcod]][i][cptcod];
8365: } /* i */
8366: } /* end bad */
8367: }else{
8368: if(fabs(sumnewm[cptcod] - 1.) <= 1.e-3) { /* good */
8369: agemingood[cptcod]=age;
8370: }else{ /* bad */
8371: for (i=1; i<=nlstate;i++){
8372: mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod];
8373: } /* i */
8374: } /* end bad */
8375: }/* end else */
8376: sum=0.;sumr=0.;
8377: for (i=1; i<=nlstate;i++){
8378: sum+=mobaverage[(int)age][i][cptcod];
8379: sumr+=mobaverage[(int)age][i][cptcod];
1.222 brouard 8380: }
1.266 brouard 8381: if(fabs(sum - 1.) > 1.e-3) { /* bad */
1.268 brouard 8382: 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 8383: } /* end bad */
8384: /* else{ /\* We found some ages summing to one, we will smooth the oldest *\/ */
8385: if(fabs(sumr - 1.) > 1.e-3) { /* bad */
1.268 brouard 8386: 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 8387: } /* end bad */
8388: }/* age */
1.266 brouard 8389:
1.222 brouard 8390:
8391: for (age=bage; age<=fage; age++){
1.235 brouard 8392: /* printf("%d %d ", cptcod, (int)age); */
1.222 brouard 8393: sumnewp[cptcod]=0.;
8394: sumnewm[cptcod]=0.;
8395: for (i=1; i<=nlstate;i++){
8396: sumnewp[cptcod]+=probs[(int)age][i][cptcod];
8397: sumnewm[cptcod]+=mobaverage[(int)age][i][cptcod];
8398: /* printf("%.4f %.4f ",probs[(int)age][i][cptcod], mobaverage[(int)age][i][cptcod]); */
8399: }
8400: /* printf("%.4f %.4f \n",sumnewp[cptcod], sumnewm[cptcod]); */
8401: }
8402: /* printf("\n"); */
8403: /* } */
1.266 brouard 8404:
1.222 brouard 8405: /* brutal averaging */
1.266 brouard 8406: /* for (i=1; i<=nlstate;i++){ */
8407: /* for (age=1; age<=bage; age++){ */
8408: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemingood[cptcod]][i][cptcod]; */
8409: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8410: /* } */
8411: /* for (age=fage; age<=AGESUP; age++){ */
8412: /* mobaverage[(int)age][i][cptcod]=mobaverage[(int)agemaxgood[cptcod]][i][cptcod]; */
8413: /* /\* printf("age=%d i=%d cptcod=%d mobaverage=%.4f \n",(int)age,i, cptcod, mobaverage[(int)age][i][cptcod]); *\/ */
8414: /* } */
8415: /* } /\* end i status *\/ */
8416: /* for (i=nlstate+1; i<=nlstate+ndeath;i++){ */
8417: /* for (age=1; age<=AGESUP; age++){ */
8418: /* /\*printf("i=%d, age=%d, cptcod=%d\n",i, (int)age, cptcod);*\/ */
8419: /* mobaverage[(int)age][i][cptcod]=0.; */
8420: /* } */
8421: /* } */
1.222 brouard 8422: }/* end cptcod */
1.266 brouard 8423: free_vector(agemaxgoodr,1, ncovcombmax);
8424: free_vector(agemaxgood,1, ncovcombmax);
8425: free_vector(agemingood,1, ncovcombmax);
8426: free_vector(agemingoodr,1, ncovcombmax);
8427: free_vector(sumnewmr,1, ncovcombmax);
1.222 brouard 8428: free_vector(sumnewm,1, ncovcombmax);
8429: free_vector(sumnewp,1, ncovcombmax);
8430: return 0;
8431: }/* End movingaverage */
1.218 brouard 8432:
1.126 brouard 8433:
1.296 brouard 8434:
1.126 brouard 8435: /************** Forecasting ******************/
1.296 brouard 8436: /* 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)*/
8437: 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){
8438: /* dateintemean, mean date of interviews
8439: dateprojd, year, month, day of starting projection
8440: dateprojf date of end of projection;year of end of projection (same day and month as proj1).
1.126 brouard 8441: agemin, agemax range of age
8442: dateprev1 dateprev2 range of dates during which prevalence is computed
8443: */
1.296 brouard 8444: /* double anprojd, mprojd, jprojd; */
8445: /* double anprojf, mprojf, jprojf; */
1.267 brouard 8446: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
1.126 brouard 8447: double agec; /* generic age */
1.296 brouard 8448: double agelim, ppij, yp,yp1,yp2;
1.126 brouard 8449: double *popeffectif,*popcount;
8450: double ***p3mat;
1.218 brouard 8451: /* double ***mobaverage; */
1.126 brouard 8452: char fileresf[FILENAMELENGTH];
8453:
8454: agelim=AGESUP;
1.211 brouard 8455: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8456: in each health status at the date of interview (if between dateprev1 and dateprev2).
8457: We still use firstpass and lastpass as another selection.
8458: */
1.214 brouard 8459: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8460: /* firstpass, lastpass, stepm, weightopt, model); */
1.126 brouard 8461:
1.201 brouard 8462: strcpy(fileresf,"F_");
8463: strcat(fileresf,fileresu);
1.126 brouard 8464: if((ficresf=fopen(fileresf,"w"))==NULL) {
8465: printf("Problem with forecast resultfile: %s\n", fileresf);
8466: fprintf(ficlog,"Problem with forecast resultfile: %s\n", fileresf);
8467: }
1.235 brouard 8468: printf("\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
8469: fprintf(ficlog,"\nComputing forecasting: result on file '%s', please wait... \n", fileresf);
1.126 brouard 8470:
1.225 brouard 8471: if (cptcoveff==0) ncodemax[cptcoveff]=1;
1.126 brouard 8472:
8473:
8474: stepsize=(int) (stepm+YEARM-1)/YEARM;
8475: if (stepm<=12) stepsize=1;
8476: if(estepm < stepm){
8477: printf ("Problem %d lower than %d\n",estepm, stepm);
8478: }
1.270 brouard 8479: else{
8480: hstepm=estepm;
8481: }
8482: if(estepm > stepm){ /* Yes every two year */
8483: stepsize=2;
8484: }
1.296 brouard 8485: hstepm=hstepm/stepm;
1.126 brouard 8486:
1.296 brouard 8487:
8488: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8489: /* fractional in yp1 *\/ */
8490: /* aintmean=yp; */
8491: /* yp2=modf((yp1*12),&yp); */
8492: /* mintmean=yp; */
8493: /* yp1=modf((yp2*30.5),&yp); */
8494: /* jintmean=yp; */
8495: /* if(jintmean==0) jintmean=1; */
8496: /* if(mintmean==0) mintmean=1; */
1.126 brouard 8497:
1.296 brouard 8498:
8499: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */
8500: /* date2dmy(dateprojd,&jprojd, &mprojd, &anprojd); */
8501: /* date2dmy(dateprojf,&jprojf, &mprojf, &anprojf); */
1.227 brouard 8502: i1=pow(2,cptcoveff);
1.126 brouard 8503: if (cptcovn < 1){i1=1;}
8504:
1.296 brouard 8505: fprintf(ficresf,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.126 brouard 8506:
8507: fprintf(ficresf,"#****** Routine prevforecast **\n");
1.227 brouard 8508:
1.126 brouard 8509: /* if (h==(int)(YEARM*yearp)){ */
1.235 brouard 8510: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8511: for(k=1; k<=i1;k++){
1.253 brouard 8512: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 8513: continue;
1.227 brouard 8514: if(invalidvarcomb[k]){
8515: printf("\nCombination (%d) projection ignored because no cases \n",k);
8516: continue;
8517: }
8518: fprintf(ficresf,"\n#****** hpijx=probability over h years, hp.jx is weighted by observed prev \n#");
8519: for(j=1;j<=cptcoveff;j++) {
8520: fprintf(ficresf," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8521: }
1.235 brouard 8522: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
1.238 brouard 8523: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
1.235 brouard 8524: }
1.227 brouard 8525: fprintf(ficresf," yearproj age");
8526: for(j=1; j<=nlstate+ndeath;j++){
8527: for(i=1; i<=nlstate;i++)
8528: fprintf(ficresf," p%d%d",i,j);
8529: fprintf(ficresf," wp.%d",j);
8530: }
1.296 brouard 8531: for (yearp=0; yearp<=(anprojf-anprojd);yearp +=stepsize) {
1.227 brouard 8532: fprintf(ficresf,"\n");
1.296 brouard 8533: fprintf(ficresf,"\n# Forecasting at date %.lf/%.lf/%.lf ",jprojd,mprojd,anprojd+yearp);
1.270 brouard 8534: /* for (agec=fage; agec>=(ageminpar-1); agec--){ */
8535: for (agec=fage; agec>=(bage); agec--){
1.227 brouard 8536: nhstepm=(int) rint((agelim-agec)*YEARM/stepm);
8537: nhstepm = nhstepm/hstepm;
8538: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8539: oldm=oldms;savm=savms;
1.268 brouard 8540: /* We compute pii at age agec over nhstepm);*/
1.235 brouard 8541: hpxij(p3mat,nhstepm,agec,hstepm,p,nlstate,stepm,oldm,savm, k,nres);
1.268 brouard 8542: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
1.227 brouard 8543: for (h=0; h<=nhstepm; h++){
8544: if (h*hstepm/YEARM*stepm ==yearp) {
1.268 brouard 8545: break;
8546: }
8547: }
8548: fprintf(ficresf,"\n");
8549: for(j=1;j<=cptcoveff;j++)
8550: fprintf(ficresf,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8551: fprintf(ficresf,"%.f %.f ",anprojd+yearp,agec+h*hstepm/YEARM*stepm);
1.268 brouard 8552:
8553: for(j=1; j<=nlstate+ndeath;j++) {
8554: ppij=0.;
8555: for(i=1; i<=nlstate;i++) {
1.278 brouard 8556: if (mobilav>=1)
8557: ppij=ppij+p3mat[i][j][h]*prev[(int)agec][i][k];
8558: else { /* even if mobilav==-1 we use mobaverage, probs may not sums to 1 */
8559: ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k];
8560: }
1.268 brouard 8561: fprintf(ficresf," %.3f", p3mat[i][j][h]);
8562: } /* end i */
8563: fprintf(ficresf," %.3f", ppij);
8564: }/* end j */
1.227 brouard 8565: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8566: } /* end agec */
1.266 brouard 8567: /* diffyear=(int) anproj1+yearp-ageminpar-1; */
8568: /*printf("Prevforecast %d+%d-%d=diffyear=%d\n",(int) anproj1, (int)yearp,(int)ageminpar,(int) anproj1-(int)ageminpar);*/
1.227 brouard 8569: } /* end yearp */
8570: } /* end k */
1.219 brouard 8571:
1.126 brouard 8572: fclose(ficresf);
1.215 brouard 8573: printf("End of Computing forecasting \n");
8574: fprintf(ficlog,"End of Computing forecasting\n");
8575:
1.126 brouard 8576: }
8577:
1.269 brouard 8578: /************** Back Forecasting ******************/
1.296 brouard 8579: /* 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){ */
8580: 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){
8581: /* back1, year, month, day of starting backprojection
1.267 brouard 8582: agemin, agemax range of age
8583: dateprev1 dateprev2 range of dates during which prevalence is computed
1.269 brouard 8584: anback2 year of end of backprojection (same day and month as back1).
8585: prevacurrent and prev are prevalences.
1.267 brouard 8586: */
8587: int yearp, stepsize, hstepm, nhstepm, j, k, cptcod, i, h, i1, k4, nres=0;
8588: double agec; /* generic age */
1.302 brouard 8589: double agelim, ppij, ppi, yp,yp1,yp2; /* ,jintmean,mintmean,aintmean;*/
1.267 brouard 8590: double *popeffectif,*popcount;
8591: double ***p3mat;
8592: /* double ***mobaverage; */
8593: char fileresfb[FILENAMELENGTH];
8594:
1.268 brouard 8595: agelim=AGEINF;
1.267 brouard 8596: /* Compute observed prevalence between dateprev1 and dateprev2 by counting the number of people
8597: in each health status at the date of interview (if between dateprev1 and dateprev2).
8598: We still use firstpass and lastpass as another selection.
8599: */
8600: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint,strstart,\ */
8601: /* firstpass, lastpass, stepm, weightopt, model); */
8602:
8603: /*Do we need to compute prevalence again?*/
8604:
8605: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
8606:
8607: strcpy(fileresfb,"FB_");
8608: strcat(fileresfb,fileresu);
8609: if((ficresfb=fopen(fileresfb,"w"))==NULL) {
8610: printf("Problem with back forecast resultfile: %s\n", fileresfb);
8611: fprintf(ficlog,"Problem with back forecast resultfile: %s\n", fileresfb);
8612: }
8613: printf("\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8614: fprintf(ficlog,"\nComputing back forecasting: result on file '%s', please wait... \n", fileresfb);
8615:
8616: if (cptcoveff==0) ncodemax[cptcoveff]=1;
8617:
8618:
8619: stepsize=(int) (stepm+YEARM-1)/YEARM;
8620: if (stepm<=12) stepsize=1;
8621: if(estepm < stepm){
8622: printf ("Problem %d lower than %d\n",estepm, stepm);
8623: }
1.270 brouard 8624: else{
8625: hstepm=estepm;
8626: }
8627: if(estepm >= stepm){ /* Yes every two year */
8628: stepsize=2;
8629: }
1.267 brouard 8630:
8631: hstepm=hstepm/stepm;
1.296 brouard 8632: /* yp1=modf(dateintmean,&yp);/\* extracts integral of datemean in yp and */
8633: /* fractional in yp1 *\/ */
8634: /* aintmean=yp; */
8635: /* yp2=modf((yp1*12),&yp); */
8636: /* mintmean=yp; */
8637: /* yp1=modf((yp2*30.5),&yp); */
8638: /* jintmean=yp; */
8639: /* if(jintmean==0) jintmean=1; */
8640: /* if(mintmean==0) jintmean=1; */
1.267 brouard 8641:
8642: i1=pow(2,cptcoveff);
8643: if (cptcovn < 1){i1=1;}
8644:
1.296 brouard 8645: fprintf(ficresfb,"# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
8646: printf("# Mean day of interviews %.lf/%.lf/%.lf (%.2f) between %.2f and %.2f \n",jintmean,mintmean,aintmean,dateintmean,dateprev1,dateprev2);
1.267 brouard 8647:
8648: fprintf(ficresfb,"#****** Routine prevbackforecast **\n");
8649:
8650: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8651: for(k=1; k<=i1;k++){
8652: if(i1 != 1 && TKresult[nres]!= k)
8653: continue;
8654: if(invalidvarcomb[k]){
8655: printf("\nCombination (%d) projection ignored because no cases \n",k);
8656: continue;
8657: }
1.268 brouard 8658: fprintf(ficresfb,"\n#****** hbijx=probability over h years, hb.jx is weighted by observed prev \n#");
1.267 brouard 8659: for(j=1;j<=cptcoveff;j++) {
8660: fprintf(ficresfb," V%d (=) %d",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8661: }
8662: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
8663: fprintf(ficresf," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
8664: }
8665: fprintf(ficresfb," yearbproj age");
8666: for(j=1; j<=nlstate+ndeath;j++){
8667: for(i=1; i<=nlstate;i++)
1.268 brouard 8668: fprintf(ficresfb," b%d%d",i,j);
8669: fprintf(ficresfb," b.%d",j);
1.267 brouard 8670: }
1.296 brouard 8671: for (yearp=0; yearp>=(anbackf-anbackd);yearp -=stepsize) {
1.267 brouard 8672: /* for (yearp=0; yearp<=(anproj2-anproj1);yearp +=stepsize) { */
8673: fprintf(ficresfb,"\n");
1.296 brouard 8674: fprintf(ficresfb,"\n# Back Forecasting at date %.lf/%.lf/%.lf ",jbackd,mbackd,anbackd+yearp);
1.273 brouard 8675: /* printf("\n# Back Forecasting at date %.lf/%.lf/%.lf ",jback1,mback1,anback1+yearp); */
1.270 brouard 8676: /* for (agec=bage; agec<=agemax-1; agec++){ /\* testing *\/ */
8677: for (agec=bage; agec<=fage; agec++){ /* testing */
1.268 brouard 8678: /* We compute bij at age agec over nhstepm, nhstepm decreases when agec increases because of agemax;*/
1.271 brouard 8679: nhstepm=(int) (agec-agelim) *YEARM/stepm;/* nhstepm=(int) rint((agec-agelim)*YEARM/stepm);*/
1.267 brouard 8680: nhstepm = nhstepm/hstepm;
8681: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8682: oldm=oldms;savm=savms;
1.268 brouard 8683: /* computes hbxij at age agec over 1 to nhstepm */
1.271 brouard 8684: /* printf("####prevbackforecast debug agec=%.2f nhstepm=%d\n",agec, nhstepm);fflush(stdout); */
1.267 brouard 8685: hbxij(p3mat,nhstepm,agec,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.268 brouard 8686: /* hpxij(p3mat,nhstepm,agec,hstepm,p, nlstate,stepm,oldm,savm, k,nres); */
8687: /* Then we print p3mat for h corresponding to the right agec+h*stepms=yearp */
8688: /* printf(" agec=%.2f\n",agec);fflush(stdout); */
1.267 brouard 8689: for (h=0; h<=nhstepm; h++){
1.268 brouard 8690: if (h*hstepm/YEARM*stepm ==-yearp) {
8691: break;
8692: }
8693: }
8694: fprintf(ficresfb,"\n");
8695: for(j=1;j<=cptcoveff;j++)
8696: fprintf(ficresfb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.296 brouard 8697: fprintf(ficresfb,"%.f %.f ",anbackd+yearp,agec-h*hstepm/YEARM*stepm);
1.268 brouard 8698: for(i=1; i<=nlstate+ndeath;i++) {
8699: ppij=0.;ppi=0.;
8700: for(j=1; j<=nlstate;j++) {
8701: /* if (mobilav==1) */
1.269 brouard 8702: ppij=ppij+p3mat[i][j][h]*prevacurrent[(int)agec][j][k];
8703: ppi=ppi+prevacurrent[(int)agec][j][k];
8704: /* ppij=ppij+p3mat[i][j][h]*mobaverage[(int)agec][j][k]; */
8705: /* ppi=ppi+mobaverage[(int)agec][j][k]; */
1.267 brouard 8706: /* else { */
8707: /* ppij=ppij+p3mat[i][j][h]*probs[(int)(agec)][i][k]; */
8708: /* } */
1.268 brouard 8709: fprintf(ficresfb," %.3f", p3mat[i][j][h]);
8710: } /* end j */
8711: if(ppi <0.99){
8712: printf("Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8713: fprintf(ficlog,"Error in prevbackforecast, prevalence doesn't sum to 1 for state %d: %3f\n",i, ppi);
8714: }
8715: fprintf(ficresfb," %.3f", ppij);
8716: }/* end j */
1.267 brouard 8717: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
8718: } /* end agec */
8719: } /* end yearp */
8720: } /* end k */
1.217 brouard 8721:
1.267 brouard 8722: /* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
1.217 brouard 8723:
1.267 brouard 8724: fclose(ficresfb);
8725: printf("End of Computing Back forecasting \n");
8726: fprintf(ficlog,"End of Computing Back forecasting\n");
1.218 brouard 8727:
1.267 brouard 8728: }
1.217 brouard 8729:
1.269 brouard 8730: /* Variance of prevalence limit: varprlim */
8731: 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 8732: /*------- Variance of forward period (stable) prevalence------*/
1.269 brouard 8733:
8734: char fileresvpl[FILENAMELENGTH];
8735: FILE *ficresvpl;
8736: double **oldm, **savm;
8737: double **varpl; /* Variances of prevalence limits by age */
8738: int i1, k, nres, j ;
8739:
8740: strcpy(fileresvpl,"VPL_");
8741: strcat(fileresvpl,fileresu);
8742: if((ficresvpl=fopen(fileresvpl,"w"))==NULL) {
1.288 brouard 8743: printf("Problem with variance of forward period (stable) prevalence resultfile: %s\n", fileresvpl);
1.269 brouard 8744: exit(0);
8745: }
1.288 brouard 8746: printf("Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(stdout);
8747: fprintf(ficlog, "Computing Variance-covariance of forward period (stable) prevalence: file '%s' ...", fileresvpl);fflush(ficlog);
1.269 brouard 8748:
8749: /*for(cptcov=1,k=0;cptcov<=i1;cptcov++){
8750: for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*/
8751:
8752: i1=pow(2,cptcoveff);
8753: if (cptcovn < 1){i1=1;}
8754:
8755: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8756: for(k=1; k<=i1;k++){
8757: if(i1 != 1 && TKresult[nres]!= k)
8758: continue;
8759: fprintf(ficresvpl,"\n#****** ");
8760: printf("\n#****** ");
8761: fprintf(ficlog,"\n#****** ");
8762: for(j=1;j<=cptcoveff;j++) {
8763: fprintf(ficresvpl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8764: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8765: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8766: }
8767: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8768: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8769: fprintf(ficresvpl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8770: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8771: }
8772: fprintf(ficresvpl,"******\n");
8773: printf("******\n");
8774: fprintf(ficlog,"******\n");
8775:
8776: varpl=matrix(1,nlstate,(int) bage, (int) fage);
8777: oldm=oldms;savm=savms;
8778: varprevlim(fileresvpl, ficresvpl, varpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, prlim, ftolpl, ncvyearp, k, strstart, nres);
8779: free_matrix(varpl,1,nlstate,(int) bage, (int)fage);
8780: /*}*/
8781: }
8782:
8783: fclose(ficresvpl);
1.288 brouard 8784: printf("done variance-covariance of forward period prevalence\n");fflush(stdout);
8785: fprintf(ficlog,"done variance-covariance of forward period prevalence\n");fflush(ficlog);
1.269 brouard 8786:
8787: }
8788: /* Variance of back prevalence: varbprlim */
8789: 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){
8790: /*------- Variance of back (stable) prevalence------*/
8791:
8792: char fileresvbl[FILENAMELENGTH];
8793: FILE *ficresvbl;
8794:
8795: double **oldm, **savm;
8796: double **varbpl; /* Variances of back prevalence limits by age */
8797: int i1, k, nres, j ;
8798:
8799: strcpy(fileresvbl,"VBL_");
8800: strcat(fileresvbl,fileresu);
8801: if((ficresvbl=fopen(fileresvbl,"w"))==NULL) {
8802: printf("Problem with variance of back (stable) prevalence resultfile: %s\n", fileresvbl);
8803: exit(0);
8804: }
8805: printf("Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(stdout);
8806: fprintf(ficlog, "Computing Variance-covariance of back (stable) prevalence: file '%s' ...", fileresvbl);fflush(ficlog);
8807:
8808:
8809: i1=pow(2,cptcoveff);
8810: if (cptcovn < 1){i1=1;}
8811:
8812: for(nres=1; nres <= nresult; nres++) /* For each resultline */
8813: for(k=1; k<=i1;k++){
8814: if(i1 != 1 && TKresult[nres]!= k)
8815: continue;
8816: fprintf(ficresvbl,"\n#****** ");
8817: printf("\n#****** ");
8818: fprintf(ficlog,"\n#****** ");
8819: for(j=1;j<=cptcoveff;j++) {
8820: fprintf(ficresvbl,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8821: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8822: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
8823: }
8824: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
8825: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8826: fprintf(ficresvbl," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8827: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
8828: }
8829: fprintf(ficresvbl,"******\n");
8830: printf("******\n");
8831: fprintf(ficlog,"******\n");
8832:
8833: varbpl=matrix(1,nlstate,(int) bage, (int) fage);
8834: oldm=oldms;savm=savms;
8835:
8836: varbrevlim(fileresvbl, ficresvbl, varbpl, matcov, p, delti, nlstate, stepm, (int) bage, (int) fage, oldm, savm, bprlim, ftolpl, mobilavproj, ncvyearp, k, strstart, nres);
8837: free_matrix(varbpl,1,nlstate,(int) bage, (int)fage);
8838: /*}*/
8839: }
8840:
8841: fclose(ficresvbl);
8842: printf("done variance-covariance of back prevalence\n");fflush(stdout);
8843: fprintf(ficlog,"done variance-covariance of back prevalence\n");fflush(ficlog);
8844:
8845: } /* End of varbprlim */
8846:
1.126 brouard 8847: /************** Forecasting *****not tested NB*************/
1.227 brouard 8848: /* 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 8849:
1.227 brouard 8850: /* int cpt, stepsize, hstepm, nhstepm, j,k,c, cptcod, i,h; */
8851: /* int *popage; */
8852: /* double calagedatem, agelim, kk1, kk2; */
8853: /* double *popeffectif,*popcount; */
8854: /* double ***p3mat,***tabpop,***tabpopprev; */
8855: /* /\* double ***mobaverage; *\/ */
8856: /* char filerespop[FILENAMELENGTH]; */
1.126 brouard 8857:
1.227 brouard 8858: /* tabpop= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8859: /* tabpopprev= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
8860: /* agelim=AGESUP; */
8861: /* calagedatem=(anpyram+mpyram/12.+jpyram/365.-dateintmean)*YEARM; */
1.126 brouard 8862:
1.227 brouard 8863: /* prevalence(probs, ageminpar, agemax, s, agev, nlstate, imx, Tvar, nbcode, ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass); */
1.126 brouard 8864:
8865:
1.227 brouard 8866: /* strcpy(filerespop,"POP_"); */
8867: /* strcat(filerespop,fileresu); */
8868: /* if((ficrespop=fopen(filerespop,"w"))==NULL) { */
8869: /* printf("Problem with forecast resultfile: %s\n", filerespop); */
8870: /* fprintf(ficlog,"Problem with forecast resultfile: %s\n", filerespop); */
8871: /* } */
8872: /* printf("Computing forecasting: result on file '%s' \n", filerespop); */
8873: /* fprintf(ficlog,"Computing forecasting: result on file '%s' \n", filerespop); */
1.126 brouard 8874:
1.227 brouard 8875: /* if (cptcoveff==0) ncodemax[cptcoveff]=1; */
1.126 brouard 8876:
1.227 brouard 8877: /* /\* if (mobilav!=0) { *\/ */
8878: /* /\* mobaverage= ma3x(1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
8879: /* /\* if (movingaverage(probs, ageminpar, fage, mobaverage,mobilav)!=0){ *\/ */
8880: /* /\* fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8881: /* /\* printf(" Error in movingaverage mobilav=%d\n",mobilav); *\/ */
8882: /* /\* } *\/ */
8883: /* /\* } *\/ */
1.126 brouard 8884:
1.227 brouard 8885: /* stepsize=(int) (stepm+YEARM-1)/YEARM; */
8886: /* if (stepm<=12) stepsize=1; */
1.126 brouard 8887:
1.227 brouard 8888: /* agelim=AGESUP; */
1.126 brouard 8889:
1.227 brouard 8890: /* hstepm=1; */
8891: /* hstepm=hstepm/stepm; */
1.218 brouard 8892:
1.227 brouard 8893: /* if (popforecast==1) { */
8894: /* if((ficpop=fopen(popfile,"r"))==NULL) { */
8895: /* printf("Problem with population file : %s\n",popfile);exit(0); */
8896: /* fprintf(ficlog,"Problem with population file : %s\n",popfile);exit(0); */
8897: /* } */
8898: /* popage=ivector(0,AGESUP); */
8899: /* popeffectif=vector(0,AGESUP); */
8900: /* popcount=vector(0,AGESUP); */
1.126 brouard 8901:
1.227 brouard 8902: /* i=1; */
8903: /* while ((c=fscanf(ficpop,"%d %lf\n",&popage[i],&popcount[i])) != EOF) i=i+1; */
1.218 brouard 8904:
1.227 brouard 8905: /* imx=i; */
8906: /* for (i=1; i<imx;i++) popeffectif[popage[i]]=popcount[i]; */
8907: /* } */
1.218 brouard 8908:
1.227 brouard 8909: /* for(cptcov=1,k=0;cptcov<=i2;cptcov++){ */
8910: /* for(cptcod=1;cptcod<=ncodemax[cptcoveff];cptcod++){ */
8911: /* k=k+1; */
8912: /* fprintf(ficrespop,"\n#******"); */
8913: /* for(j=1;j<=cptcoveff;j++) { */
8914: /* fprintf(ficrespop," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */
8915: /* } */
8916: /* fprintf(ficrespop,"******\n"); */
8917: /* fprintf(ficrespop,"# Age"); */
8918: /* for(j=1; j<=nlstate+ndeath;j++) fprintf(ficrespop," P.%d",j); */
8919: /* if (popforecast==1) fprintf(ficrespop," [Population]"); */
1.126 brouard 8920:
1.227 brouard 8921: /* for (cpt=0; cpt<=0;cpt++) { */
8922: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
1.126 brouard 8923:
1.227 brouard 8924: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8925: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8926: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8927:
1.227 brouard 8928: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8929: /* oldm=oldms;savm=savms; */
8930: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.218 brouard 8931:
1.227 brouard 8932: /* for (h=0; h<=nhstepm; h++){ */
8933: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8934: /* fprintf(ficrespop,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8935: /* } */
8936: /* for(j=1; j<=nlstate+ndeath;j++) { */
8937: /* kk1=0.;kk2=0; */
8938: /* for(i=1; i<=nlstate;i++) { */
8939: /* if (mobilav==1) */
8940: /* kk1=kk1+p3mat[i][j][h]*mobaverage[(int)agedeb+1][i][cptcod]; */
8941: /* else { */
8942: /* kk1=kk1+p3mat[i][j][h]*probs[(int)(agedeb+1)][i][cptcod]; */
8943: /* } */
8944: /* } */
8945: /* if (h==(int)(calagedatem+12*cpt)){ */
8946: /* tabpop[(int)(agedeb)][j][cptcod]=kk1; */
8947: /* /\*fprintf(ficrespop," %.3f", kk1); */
8948: /* if (popforecast==1) fprintf(ficrespop," [%.f]", kk1*popeffectif[(int)agedeb+1]);*\/ */
8949: /* } */
8950: /* } */
8951: /* for(i=1; i<=nlstate;i++){ */
8952: /* kk1=0.; */
8953: /* for(j=1; j<=nlstate;j++){ */
8954: /* kk1= kk1+tabpop[(int)(agedeb)][j][cptcod]; */
8955: /* } */
8956: /* tabpopprev[(int)(agedeb)][i][cptcod]=tabpop[(int)(agedeb)][i][cptcod]/kk1*popeffectif[(int)(agedeb+(calagedatem+12*cpt)*hstepm/YEARM*stepm-1)]; */
8957: /* } */
1.218 brouard 8958:
1.227 brouard 8959: /* if (h==(int)(calagedatem+12*cpt)) */
8960: /* for(j=1; j<=nlstate;j++) */
8961: /* fprintf(ficrespop," %15.2f",tabpopprev[(int)(agedeb+1)][j][cptcod]); */
8962: /* } */
8963: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8964: /* } */
8965: /* } */
1.218 brouard 8966:
1.227 brouard 8967: /* /\******\/ */
1.218 brouard 8968:
1.227 brouard 8969: /* for (cpt=1; cpt<=(anpyram1-anpyram);cpt++) { */
8970: /* fprintf(ficrespop,"\n\n# Forecasting at date %.lf/%.lf/%.lf ",jpyram,mpyram,anpyram+cpt); */
8971: /* for (agedeb=(fage-((int)calagedatem %12/12.)); agedeb>=(ageminpar-((int)calagedatem %12)/12.); agedeb--){ */
8972: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); */
8973: /* nhstepm = nhstepm/hstepm; */
1.126 brouard 8974:
1.227 brouard 8975: /* p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8976: /* oldm=oldms;savm=savms; */
8977: /* hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
8978: /* for (h=0; h<=nhstepm; h++){ */
8979: /* if (h==(int) (calagedatem+YEARM*cpt)) { */
8980: /* fprintf(ficresf,"\n %3.f ",agedeb+h*hstepm/YEARM*stepm); */
8981: /* } */
8982: /* for(j=1; j<=nlstate+ndeath;j++) { */
8983: /* kk1=0.;kk2=0; */
8984: /* for(i=1; i<=nlstate;i++) { */
8985: /* kk1=kk1+p3mat[i][j][h]*tabpopprev[(int)agedeb+1][i][cptcod]; */
8986: /* } */
8987: /* if (h==(int)(calagedatem+12*cpt)) fprintf(ficresf," %15.2f", kk1); */
8988: /* } */
8989: /* } */
8990: /* free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); */
8991: /* } */
8992: /* } */
8993: /* } */
8994: /* } */
1.218 brouard 8995:
1.227 brouard 8996: /* /\* if (mobilav!=0) free_ma3x(mobaverage,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); *\/ */
1.218 brouard 8997:
1.227 brouard 8998: /* if (popforecast==1) { */
8999: /* free_ivector(popage,0,AGESUP); */
9000: /* free_vector(popeffectif,0,AGESUP); */
9001: /* free_vector(popcount,0,AGESUP); */
9002: /* } */
9003: /* free_ma3x(tabpop,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9004: /* free_ma3x(tabpopprev,1, AGESUP,1,NCOVMAX, 1,NCOVMAX); */
9005: /* fclose(ficrespop); */
9006: /* } /\* End of popforecast *\/ */
1.218 brouard 9007:
1.126 brouard 9008: int fileappend(FILE *fichier, char *optionfich)
9009: {
9010: if((fichier=fopen(optionfich,"a"))==NULL) {
9011: printf("Problem with file: %s\n", optionfich);
9012: fprintf(ficlog,"Problem with file: %s\n", optionfich);
9013: return (0);
9014: }
9015: fflush(fichier);
9016: return (1);
9017: }
9018:
9019:
9020: /**************** function prwizard **********************/
9021: void prwizard(int ncovmodel, int nlstate, int ndeath, char model[], FILE *ficparo)
9022: {
9023:
9024: /* Wizard to print covariance matrix template */
9025:
1.164 brouard 9026: char ca[32], cb[32];
9027: int i,j, k, li, lj, lk, ll, jj, npar, itimes;
1.126 brouard 9028: int numlinepar;
9029:
9030: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9031: fprintf(ficparo,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
9032: for(i=1; i <=nlstate; i++){
9033: jj=0;
9034: for(j=1; j <=nlstate+ndeath; j++){
9035: if(j==i) continue;
9036: jj++;
9037: /*ca[0]= k+'a'-1;ca[1]='\0';*/
9038: printf("%1d%1d",i,j);
9039: fprintf(ficparo,"%1d%1d",i,j);
9040: for(k=1; k<=ncovmodel;k++){
9041: /* printf(" %lf",param[i][j][k]); */
9042: /* fprintf(ficparo," %lf",param[i][j][k]); */
9043: printf(" 0.");
9044: fprintf(ficparo," 0.");
9045: }
9046: printf("\n");
9047: fprintf(ficparo,"\n");
9048: }
9049: }
9050: printf("# Scales (for hessian or gradient estimation)\n");
9051: fprintf(ficparo,"# Scales (for hessian or gradient estimation)\n");
9052: npar= (nlstate+ndeath-1)*nlstate*ncovmodel; /* Number of parameters*/
9053: for(i=1; i <=nlstate; i++){
9054: jj=0;
9055: for(j=1; j <=nlstate+ndeath; j++){
9056: if(j==i) continue;
9057: jj++;
9058: fprintf(ficparo,"%1d%1d",i,j);
9059: printf("%1d%1d",i,j);
9060: fflush(stdout);
9061: for(k=1; k<=ncovmodel;k++){
9062: /* printf(" %le",delti3[i][j][k]); */
9063: /* fprintf(ficparo," %le",delti3[i][j][k]); */
9064: printf(" 0.");
9065: fprintf(ficparo," 0.");
9066: }
9067: numlinepar++;
9068: printf("\n");
9069: fprintf(ficparo,"\n");
9070: }
9071: }
9072: printf("# Covariance matrix\n");
9073: /* # 121 Var(a12)\n\ */
9074: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9075: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
9076: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
9077: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
9078: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
9079: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
9080: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9081: fflush(stdout);
9082: fprintf(ficparo,"# Covariance matrix\n");
9083: /* # 121 Var(a12)\n\ */
9084: /* # 122 Cov(b12,a12) Var(b12)\n\ */
9085: /* # ...\n\ */
9086: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
9087:
9088: for(itimes=1;itimes<=2;itimes++){
9089: jj=0;
9090: for(i=1; i <=nlstate; i++){
9091: for(j=1; j <=nlstate+ndeath; j++){
9092: if(j==i) continue;
9093: for(k=1; k<=ncovmodel;k++){
9094: jj++;
9095: ca[0]= k+'a'-1;ca[1]='\0';
9096: if(itimes==1){
9097: printf("#%1d%1d%d",i,j,k);
9098: fprintf(ficparo,"#%1d%1d%d",i,j,k);
9099: }else{
9100: printf("%1d%1d%d",i,j,k);
9101: fprintf(ficparo,"%1d%1d%d",i,j,k);
9102: /* printf(" %.5le",matcov[i][j]); */
9103: }
9104: ll=0;
9105: for(li=1;li <=nlstate; li++){
9106: for(lj=1;lj <=nlstate+ndeath; lj++){
9107: if(lj==li) continue;
9108: for(lk=1;lk<=ncovmodel;lk++){
9109: ll++;
9110: if(ll<=jj){
9111: cb[0]= lk +'a'-1;cb[1]='\0';
9112: if(ll<jj){
9113: if(itimes==1){
9114: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9115: fprintf(ficparo," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
9116: }else{
9117: printf(" 0.");
9118: fprintf(ficparo," 0.");
9119: }
9120: }else{
9121: if(itimes==1){
9122: printf(" Var(%s%1d%1d)",ca,i,j);
9123: fprintf(ficparo," Var(%s%1d%1d)",ca,i,j);
9124: }else{
9125: printf(" 0.");
9126: fprintf(ficparo," 0.");
9127: }
9128: }
9129: }
9130: } /* end lk */
9131: } /* end lj */
9132: } /* end li */
9133: printf("\n");
9134: fprintf(ficparo,"\n");
9135: numlinepar++;
9136: } /* end k*/
9137: } /*end j */
9138: } /* end i */
9139: } /* end itimes */
9140:
9141: } /* end of prwizard */
9142: /******************* Gompertz Likelihood ******************************/
9143: double gompertz(double x[])
9144: {
1.302 brouard 9145: double A=0.0,B=0.,L=0.0,sump=0.,num=0.;
1.126 brouard 9146: int i,n=0; /* n is the size of the sample */
9147:
1.220 brouard 9148: for (i=1;i<=imx ; i++) {
1.126 brouard 9149: sump=sump+weight[i];
9150: /* sump=sump+1;*/
9151: num=num+1;
9152: }
1.302 brouard 9153: L=0.0;
9154: /* agegomp=AGEGOMP; */
1.126 brouard 9155: /* for (i=0; i<=imx; i++)
9156: 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]);*/
9157:
1.302 brouard 9158: for (i=1;i<=imx ; i++) {
9159: /* mu(a)=mu(agecomp)*exp(teta*(age-agegomp))
9160: mu(a)=x[1]*exp(x[2]*(age-agegomp)); x[1] and x[2] are per year.
9161: * L= Product mu(agedeces)exp(-\int_ageexam^agedc mu(u) du ) for a death between agedc (in month)
9162: * and agedc +1 month, cens[i]=0: log(x[1]/YEARM)
9163: * +
9164: * exp(-\int_ageexam^agecens mu(u) du ) when censored, cens[i]=1
9165: */
9166: if (wav[i] > 1 || agedc[i] < AGESUP) {
9167: if (cens[i] == 1){
9168: A=-x[1]/(x[2])*(exp(x[2]*(agecens[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)));
9169: } else if (cens[i] == 0){
1.126 brouard 9170: A=-x[1]/(x[2])*(exp(x[2]*(agedc[i]-agegomp))-exp(x[2]*(ageexmed[i]-agegomp)))
1.302 brouard 9171: +log(x[1]/YEARM) +x[2]*(agedc[i]-agegomp)+log(YEARM);
9172: } else
9173: printf("Gompertz cens[%d] neither 1 nor 0\n",i);
1.126 brouard 9174: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
1.302 brouard 9175: L=L+A*weight[i];
1.126 brouard 9176: /* 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 9177: }
9178: }
1.126 brouard 9179:
1.302 brouard 9180: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
1.126 brouard 9181:
9182: return -2*L*num/sump;
9183: }
9184:
1.136 brouard 9185: #ifdef GSL
9186: /******************* Gompertz_f Likelihood ******************************/
9187: double gompertz_f(const gsl_vector *v, void *params)
9188: {
1.302 brouard 9189: double A=0.,B=0.,LL=0.0,sump=0.,num=0.;
1.136 brouard 9190: double *x= (double *) v->data;
9191: int i,n=0; /* n is the size of the sample */
9192:
9193: for (i=0;i<=imx-1 ; i++) {
9194: sump=sump+weight[i];
9195: /* sump=sump+1;*/
9196: num=num+1;
9197: }
9198:
9199:
9200: /* for (i=0; i<=imx; i++)
9201: 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]);*/
9202: printf("x[0]=%lf x[1]=%lf\n",x[0],x[1]);
9203: for (i=1;i<=imx ; i++)
9204: {
9205: if (cens[i] == 1 && wav[i]>1)
9206: A=-x[0]/(x[1])*(exp(x[1]*(agecens[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)));
9207:
9208: if (cens[i] == 0 && wav[i]>1)
9209: A=-x[0]/(x[1])*(exp(x[1]*(agedc[i]-agegomp))-exp(x[1]*(ageexmed[i]-agegomp)))
9210: +log(x[0]/YEARM)+x[1]*(agedc[i]-agegomp)+log(YEARM);
9211:
9212: /*if (wav[i] > 1 && agecens[i] > 15) {*/ /* ??? */
9213: if (wav[i] > 1 ) { /* ??? */
9214: LL=LL+A*weight[i];
9215: /* 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]);*/
9216: }
9217: }
9218:
9219: /*printf("x1=%2.9f x2=%2.9f x3=%2.9f L=%f\n",x[1],x[2],x[3],L);*/
9220: printf("x[0]=%lf x[1]=%lf -2*LL*num/sump=%lf\n",x[0],x[1],-2*LL*num/sump);
9221:
9222: return -2*LL*num/sump;
9223: }
9224: #endif
9225:
1.126 brouard 9226: /******************* Printing html file ***********/
1.201 brouard 9227: void printinghtmlmort(char fileresu[], char title[], char datafile[], int firstpass, \
1.126 brouard 9228: int lastpass, int stepm, int weightopt, char model[],\
9229: int imx, double p[],double **matcov,double agemortsup){
9230: int i,k;
9231:
9232: fprintf(fichtm,"<ul><li><h4>Result files </h4>\n Force of mortality. Parameters of the Gompertz fit (with confidence interval in brackets):<br>");
9233: fprintf(fichtm," mu(age) =%lf*exp(%lf*(age-%d)) per year<br><br>",p[1],p[2],agegomp);
9234: for (i=1;i<=2;i++)
9235: 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 9236: fprintf(fichtm,"<br><br><img src=\"graphmort.svg\">");
1.126 brouard 9237: fprintf(fichtm,"</ul>");
9238:
9239: fprintf(fichtm,"<ul><li><h4>Life table</h4>\n <br>");
9240:
9241: 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>");
9242:
9243: for (k=agegomp;k<(agemortsup-2);k++)
9244: 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]);
9245:
9246:
9247: fflush(fichtm);
9248: }
9249:
9250: /******************* Gnuplot file **************/
1.201 brouard 9251: void printinggnuplotmort(char fileresu[], char optionfilefiname[], double ageminpar, double agemaxpar, double fage , char pathc[], double p[]){
1.126 brouard 9252:
9253: char dirfileres[132],optfileres[132];
1.164 brouard 9254:
1.126 brouard 9255: int ng;
9256:
9257:
9258: /*#ifdef windows */
9259: fprintf(ficgp,"cd \"%s\" \n",pathc);
9260: /*#endif */
9261:
9262:
9263: strcpy(dirfileres,optionfilefiname);
9264: strcpy(optfileres,"vpl");
1.199 brouard 9265: fprintf(ficgp,"set out \"graphmort.svg\"\n ");
1.126 brouard 9266: fprintf(ficgp,"set xlabel \"Age\"\n set ylabel \"Force of mortality (per year)\" \n ");
1.199 brouard 9267: fprintf(ficgp, "set ter svg size 640, 480\n set log y\n");
1.145 brouard 9268: /* fprintf(ficgp, "set size 0.65,0.65\n"); */
1.126 brouard 9269: fprintf(ficgp,"plot [%d:100] %lf*exp(%lf*(x-%d))",agegomp,p[1],p[2],agegomp);
9270:
9271: }
9272:
1.136 brouard 9273: int readdata(char datafile[], int firstobs, int lastobs, int *imax)
9274: {
1.126 brouard 9275:
1.136 brouard 9276: /*-------- data file ----------*/
9277: FILE *fic;
9278: char dummy[]=" ";
1.240 brouard 9279: int i=0, j=0, n=0, iv=0, v;
1.223 brouard 9280: int lstra;
1.136 brouard 9281: int linei, month, year,iout;
1.302 brouard 9282: int noffset=0; /* This is the offset if BOM data file */
1.136 brouard 9283: char line[MAXLINE], linetmp[MAXLINE];
1.164 brouard 9284: char stra[MAXLINE], strb[MAXLINE];
1.136 brouard 9285: char *stratrunc;
1.223 brouard 9286:
1.240 brouard 9287: DummyV=ivector(1,NCOVMAX); /* 1 to 3 */
9288: FixedV=ivector(1,NCOVMAX); /* 1 to 3 */
1.126 brouard 9289:
1.240 brouard 9290: for(v=1; v <=ncovcol;v++){
9291: DummyV[v]=0;
9292: FixedV[v]=0;
9293: }
9294: for(v=ncovcol+1; v <=ncovcol+nqv;v++){
9295: DummyV[v]=1;
9296: FixedV[v]=0;
9297: }
9298: for(v=ncovcol+nqv+1; v <=ncovcol+nqv+ntv;v++){
9299: DummyV[v]=0;
9300: FixedV[v]=1;
9301: }
9302: for(v=ncovcol+nqv+ntv+1; v <=ncovcol+nqv+ntv+nqtv;v++){
9303: DummyV[v]=1;
9304: FixedV[v]=1;
9305: }
9306: for(v=1; v <=ncovcol+nqv+ntv+nqtv;v++){
9307: printf("Covariate type in the data: V%d, DummyV(V%d)=%d, FixedV(V%d)=%d\n",v,v,DummyV[v],v,FixedV[v]);
9308: 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]);
9309: }
1.126 brouard 9310:
1.136 brouard 9311: if((fic=fopen(datafile,"r"))==NULL) {
1.218 brouard 9312: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9313: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
1.136 brouard 9314: }
1.126 brouard 9315:
1.302 brouard 9316: /* Is it a BOM UTF-8 Windows file? */
9317: /* First data line */
9318: linei=0;
9319: while(fgets(line, MAXLINE, fic)) {
9320: noffset=0;
9321: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
9322: {
9323: noffset=noffset+3;
9324: printf("# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);fflush(stdout);
9325: fprintf(ficlog,"# Data file '%s' is an UTF8 BOM file, please convert to UTF8 or ascii file and rerun.\n",datafile);
9326: fflush(ficlog); return 1;
9327: }
9328: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
9329: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
9330: {
9331: noffset=noffset+2;
1.304 brouard 9332: 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);
9333: 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 9334: fflush(ficlog); return 1;
9335: }
9336: else if( line[0] == 0 && line[1] == 0)
9337: {
9338: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
9339: noffset=noffset+4;
1.304 brouard 9340: 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);
9341: 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 9342: fflush(ficlog); return 1;
9343: }
9344: } else{
9345: ;/*printf(" Not a BOM file\n");*/
9346: }
9347: /* If line starts with a # it is a comment */
9348: if (line[noffset] == '#') {
9349: linei=linei+1;
9350: break;
9351: }else{
9352: break;
9353: }
9354: }
9355: fclose(fic);
9356: if((fic=fopen(datafile,"r"))==NULL) {
9357: printf("Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(stdout);
9358: fprintf(ficlog,"Problem while opening datafile: %s with errno='%s'\n", datafile,strerror(errno));fflush(ficlog);return 1;
9359: }
9360: /* Not a Bom file */
9361:
1.136 brouard 9362: i=1;
9363: while ((fgets(line, MAXLINE, fic) != NULL) &&((i >= firstobs) && (i <=lastobs))) {
9364: linei=linei+1;
9365: for(j=strlen(line); j>=0;j--){ /* Untabifies line */
9366: if(line[j] == '\t')
9367: line[j] = ' ';
9368: }
9369: for(j=strlen(line)-1; (line[j]==' ')||(line[j]==10)||(line[j]==13);j--){
9370: ;
9371: };
9372: line[j+1]=0; /* Trims blanks at end of line */
9373: if(line[0]=='#'){
9374: fprintf(ficlog,"Comment line\n%s\n",line);
9375: printf("Comment line\n%s\n",line);
9376: continue;
9377: }
9378: trimbb(linetmp,line); /* Trims multiple blanks in line */
1.164 brouard 9379: strcpy(line, linetmp);
1.223 brouard 9380:
9381: /* Loops on waves */
9382: for (j=maxwav;j>=1;j--){
9383: for (iv=nqtv;iv>=1;iv--){ /* Loop on time varying quantitative variables */
1.238 brouard 9384: cutv(stra, strb, line, ' ');
9385: if(strb[0]=='.') { /* Missing value */
9386: lval=-1;
9387: cotqvar[j][iv][i]=-1; /* 0.0/0.0 */
9388: cotvar[j][ntv+iv][i]=-1; /* For performance reasons */
9389: if(isalpha(strb[1])) { /* .m or .d Really Missing value */
9390: 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);
9391: 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);
9392: return 1;
9393: }
9394: }else{
9395: errno=0;
9396: /* what_kind_of_number(strb); */
9397: dval=strtod(strb,&endptr);
9398: /* if( strb[0]=='\0' || (*endptr != '\0')){ */
9399: /* if(strb != endptr && *endptr == '\0') */
9400: /* dval=dlval; */
9401: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9402: if( strb[0]=='\0' || (*endptr != '\0')){
9403: 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);
9404: 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);
9405: return 1;
9406: }
9407: cotqvar[j][iv][i]=dval;
9408: cotvar[j][ntv+iv][i]=dval;
9409: }
9410: strcpy(line,stra);
1.223 brouard 9411: }/* end loop ntqv */
1.225 brouard 9412:
1.223 brouard 9413: for (iv=ntv;iv>=1;iv--){ /* Loop on time varying dummies */
1.238 brouard 9414: cutv(stra, strb, line, ' ');
9415: if(strb[0]=='.') { /* Missing value */
9416: lval=-1;
9417: }else{
9418: errno=0;
9419: lval=strtol(strb,&endptr,10);
9420: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9421: if( strb[0]=='\0' || (*endptr != '\0')){
9422: 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);
9423: 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);
9424: return 1;
9425: }
9426: }
9427: if(lval <-1 || lval >1){
9428: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9429: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9430: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9431: For example, for multinomial values like 1, 2 and 3,\n \
9432: build V1=0 V2=0 for the reference value (1),\n \
9433: V1=1 V2=0 for (2) \n \
1.223 brouard 9434: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9435: output of IMaCh is often meaningless.\n \
1.223 brouard 9436: Exiting.\n",lval,linei, i,line,j);
1.238 brouard 9437: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.223 brouard 9438: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9439: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.238 brouard 9440: For example, for multinomial values like 1, 2 and 3,\n \
9441: build V1=0 V2=0 for the reference value (1),\n \
9442: V1=1 V2=0 for (2) \n \
1.223 brouard 9443: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.238 brouard 9444: output of IMaCh is often meaningless.\n \
1.223 brouard 9445: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.238 brouard 9446: return 1;
9447: }
9448: cotvar[j][iv][i]=(double)(lval);
9449: strcpy(line,stra);
1.223 brouard 9450: }/* end loop ntv */
1.225 brouard 9451:
1.223 brouard 9452: /* Statuses at wave */
1.137 brouard 9453: cutv(stra, strb, line, ' ');
1.223 brouard 9454: if(strb[0]=='.') { /* Missing value */
1.238 brouard 9455: lval=-1;
1.136 brouard 9456: }else{
1.238 brouard 9457: errno=0;
9458: lval=strtol(strb,&endptr,10);
9459: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN))*/
9460: if( strb[0]=='\0' || (*endptr != '\0')){
9461: 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);
9462: 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);
9463: return 1;
9464: }
1.136 brouard 9465: }
1.225 brouard 9466:
1.136 brouard 9467: s[j][i]=lval;
1.225 brouard 9468:
1.223 brouard 9469: /* Date of Interview */
1.136 brouard 9470: strcpy(line,stra);
9471: cutv(stra, strb,line,' ');
1.169 brouard 9472: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9473: }
1.169 brouard 9474: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.225 brouard 9475: month=99;
9476: year=9999;
1.136 brouard 9477: }else{
1.225 brouard 9478: 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);
9479: 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);
9480: return 1;
1.136 brouard 9481: }
9482: anint[j][i]= (double) year;
1.302 brouard 9483: mint[j][i]= (double)month;
9484: /* if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){ */
9485: /* 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]); */
9486: /* 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]); */
9487: /* } */
1.136 brouard 9488: strcpy(line,stra);
1.223 brouard 9489: } /* End loop on waves */
1.225 brouard 9490:
1.223 brouard 9491: /* Date of death */
1.136 brouard 9492: cutv(stra, strb,line,' ');
1.169 brouard 9493: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9494: }
1.169 brouard 9495: else if( (iout=sscanf(strb,"%s.",dummy)) != 0){
1.136 brouard 9496: month=99;
9497: year=9999;
9498: }else{
1.141 brouard 9499: 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 9500: 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);
9501: return 1;
1.136 brouard 9502: }
9503: andc[i]=(double) year;
9504: moisdc[i]=(double) month;
9505: strcpy(line,stra);
9506:
1.223 brouard 9507: /* Date of birth */
1.136 brouard 9508: cutv(stra, strb,line,' ');
1.169 brouard 9509: if( (iout=sscanf(strb,"%d/%d",&month, &year)) != 0){
1.136 brouard 9510: }
1.169 brouard 9511: else if( (iout=sscanf(strb,"%s.", dummy)) != 0){
1.136 brouard 9512: month=99;
9513: year=9999;
9514: }else{
1.141 brouard 9515: 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);
9516: 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 9517: return 1;
1.136 brouard 9518: }
9519: if (year==9999) {
1.141 brouard 9520: 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);
9521: 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 9522: return 1;
9523:
1.136 brouard 9524: }
9525: annais[i]=(double)(year);
1.302 brouard 9526: moisnais[i]=(double)(month);
9527: for (j=1;j<=maxwav;j++){
9528: if( (int)anint[j][i]+ (int)(mint[j][i])/12. < (int) (moisnais[i]/12.+annais[i])){
9529: 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]);
9530: 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]);
9531: }
9532: }
9533:
1.136 brouard 9534: strcpy(line,stra);
1.225 brouard 9535:
1.223 brouard 9536: /* Sample weight */
1.136 brouard 9537: cutv(stra, strb,line,' ');
9538: errno=0;
9539: dval=strtod(strb,&endptr);
9540: if( strb[0]=='\0' || (*endptr != '\0')){
1.141 brouard 9541: printf("Error reading data around '%f' at line number %d, \"%s\" for individual %d\nShould be a weight. Exiting.\n",dval, i,line,linei);
9542: 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 9543: fflush(ficlog);
9544: return 1;
9545: }
9546: weight[i]=dval;
9547: strcpy(line,stra);
1.225 brouard 9548:
1.223 brouard 9549: for (iv=nqv;iv>=1;iv--){ /* Loop on fixed quantitative variables */
9550: cutv(stra, strb, line, ' ');
9551: if(strb[0]=='.') { /* Missing value */
1.225 brouard 9552: lval=-1;
1.311 brouard 9553: coqvar[iv][i]=NAN;
9554: covar[ncovcol+iv][i]=NAN; /* including qvar in standard covar for performance reasons */
1.223 brouard 9555: }else{
1.225 brouard 9556: errno=0;
9557: /* what_kind_of_number(strb); */
9558: dval=strtod(strb,&endptr);
9559: /* if(strb != endptr && *endptr == '\0') */
9560: /* dval=dlval; */
9561: /* if (errno == ERANGE && (lval == LONG_MAX || lval == LONG_MIN)) */
9562: if( strb[0]=='\0' || (*endptr != '\0')){
9563: 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);
9564: 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);
9565: return 1;
9566: }
9567: coqvar[iv][i]=dval;
1.226 brouard 9568: covar[ncovcol+iv][i]=dval; /* including qvar in standard covar for performance reasons */
1.223 brouard 9569: }
9570: strcpy(line,stra);
9571: }/* end loop nqv */
1.136 brouard 9572:
1.223 brouard 9573: /* Covariate values */
1.136 brouard 9574: for (j=ncovcol;j>=1;j--){
9575: cutv(stra, strb,line,' ');
1.223 brouard 9576: if(strb[0]=='.') { /* Missing covariate value */
1.225 brouard 9577: lval=-1;
1.136 brouard 9578: }else{
1.225 brouard 9579: errno=0;
9580: lval=strtol(strb,&endptr,10);
9581: if( strb[0]=='\0' || (*endptr != '\0')){
9582: 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);
9583: 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);
9584: return 1;
9585: }
1.136 brouard 9586: }
9587: if(lval <-1 || lval >1){
1.225 brouard 9588: printf("Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9589: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9590: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9591: For example, for multinomial values like 1, 2 and 3,\n \
9592: build V1=0 V2=0 for the reference value (1),\n \
9593: V1=1 V2=0 for (2) \n \
1.136 brouard 9594: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9595: output of IMaCh is often meaningless.\n \
1.136 brouard 9596: Exiting.\n",lval,linei, i,line,j);
1.225 brouard 9597: fprintf(ficlog,"Error reading data around '%ld' at line number %d for individual %d, '%s'\n \
1.136 brouard 9598: Should be a value of %d(nth) covariate (0 should be the value for the reference and 1\n \
9599: for the alternative. IMaCh does not build design variables automatically, do it yourself.\n \
1.225 brouard 9600: For example, for multinomial values like 1, 2 and 3,\n \
9601: build V1=0 V2=0 for the reference value (1),\n \
9602: V1=1 V2=0 for (2) \n \
1.136 brouard 9603: and V1=0 V2=1 for (3). V1=1 V2=1 should not exist and the corresponding\n \
1.225 brouard 9604: output of IMaCh is often meaningless.\n \
1.136 brouard 9605: Exiting.\n",lval,linei, i,line,j);fflush(ficlog);
1.225 brouard 9606: return 1;
1.136 brouard 9607: }
9608: covar[j][i]=(double)(lval);
9609: strcpy(line,stra);
9610: }
9611: lstra=strlen(stra);
1.225 brouard 9612:
1.136 brouard 9613: if(lstra > 9){ /* More than 2**32 or max of what printf can write with %ld */
9614: stratrunc = &(stra[lstra-9]);
9615: num[i]=atol(stratrunc);
9616: }
9617: else
9618: num[i]=atol(stra);
9619: /*if((s[2][i]==2) && (s[3][i]==-1)&&(s[4][i]==9)){
9620: 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;}*/
9621:
9622: i=i+1;
9623: } /* End loop reading data */
1.225 brouard 9624:
1.136 brouard 9625: *imax=i-1; /* Number of individuals */
9626: fclose(fic);
1.225 brouard 9627:
1.136 brouard 9628: return (0);
1.164 brouard 9629: /* endread: */
1.225 brouard 9630: printf("Exiting readdata: ");
9631: fclose(fic);
9632: return (1);
1.223 brouard 9633: }
1.126 brouard 9634:
1.234 brouard 9635: void removefirstspace(char **stri){/*, char stro[]) {*/
1.230 brouard 9636: char *p1 = *stri, *p2 = *stri;
1.235 brouard 9637: while (*p2 == ' ')
1.234 brouard 9638: p2++;
9639: /* while ((*p1++ = *p2++) !=0) */
9640: /* ; */
9641: /* do */
9642: /* while (*p2 == ' ') */
9643: /* p2++; */
9644: /* while (*p1++ == *p2++); */
9645: *stri=p2;
1.145 brouard 9646: }
9647:
1.235 brouard 9648: int decoderesult ( char resultline[], int nres)
1.230 brouard 9649: /**< This routine decode one result line and returns the combination # of dummy covariates only **/
9650: {
1.235 brouard 9651: int j=0, k=0, k1=0, k2=0, k3=0, k4=0, match=0, k2q=0, k3q=0, k4q=0;
1.230 brouard 9652: char resultsav[MAXLINE];
1.234 brouard 9653: int resultmodel[MAXLINE];
9654: int modelresult[MAXLINE];
1.230 brouard 9655: char stra[80], strb[80], strc[80], strd[80],stre[80];
9656:
1.234 brouard 9657: removefirstspace(&resultline);
1.230 brouard 9658:
9659: if (strstr(resultline,"v") !=0){
9660: printf("Error. 'v' must be in upper case 'V' result: %s ",resultline);
9661: fprintf(ficlog,"Error. 'v' must be in upper case result: %s ",resultline);fflush(ficlog);
9662: return 1;
9663: }
9664: trimbb(resultsav, resultline);
9665: if (strlen(resultsav) >1){
9666: j=nbocc(resultsav,'='); /**< j=Number of covariate values'=' */
9667: }
1.253 brouard 9668: if(j == 0){ /* Resultline but no = */
9669: TKresult[nres]=0; /* Combination for the nresult and the model */
9670: return (0);
9671: }
1.234 brouard 9672: if( j != cptcovs ){ /* Be careful if a variable is in a product but not single */
1.310 brouard 9673: printf("ERROR: the number of variables in the resultline, %d, differs from the number of variables used in the model line, %d.\n",j, cptcovs);
9674: fprintf(ficlog,"ERROR: the number of variables in the resultline, %d, differs from the number of variables used in the model line, %d.\n",j, cptcovs);
1.234 brouard 9675: }
9676: for(k=1; k<=j;k++){ /* Loop on any covariate of the result line */
9677: if(nbocc(resultsav,'=') >1){
9678: cutl(stra,strb,resultsav,' '); /* keeps in strb after the first ' '
1.310 brouard 9679: resultsav= V4=1 V5=25.1 V3=0 stra= V5=25.1 V3=0 strb= V4=1 */
1.234 brouard 9680: cutl(strc,strd,strb,'='); /* strb:V4=1 strc=1 strd=V4 */
9681: }else
9682: cutl(strc,strd,resultsav,'=');
1.230 brouard 9683: Tvalsel[k]=atof(strc); /* 1 */
1.234 brouard 9684:
1.230 brouard 9685: cutl(strc,stre,strd,'V'); /* strd='V4' strc=4 stre='V' */;
9686: Tvarsel[k]=atoi(strc);
9687: /* Typevarsel[k]=1; /\* 1 for age product *\/ */
9688: /* cptcovsel++; */
9689: if (nbocc(stra,'=') >0)
9690: strcpy(resultsav,stra); /* and analyzes it */
9691: }
1.235 brouard 9692: /* Checking for missing or useless values in comparison of current model needs */
1.236 brouard 9693: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9694: if(Typevar[k1]==0){ /* Single covariate in model */
1.234 brouard 9695: match=0;
1.236 brouard 9696: for(k2=1; k2 <=j;k2++){/* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.237 brouard 9697: if(Tvar[k1]==Tvarsel[k2]) {/* Tvar[1]=5 == Tvarsel[2]=5 */
1.236 brouard 9698: modelresult[k2]=k1;/* modelresult[2]=1 modelresult[1]=2 modelresult[3]=3 modelresult[6]=4 modelresult[9]=5 */
1.234 brouard 9699: match=1;
9700: break;
9701: }
9702: }
9703: if(match == 0){
1.310 brouard 9704: printf("Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9705: fprintf(ficlog,"Error in result line: V%d is missing in result: %s according to model=%s\n",k1, resultline, model);
9706: return 1;
1.234 brouard 9707: }
9708: }
9709: }
1.235 brouard 9710: /* Checking for missing or useless values in comparison of current model needs */
9711: for(k2=1; k2 <=j;k2++){ /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
1.234 brouard 9712: match=0;
1.235 brouard 9713: for(k1=1; k1<= cptcovt ;k1++){ /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9714: if(Typevar[k1]==0){ /* Single */
1.237 brouard 9715: if(Tvar[k1]==Tvarsel[k2]) { /* Tvar[2]=4 == Tvarsel[1]=4 */
1.235 brouard 9716: resultmodel[k1]=k2; /* resultmodel[2]=1 resultmodel[1]=2 resultmodel[3]=3 resultmodel[6]=4 resultmodel[9]=5 */
1.234 brouard 9717: ++match;
9718: }
9719: }
9720: }
9721: if(match == 0){
9722: printf("Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
1.310 brouard 9723: fprintf(ficlog,"Error in result line: %d value missing; result: %s, model=%s\n",k1, resultline, model);
9724: return 1;
1.234 brouard 9725: }else if(match > 1){
9726: printf("Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
1.310 brouard 9727: fprintf(ficlog,"Error in result line: %d doubled; result: %s, model=%s\n",k2, resultline, model);
9728: return 1;
1.234 brouard 9729: }
9730: }
1.235 brouard 9731:
1.234 brouard 9732: /* We need to deduce which combination number is chosen and save quantitative values */
1.235 brouard 9733: /* model line V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
9734: /* result line V4=1 V5=25.1 V3=0 V2=8 V1=1 */
9735: /* should give a combination of dummy V4=1, V3=0, V1=1 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 5 + (1offset) = 6*/
9736: /* result line V4=1 V5=24.1 V3=1 V2=8 V1=0 */
9737: /* should give a combination of dummy V4=1, V3=1, V1=0 => V4*2**(0) + V3*2**(1) + V1*2**(2) = 3 + (1offset) = 4*/
9738: /* 1 0 0 0 */
9739: /* 2 1 0 0 */
9740: /* 3 0 1 0 */
9741: /* 4 1 1 0 */ /* V4=1, V3=1, V1=0 */
9742: /* 5 0 0 1 */
9743: /* 6 1 0 1 */ /* V4=1, V3=0, V1=1 */
9744: /* 7 0 1 1 */
9745: /* 8 1 1 1 */
1.237 brouard 9746: /* V(Tvresult)=Tresult V4=1 V3=0 V1=1 Tresult[nres=1][2]=0 */
9747: /* V(Tvqresult)=Tqresult V5=25.1 V2=8 Tqresult[nres=1][1]=25.1 */
9748: /* V5*age V5 known which value for nres? */
9749: /* Tqinvresult[2]=8 Tqinvresult[1]=25.1 */
1.235 brouard 9750: for(k1=1, k=0, k4=0, k4q=0; k1 <=cptcovt;k1++){ /* model line */
9751: if( Dummy[k1]==0 && Typevar[k1]==0 ){ /* Single dummy */
1.237 brouard 9752: k3= resultmodel[k1]; /* resultmodel[2(V4)] = 1=k3 */
1.235 brouard 9753: k2=(int)Tvarsel[k3]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
9754: k+=Tvalsel[k3]*pow(2,k4); /* Tvalsel[1]=1 */
1.237 brouard 9755: Tresult[nres][k4+1]=Tvalsel[k3];/* Tresult[nres][1]=1(V4=1) Tresult[nres][2]=0(V3=0) */
9756: Tvresult[nres][k4+1]=(int)Tvarsel[k3];/* Tvresult[nres][1]=4 Tvresult[nres][3]=1 */
9757: Tinvresult[nres][(int)Tvarsel[k3]]=Tvalsel[k3]; /* Tinvresult[nres][4]=1 */
1.235 brouard 9758: printf("Decoderesult Dummy k=%d, V(k2=V%d)= Tvalsel[%d]=%d, 2**(%d)\n",k, k2, k3, (int)Tvalsel[k3], k4);
9759: k4++;;
9760: } else if( Dummy[k1]==1 && Typevar[k1]==0 ){ /* Single quantitative */
9761: k3q= resultmodel[k1]; /* resultmodel[2] = 1=k3 */
9762: k2q=(int)Tvarsel[k3q]; /* Tvarsel[resultmodel[2]]= Tvarsel[1] = 4=k2 */
1.237 brouard 9763: Tqresult[nres][k4q+1]=Tvalsel[k3q]; /* Tqresult[nres][1]=25.1 */
9764: Tvqresult[nres][k4q+1]=(int)Tvarsel[k3q]; /* Tvqresult[nres][1]=5 */
9765: Tqinvresult[nres][(int)Tvarsel[k3q]]=Tvalsel[k3q]; /* Tqinvresult[nres][5]=25.1 */
1.235 brouard 9766: printf("Decoderesult Quantitative nres=%d, V(k2q=V%d)= Tvalsel[%d]=%d, Tvarsel[%d]=%f\n",nres, k2q, k3q, Tvarsel[k3q], k3q, Tvalsel[k3q]);
9767: k4q++;;
9768: }
9769: }
1.234 brouard 9770:
1.235 brouard 9771: TKresult[nres]=++k; /* Combination for the nresult and the model */
1.230 brouard 9772: return (0);
9773: }
1.235 brouard 9774:
1.230 brouard 9775: int decodemodel( char model[], int lastobs)
9776: /**< This routine decodes the model and returns:
1.224 brouard 9777: * Model V1+V2+V3+V8+V7*V8+V5*V6+V8*age+V3*age+age*age
9778: * - nagesqr = 1 if age*age in the model, otherwise 0.
9779: * - cptcovt total number of covariates of the model nbocc(+)+1 = 8 excepting constant and age and age*age
9780: * - cptcovn or number of covariates k of the models excluding age*products =6 and age*age
9781: * - cptcovage number of covariates with age*products =2
9782: * - cptcovs number of simple covariates
9783: * - 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
9784: * which is a new column after the 9 (ncovcol) variables.
9785: * - if k is a product Vn*Vm covar[k][i] is filled with correct values for each individual
9786: * - Tprod[l] gives the kth covariates of the product Vn*Vm l=1 to cptcovprod-cptcovage
9787: * Tprod[1]@2 {5, 6}: position of first product V7*V8 is 5, and second V5*V6 is 6.
9788: * - Tvard[k] p Tvard[1][1]@4 {7, 8, 5, 6} for V7*V8 and V5*V6 .
9789: */
1.136 brouard 9790: {
1.238 brouard 9791: int i, j, k, ks, v;
1.227 brouard 9792: int j1, k1, k2, k3, k4;
1.136 brouard 9793: char modelsav[80];
1.145 brouard 9794: char stra[80], strb[80], strc[80], strd[80],stre[80];
1.187 brouard 9795: char *strpt;
1.136 brouard 9796:
1.145 brouard 9797: /*removespace(model);*/
1.136 brouard 9798: if (strlen(model) >1){ /* If there is at least 1 covariate */
1.145 brouard 9799: j=0, j1=0, k1=0, k2=-1, ks=0, cptcovn=0;
1.137 brouard 9800: if (strstr(model,"AGE") !=0){
1.192 brouard 9801: printf("Error. AGE must be in lower case 'age' model=1+age+%s. ",model);
9802: fprintf(ficlog,"Error. AGE must be in lower case model=1+age+%s. ",model);fflush(ficlog);
1.136 brouard 9803: return 1;
9804: }
1.141 brouard 9805: if (strstr(model,"v") !=0){
9806: printf("Error. 'v' must be in upper case 'V' model=%s ",model);
9807: fprintf(ficlog,"Error. 'v' must be in upper case model=%s ",model);fflush(ficlog);
9808: return 1;
9809: }
1.187 brouard 9810: strcpy(modelsav,model);
9811: if ((strpt=strstr(model,"age*age")) !=0){
9812: printf(" strpt=%s, model=%s\n",strpt, model);
9813: if(strpt != model){
1.234 brouard 9814: printf("Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9815: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9816: corresponding column of parameters.\n",model);
1.234 brouard 9817: fprintf(ficlog,"Error in model: 'model=%s'; 'age*age' should in first place before other covariates\n \
1.192 brouard 9818: 'model=1+age+age*age+V1.' or 'model=1+age+age*age+V1+V1*age.', please swap as well as \n \
1.187 brouard 9819: corresponding column of parameters.\n",model); fflush(ficlog);
1.234 brouard 9820: return 1;
1.225 brouard 9821: }
1.187 brouard 9822: nagesqr=1;
9823: if (strstr(model,"+age*age") !=0)
1.234 brouard 9824: substrchaine(modelsav, model, "+age*age");
1.187 brouard 9825: else if (strstr(model,"age*age+") !=0)
1.234 brouard 9826: substrchaine(modelsav, model, "age*age+");
1.187 brouard 9827: else
1.234 brouard 9828: substrchaine(modelsav, model, "age*age");
1.187 brouard 9829: }else
9830: nagesqr=0;
9831: if (strlen(modelsav) >1){
9832: j=nbocc(modelsav,'+'); /**< j=Number of '+' */
9833: j1=nbocc(modelsav,'*'); /**< j1=Number of '*' */
1.224 brouard 9834: cptcovs=j+1-j1; /**< Number of simple covariates V1+V1*age+V3 +V3*V4+age*age=> V1 + V3 =5-3=2 */
1.187 brouard 9835: cptcovt= j+1; /* Number of total covariates in the model, not including
1.225 brouard 9836: * cst, age and age*age
9837: * V1+V1*age+ V3 + V3*V4+age*age=> 3+1=4*/
9838: /* including age products which are counted in cptcovage.
9839: * but the covariates which are products must be treated
9840: * separately: ncovn=4- 2=2 (V1+V3). */
1.187 brouard 9841: cptcovprod=j1; /**< Number of products V1*V2 +v3*age = 2 */
9842: cptcovprodnoage=0; /**< Number of covariate products without age: V3*V4 =1 */
1.225 brouard 9843:
9844:
1.187 brouard 9845: /* Design
9846: * V1 V2 V3 V4 V5 V6 V7 V8 V9 Weight
9847: * < ncovcol=8 >
9848: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8
9849: * k= 1 2 3 4 5 6 7 8
9850: * cptcovn number of covariates (not including constant and age ) = # of + plus 1 = 7+1=8
9851: * covar[k,i], value of kth covariate if not including age for individual i:
1.224 brouard 9852: * covar[1][i]= (V1), covar[4][i]=(V4), covar[8][i]=(V8)
9853: * Tvar[k] # of the kth covariate: Tvar[1]=2 Tvar[2]=1 Tvar[4]=3 Tvar[8]=8
1.187 brouard 9854: * if multiplied by age: V3*age Tvar[3=V3*age]=3 (V3) Tvar[7]=8 and
9855: * Tage[++cptcovage]=k
9856: * if products, new covar are created after ncovcol with k1
9857: * Tvar[k]=ncovcol+k1; # of the kth covariate product: Tvar[5]=ncovcol+1=10 Tvar[6]=ncovcol+1=11
9858: * Tprod[k1]=k; Tprod[1]=5 Tprod[2]= 6; gives the position of the k1th product
9859: * 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
9860: * Tvar[cptcovn+k2]=Tvard[k1][1];Tvar[cptcovn+k2+1]=Tvard[k1][2];
9861: * Tvar[8+1]=5;Tvar[8+2]=6;Tvar[8+3]=7;Tvar[8+4]=8 inverted
9862: * V1 V2 V3 V4 V5 V6 V7 V8 V9 V10 V11
9863: * < ncovcol=8 >
9864: * Model V2 + V1 + V3*age + V3 + V5*V6 + V7*V8 + V8*age + V8 d1 d1 d2 d2
9865: * k= 1 2 3 4 5 6 7 8 9 10 11 12
9866: * Tvar[k]= 2 1 3 3 10 11 8 8 5 6 7 8
9867: * p Tvar[1]@12={2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9868: * p Tprod[1]@2={ 6, 5}
9869: *p Tvard[1][1]@4= {7, 8, 5, 6}
9870: * covar[k][i]= V2 V1 ? V3 V5*V6? V7*V8? ? V8
9871: * cov[Tage[kk]+2]=covar[Tvar[Tage[kk]]][i]*cov[2];
9872: *How to reorganize?
9873: * Model V1 + V2 + V3 + V8 + V5*V6 + V7*V8 + V3*age + V8*age
9874: * Tvars {2, 1, 3, 3, 11, 10, 8, 8, 7, 8, 5, 6}
9875: * {2, 1, 4, 8, 5, 6, 3, 7}
9876: * Struct []
9877: */
1.225 brouard 9878:
1.187 brouard 9879: /* This loop fills the array Tvar from the string 'model'.*/
9880: /* j is the number of + signs in the model V1+V2+V3 j=2 i=3 to 1 */
9881: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4 */
9882: /* k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tage[cptcovage=1]=4 */
9883: /* k=3 V4 Tvar[k=3]= 4 (from V4) */
9884: /* k=2 V1 Tvar[k=2]= 1 (from V1) */
9885: /* k=1 Tvar[1]=2 (from V2) */
9886: /* k=5 Tvar[5] */
9887: /* for (k=1; k<=cptcovn;k++) { */
1.198 brouard 9888: /* cov[2+k]=nbcode[Tvar[k]][codtabm(ij,Tvar[k])]; */
1.187 brouard 9889: /* } */
1.198 brouard 9890: /* for (k=1; k<=cptcovage;k++) cov[2+Tage[k]]=nbcode[Tvar[Tage[k]]][codtabm(ij,Tvar[Tage[k])]]*cov[2]; */
1.187 brouard 9891: /*
9892: * Treating invertedly V2+V1+V3*age+V2*V4 is as if written V2*V4 +V3*age + V1 + V2 */
1.227 brouard 9893: for(k=cptcovt; k>=1;k--){ /**< Number of covariates not including constant and age, neither age*age*/
9894: Tvar[k]=0; Tprod[k]=0; Tposprod[k]=0;
9895: }
1.187 brouard 9896: cptcovage=0;
9897: for(k=1; k<=cptcovt;k++){ /* Loop on total covariates of the model */
1.234 brouard 9898: cutl(stra,strb,modelsav,'+'); /* keeps in strb after the first '+'
1.225 brouard 9899: modelsav==V2+V1+V4+V3*age strb=V3*age stra=V2+V1+V4 */
1.234 brouard 9900: if (nbocc(modelsav,'+')==0) strcpy(strb,modelsav); /* and analyzes it */
9901: /* printf("i=%d a=%s b=%s sav=%s\n",i, stra,strb,modelsav);*/
9902: /*scanf("%d",i);*/
9903: if (strchr(strb,'*')) { /**< Model includes a product V2+V1+V4+V3*age strb=V3*age */
9904: cutl(strc,strd,strb,'*'); /**< strd*strc Vm*Vn: strb=V3*age(input) strc=age strd=V3 ; V3*V2 strc=V2, strd=V3 */
9905: if (strcmp(strc,"age")==0) { /**< Model includes age: Vn*age */
9906: /* covar is not filled and then is empty */
9907: cptcovprod--;
9908: cutl(stre,strb,strd,'V'); /* strd=V3(input): stre="3" */
9909: Tvar[k]=atoi(stre); /* V2+V1+V4+V3*age Tvar[4]=3 ; V1+V2*age Tvar[2]=2; V1+V1*age Tvar[2]=1 */
9910: Typevar[k]=1; /* 1 for age product */
9911: cptcovage++; /* Sums the number of covariates which include age as a product */
9912: Tage[cptcovage]=k; /* Tvar[4]=3, Tage[1] = 4 or V1+V1*age Tvar[2]=1, Tage[1]=2 */
9913: /*printf("stre=%s ", stre);*/
9914: } else if (strcmp(strd,"age")==0) { /* or age*Vn */
9915: cptcovprod--;
9916: cutl(stre,strb,strc,'V');
9917: Tvar[k]=atoi(stre);
9918: Typevar[k]=1; /* 1 for age product */
9919: cptcovage++;
9920: Tage[cptcovage]=k;
9921: } else { /* Age is not in the model product V2+V1+V1*V4+V3*age+V3*V2 strb=V3*V2*/
9922: /* loops on k1=1 (V3*V2) and k1=2 V4*V3 */
9923: cptcovn++;
9924: cptcovprodnoage++;k1++;
9925: cutl(stre,strb,strc,'V'); /* strc= Vn, stre is n; strb=V3*V2 stre=3 strc=*/
9926: Tvar[k]=ncovcol+nqv+ntv+nqtv+k1; /* For model-covariate k tells which data-covariate to use but
9927: because this model-covariate is a construction we invent a new column
9928: which is after existing variables ncovcol+nqv+ntv+nqtv + k1
9929: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2
9930: Tvar[3=V1*V4]=4+1 Tvar[5=V3*V2]=4 + 2= 6, etc */
9931: Typevar[k]=2; /* 2 for double fixed dummy covariates */
9932: cutl(strc,strb,strd,'V'); /* strd was Vm, strc is m */
9933: Tprod[k1]=k; /* Tprod[1]=3(=V1*V4) for V2+V1+V1*V4+age*V3+V3*V2 */
9934: Tposprod[k]=k1; /* Tpsprod[3]=1, Tposprod[2]=5 */
9935: Tvard[k1][1] =atoi(strc); /* m 1 for V1*/
9936: Tvard[k1][2] =atoi(stre); /* n 4 for V4*/
9937: k2=k2+2; /* k2 is initialize to -1, We want to store the n and m in Vn*Vm at the end of Tvar */
9938: /* Tvar[cptcovt+k2]=Tvard[k1][1]; /\* Tvar[(cptcovt=4+k2=1)=5]= 1 (V1) *\/ */
9939: /* Tvar[cptcovt+k2+1]=Tvard[k1][2]; /\* Tvar[(cptcovt=4+(k2=1)+1)=6]= 4 (V4) *\/ */
1.225 brouard 9940: /*ncovcol=4 and model=V2+V1+V1*V4+age*V3+V3*V2, Tvar[3]=5, Tvar[4]=6, cptcovt=5 */
1.234 brouard 9941: /* 1 2 3 4 5 | Tvar[5+1)=1, Tvar[7]=2 */
9942: for (i=1; i<=lastobs;i++){
9943: /* Computes the new covariate which is a product of
9944: covar[n][i]* covar[m][i] and stores it at ncovol+k1 May not be defined */
9945: covar[ncovcol+k1][i]=covar[atoi(stre)][i]*covar[atoi(strc)][i];
9946: }
9947: } /* End age is not in the model */
9948: } /* End if model includes a product */
9949: else { /* no more sum */
9950: /*printf("d=%s c=%s b=%s\n", strd,strc,strb);*/
9951: /* scanf("%d",i);*/
9952: cutl(strd,strc,strb,'V');
9953: ks++; /**< Number of simple covariates dummy or quantitative, fixe or varying */
9954: cptcovn++; /** V4+V3+V5: V4 and V3 timevarying dummy covariates, V5 timevarying quantitative */
9955: Tvar[k]=atoi(strd);
9956: Typevar[k]=0; /* 0 for simple covariates */
9957: }
9958: strcpy(modelsav,stra); /* modelsav=V2+V1+V4 stra=V2+V1+V4 */
1.223 brouard 9959: /*printf("a=%s b=%s sav=%s\n", stra,strb,modelsav);
1.225 brouard 9960: scanf("%d",i);*/
1.187 brouard 9961: } /* end of loop + on total covariates */
9962: } /* end if strlen(modelsave == 0) age*age might exist */
9963: } /* end if strlen(model == 0) */
1.136 brouard 9964:
9965: /*The number n of Vn is stored in Tvar. cptcovage =number of age covariate. Tage gives the position of age. cptcovprod= number of products.
9966: If model=V1+V1*age then Tvar[1]=1 Tvar[2]=1 cptcovage=1 Tage[1]=2 cptcovprod=0*/
1.225 brouard 9967:
1.136 brouard 9968: /* printf("tvar1=%d tvar2=%d tvar3=%d cptcovage=%d Tage=%d",Tvar[1],Tvar[2],Tvar[3],cptcovage,Tage[1]);
1.225 brouard 9969: printf("cptcovprod=%d ", cptcovprod);
9970: fprintf(ficlog,"cptcovprod=%d ", cptcovprod);
9971: scanf("%d ",i);*/
9972:
9973:
1.230 brouard 9974: /* Until here, decodemodel knows only the grammar (simple, product, age*) of the model but not what kind
9975: of variable (dummy vs quantitative, fixed vs time varying) is behind. But we know the # of each. */
1.226 brouard 9976: /* ncovcol= 1, nqv=1 | ntv=2, nqtv= 1 = 5 possible variables data: 2 fixed 3, varying
9977: model= V5 + V4 +V3 + V4*V3 + V5*age + V2 + V1*V2 + V1*age + V5*age, V1 is not used saving its place
9978: k = 1 2 3 4 5 6 7 8 9
9979: Tvar[k]= 5 4 3 1+1+2+1+1=6 5 2 7 1 5
9980: Typevar[k]= 0 0 0 2 1 0 2 1 1
1.227 brouard 9981: Fixed[k] 1 1 1 1 3 0 0 or 2 2 3
9982: Dummy[k] 1 0 0 0 3 1 1 2 3
9983: Tmodelind[combination of covar]=k;
1.225 brouard 9984: */
9985: /* Dispatching between quantitative and time varying covariates */
1.226 brouard 9986: /* If Tvar[k] >ncovcol it is a product */
1.225 brouard 9987: /* 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 9988: /* Computing effective variables, ie used by the model, that is from the cptcovt variables */
1.227 brouard 9989: printf("Model=%s\n\
9990: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9991: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9992: 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);
9993: fprintf(ficlog,"Model=%s\n\
9994: Typevar: 0 for simple covariate (dummy, quantitative, fixed or varying), 1 for age product, 2 for product \n\
9995: Fixed[k] 0=fixed (product or simple), 1 varying, 2 fixed with age product, 3 varying with age product \n\
9996: 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 9997: for(k=-1;k<=cptcovt; k++){ Fixed[k]=0; Dummy[k]=0;}
1.234 brouard 9998: 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 */
9999: if (Tvar[k] <=ncovcol && Typevar[k]==0 ){ /* Simple fixed dummy (<=ncovcol) covariates */
1.227 brouard 10000: Fixed[k]= 0;
10001: Dummy[k]= 0;
1.225 brouard 10002: ncoveff++;
1.232 brouard 10003: ncovf++;
1.234 brouard 10004: nsd++;
10005: modell[k].maintype= FTYPE;
10006: TvarsD[nsd]=Tvar[k];
10007: TvarsDind[nsd]=k;
10008: TvarF[ncovf]=Tvar[k];
10009: TvarFind[ncovf]=k;
10010: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10011: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
10012: }else if( Tvar[k] <=ncovcol && Typevar[k]==2){ /* Product of fixed dummy (<=ncovcol) covariates */
10013: Fixed[k]= 0;
10014: Dummy[k]= 0;
10015: ncoveff++;
10016: ncovf++;
10017: modell[k].maintype= FTYPE;
10018: TvarF[ncovf]=Tvar[k];
10019: TvarFind[ncovf]=k;
1.230 brouard 10020: TvarFD[ncoveff]=Tvar[k]; /* TvarFD[1]=V1 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.231 brouard 10021: TvarFDind[ncoveff]=k; /* TvarFDind[1]=9 in V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 */
1.240 brouard 10022: }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 10023: Fixed[k]= 0;
10024: Dummy[k]= 1;
1.230 brouard 10025: nqfveff++;
1.234 brouard 10026: modell[k].maintype= FTYPE;
10027: modell[k].subtype= FQ;
10028: nsq++;
10029: TvarsQ[nsq]=Tvar[k];
10030: TvarsQind[nsq]=k;
1.232 brouard 10031: ncovf++;
1.234 brouard 10032: TvarF[ncovf]=Tvar[k];
10033: TvarFind[ncovf]=k;
1.231 brouard 10034: 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 10035: 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 10036: }else if( Tvar[k] <=ncovcol+nqv+ntv && Typevar[k]==0){/* Only simple time varying dummy variables */
1.227 brouard 10037: Fixed[k]= 1;
10038: Dummy[k]= 0;
1.225 brouard 10039: ntveff++; /* Only simple time varying dummy variable */
1.234 brouard 10040: modell[k].maintype= VTYPE;
10041: modell[k].subtype= VD;
10042: nsd++;
10043: TvarsD[nsd]=Tvar[k];
10044: TvarsDind[nsd]=k;
10045: ncovv++; /* Only simple time varying variables */
10046: TvarV[ncovv]=Tvar[k];
1.242 brouard 10047: 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 10048: 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 */
10049: 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 10050: 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);
10051: printf("Quasi TmodelInvind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv);
1.231 brouard 10052: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv && Typevar[k]==0){ /* Only simple time varying quantitative variable V5*/
1.234 brouard 10053: Fixed[k]= 1;
10054: Dummy[k]= 1;
10055: nqtveff++;
10056: modell[k].maintype= VTYPE;
10057: modell[k].subtype= VQ;
10058: ncovv++; /* Only simple time varying variables */
10059: nsq++;
10060: TvarsQ[nsq]=Tvar[k];
10061: TvarsQind[nsq]=k;
10062: TvarV[ncovv]=Tvar[k];
1.242 brouard 10063: 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 10064: 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 */
10065: 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 10066: TmodelInvQind[nqtveff]=Tvar[k]- ncovcol-nqv-ntv;/* Only simple time varying quantitative variable */
10067: /* Tmodeliqind[k]=nqtveff;/\* Only simple time varying quantitative variable *\/ */
10068: 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 10069: printf("Quasi TmodelInvQind[%d]=%d\n",k,Tvar[k]- ncovcol-nqv-ntv);
1.227 brouard 10070: }else if (Typevar[k] == 1) { /* product with age */
1.234 brouard 10071: ncova++;
10072: TvarA[ncova]=Tvar[k];
10073: TvarAind[ncova]=k;
1.231 brouard 10074: if (Tvar[k] <=ncovcol ){ /* Product age with fixed dummy covariatee */
1.240 brouard 10075: Fixed[k]= 2;
10076: Dummy[k]= 2;
10077: modell[k].maintype= ATYPE;
10078: modell[k].subtype= APFD;
10079: /* ncoveff++; */
1.227 brouard 10080: }else if( Tvar[k] <=ncovcol+nqv) { /* Remind that product Vn*Vm are added in k*/
1.240 brouard 10081: Fixed[k]= 2;
10082: Dummy[k]= 3;
10083: modell[k].maintype= ATYPE;
10084: modell[k].subtype= APFQ; /* Product age * fixed quantitative */
10085: /* nqfveff++; /\* Only simple fixed quantitative variable *\/ */
1.227 brouard 10086: }else if( Tvar[k] <=ncovcol+nqv+ntv ){
1.240 brouard 10087: Fixed[k]= 3;
10088: Dummy[k]= 2;
10089: modell[k].maintype= ATYPE;
10090: modell[k].subtype= APVD; /* Product age * varying dummy */
10091: /* ntveff++; /\* Only simple time varying dummy variable *\/ */
1.227 brouard 10092: }else if( Tvar[k] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10093: Fixed[k]= 3;
10094: Dummy[k]= 3;
10095: modell[k].maintype= ATYPE;
10096: modell[k].subtype= APVQ; /* Product age * varying quantitative */
10097: /* nqtveff++;/\* Only simple time varying quantitative variable *\/ */
1.227 brouard 10098: }
10099: }else if (Typevar[k] == 2) { /* product without age */
10100: k1=Tposprod[k];
10101: if(Tvard[k1][1] <=ncovcol){
1.240 brouard 10102: if(Tvard[k1][2] <=ncovcol){
10103: Fixed[k]= 1;
10104: Dummy[k]= 0;
10105: modell[k].maintype= FTYPE;
10106: modell[k].subtype= FPDD; /* Product fixed dummy * fixed dummy */
10107: ncovf++; /* Fixed variables without age */
10108: TvarF[ncovf]=Tvar[k];
10109: TvarFind[ncovf]=k;
10110: }else if(Tvard[k1][2] <=ncovcol+nqv){
10111: Fixed[k]= 0; /* or 2 ?*/
10112: Dummy[k]= 1;
10113: modell[k].maintype= FTYPE;
10114: modell[k].subtype= FPDQ; /* Product fixed dummy * fixed quantitative */
10115: ncovf++; /* Varying variables without age */
10116: TvarF[ncovf]=Tvar[k];
10117: TvarFind[ncovf]=k;
10118: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10119: Fixed[k]= 1;
10120: Dummy[k]= 0;
10121: modell[k].maintype= VTYPE;
10122: modell[k].subtype= VPDD; /* Product fixed dummy * varying dummy */
10123: ncovv++; /* Varying variables without age */
10124: TvarV[ncovv]=Tvar[k];
10125: TvarVind[ncovv]=k;
10126: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10127: Fixed[k]= 1;
10128: Dummy[k]= 1;
10129: modell[k].maintype= VTYPE;
10130: modell[k].subtype= VPDQ; /* Product fixed dummy * varying quantitative */
10131: ncovv++; /* Varying variables without age */
10132: TvarV[ncovv]=Tvar[k];
10133: TvarVind[ncovv]=k;
10134: }
1.227 brouard 10135: }else if(Tvard[k1][1] <=ncovcol+nqv){
1.240 brouard 10136: if(Tvard[k1][2] <=ncovcol){
10137: Fixed[k]= 0; /* or 2 ?*/
10138: Dummy[k]= 1;
10139: modell[k].maintype= FTYPE;
10140: modell[k].subtype= FPDQ; /* Product fixed quantitative * fixed dummy */
10141: ncovf++; /* Fixed variables without age */
10142: TvarF[ncovf]=Tvar[k];
10143: TvarFind[ncovf]=k;
10144: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10145: Fixed[k]= 1;
10146: Dummy[k]= 1;
10147: modell[k].maintype= VTYPE;
10148: modell[k].subtype= VPDQ; /* Product fixed quantitative * varying dummy */
10149: ncovv++; /* Varying variables without age */
10150: TvarV[ncovv]=Tvar[k];
10151: TvarVind[ncovv]=k;
10152: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10153: Fixed[k]= 1;
10154: Dummy[k]= 1;
10155: modell[k].maintype= VTYPE;
10156: modell[k].subtype= VPQQ; /* Product fixed quantitative * varying quantitative */
10157: ncovv++; /* Varying variables without age */
10158: TvarV[ncovv]=Tvar[k];
10159: TvarVind[ncovv]=k;
10160: ncovv++; /* Varying variables without age */
10161: TvarV[ncovv]=Tvar[k];
10162: TvarVind[ncovv]=k;
10163: }
1.227 brouard 10164: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv){
1.240 brouard 10165: if(Tvard[k1][2] <=ncovcol){
10166: Fixed[k]= 1;
10167: Dummy[k]= 1;
10168: modell[k].maintype= VTYPE;
10169: modell[k].subtype= VPDD; /* Product time varying dummy * fixed dummy */
10170: ncovv++; /* Varying variables without age */
10171: TvarV[ncovv]=Tvar[k];
10172: TvarVind[ncovv]=k;
10173: }else if(Tvard[k1][2] <=ncovcol+nqv){
10174: Fixed[k]= 1;
10175: Dummy[k]= 1;
10176: modell[k].maintype= VTYPE;
10177: modell[k].subtype= VPDQ; /* Product time varying dummy * fixed quantitative */
10178: ncovv++; /* Varying variables without age */
10179: TvarV[ncovv]=Tvar[k];
10180: TvarVind[ncovv]=k;
10181: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10182: Fixed[k]= 1;
10183: Dummy[k]= 0;
10184: modell[k].maintype= VTYPE;
10185: modell[k].subtype= VPDD; /* Product time varying dummy * time varying dummy */
10186: ncovv++; /* Varying variables without age */
10187: TvarV[ncovv]=Tvar[k];
10188: TvarVind[ncovv]=k;
10189: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10190: Fixed[k]= 1;
10191: Dummy[k]= 1;
10192: modell[k].maintype= VTYPE;
10193: modell[k].subtype= VPDQ; /* Product time varying dummy * time varying quantitative */
10194: ncovv++; /* Varying variables without age */
10195: TvarV[ncovv]=Tvar[k];
10196: TvarVind[ncovv]=k;
10197: }
1.227 brouard 10198: }else if(Tvard[k1][1] <=ncovcol+nqv+ntv+nqtv){
1.240 brouard 10199: if(Tvard[k1][2] <=ncovcol){
10200: Fixed[k]= 1;
10201: Dummy[k]= 1;
10202: modell[k].maintype= VTYPE;
10203: modell[k].subtype= VPDQ; /* Product time varying quantitative * fixed dummy */
10204: ncovv++; /* Varying variables without age */
10205: TvarV[ncovv]=Tvar[k];
10206: TvarVind[ncovv]=k;
10207: }else if(Tvard[k1][2] <=ncovcol+nqv){
10208: Fixed[k]= 1;
10209: Dummy[k]= 1;
10210: modell[k].maintype= VTYPE;
10211: modell[k].subtype= VPQQ; /* Product time varying quantitative * fixed quantitative */
10212: ncovv++; /* Varying variables without age */
10213: TvarV[ncovv]=Tvar[k];
10214: TvarVind[ncovv]=k;
10215: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv){
10216: Fixed[k]= 1;
10217: Dummy[k]= 1;
10218: modell[k].maintype= VTYPE;
10219: modell[k].subtype= VPDQ; /* Product time varying quantitative * time varying dummy */
10220: ncovv++; /* Varying variables without age */
10221: TvarV[ncovv]=Tvar[k];
10222: TvarVind[ncovv]=k;
10223: }else if(Tvard[k1][2] <=ncovcol+nqv+ntv+nqtv){
10224: Fixed[k]= 1;
10225: Dummy[k]= 1;
10226: modell[k].maintype= VTYPE;
10227: modell[k].subtype= VPQQ; /* Product time varying quantitative * time varying quantitative */
10228: ncovv++; /* Varying variables without age */
10229: TvarV[ncovv]=Tvar[k];
10230: TvarVind[ncovv]=k;
10231: }
1.227 brouard 10232: }else{
1.240 brouard 10233: printf("Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10234: fprintf(ficlog,"Error unknown type of covariate: Tvard[%d][1]=%d,Tvard[%d][2]=%d\n",k1,Tvard[k1][1],k1,Tvard[k1][2]);
10235: } /*end k1*/
1.225 brouard 10236: }else{
1.226 brouard 10237: printf("Error, current version can't treat for performance reasons, Tvar[%d]=%d, Typevar[%d]=%d\n", k, Tvar[k], k, Typevar[k]);
10238: 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 10239: }
1.227 brouard 10240: 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 10241: printf(" modell[%d].maintype=%d, modell[%d].subtype=%d\n",k,modell[k].maintype,k,modell[k].subtype);
1.227 brouard 10242: 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]);
10243: }
10244: /* Searching for doublons in the model */
10245: for(k1=1; k1<= cptcovt;k1++){
10246: for(k2=1; k2 <k1;k2++){
1.285 brouard 10247: /* if((Typevar[k1]==Typevar[k2]) && (Fixed[Tvar[k1]]==Fixed[Tvar[k2]]) && (Dummy[Tvar[k1]]==Dummy[Tvar[k2]] )){ */
10248: if((Typevar[k1]==Typevar[k2]) && (Fixed[k1]==Fixed[k2]) && (Dummy[k1]==Dummy[k2] )){
1.234 brouard 10249: if((Typevar[k1] == 0 || Typevar[k1] == 1)){ /* Simple or age product */
10250: if(Tvar[k1]==Tvar[k2]){
1.285 brouard 10251: 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]);
10252: 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 10253: return(1);
10254: }
10255: }else if (Typevar[k1] ==2){
10256: k3=Tposprod[k1];
10257: k4=Tposprod[k2];
10258: 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])) ){
10259: 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]]);
10260: 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);
10261: return(1);
10262: }
10263: }
1.227 brouard 10264: }
10265: }
1.225 brouard 10266: }
10267: printf("ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
10268: fprintf(ficlog,"ncoveff=%d, nqfveff=%d, ntveff=%d, nqtveff=%d, cptcovn=%d\n",ncoveff,nqfveff,ntveff,nqtveff,cptcovn);
1.234 brouard 10269: printf("ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd,nsq);
10270: fprintf(ficlog,"ncovf=%d, ncovv=%d, ncova=%d, nsd=%d, nsq=%d\n",ncovf,ncovv,ncova,nsd, nsq);
1.137 brouard 10271: return (0); /* with covar[new additional covariate if product] and Tage if age */
1.164 brouard 10272: /*endread:*/
1.225 brouard 10273: printf("Exiting decodemodel: ");
10274: return (1);
1.136 brouard 10275: }
10276:
1.169 brouard 10277: int calandcheckages(int imx, int maxwav, double *agemin, double *agemax, int *nberr, int *nbwarn )
1.248 brouard 10278: {/* Check ages at death */
1.136 brouard 10279: int i, m;
1.218 brouard 10280: int firstone=0;
10281:
1.136 brouard 10282: for (i=1; i<=imx; i++) {
10283: for(m=2; (m<= maxwav); m++) {
10284: if (((int)mint[m][i]== 99) && (s[m][i] <= nlstate)){
10285: anint[m][i]=9999;
1.216 brouard 10286: if (s[m][i] != -2) /* Keeping initial status of unknown vital status */
10287: s[m][i]=-1;
1.136 brouard 10288: }
10289: if((int)moisdc[i]==99 && (int)andc[i]==9999 && s[m][i]>nlstate){
1.260 brouard 10290: *nberr = *nberr + 1;
1.218 brouard 10291: if(firstone == 0){
10292: firstone=1;
1.260 brouard 10293: 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 10294: }
1.262 brouard 10295: 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 10296: s[m][i]=-1; /* Droping the death status */
1.136 brouard 10297: }
10298: if((int)moisdc[i]==99 && (int)andc[i]!=9999 && s[m][i]>nlstate){
1.169 brouard 10299: (*nberr)++;
1.259 brouard 10300: 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 10301: 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 10302: s[m][i]=-2; /* We prefer to skip it (and to skip it in version 0.8a1 too */
1.136 brouard 10303: }
10304: }
10305: }
10306:
10307: for (i=1; i<=imx; i++) {
10308: agedc[i]=(moisdc[i]/12.+andc[i])-(moisnais[i]/12.+annais[i]);
10309: for(m=firstpass; (m<= lastpass); m++){
1.214 brouard 10310: 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 10311: if (s[m][i] >= nlstate+1) {
1.169 brouard 10312: if(agedc[i]>0){
10313: if((int)moisdc[i]!=99 && (int)andc[i]!=9999){
1.136 brouard 10314: agev[m][i]=agedc[i];
1.214 brouard 10315: /*if(moisdc[i]==99 && andc[i]==9999) s[m][i]=-1;*/
1.169 brouard 10316: }else {
1.136 brouard 10317: if ((int)andc[i]!=9999){
10318: nbwarn++;
10319: printf("Warning negative age at death: %ld line:%d\n",num[i],i);
10320: fprintf(ficlog,"Warning negative age at death: %ld line:%d\n",num[i],i);
10321: agev[m][i]=-1;
10322: }
10323: }
1.169 brouard 10324: } /* agedc > 0 */
1.214 brouard 10325: } /* end if */
1.136 brouard 10326: else if(s[m][i] !=9){ /* Standard case, age in fractional
10327: years but with the precision of a month */
10328: agev[m][i]=(mint[m][i]/12.+1./24.+anint[m][i])-(moisnais[i]/12.+1./24.+annais[i]);
10329: if((int)mint[m][i]==99 || (int)anint[m][i]==9999)
10330: agev[m][i]=1;
10331: else if(agev[m][i] < *agemin){
10332: *agemin=agev[m][i];
10333: printf(" Min anint[%d][%d]=%.2f annais[%d]=%.2f, agemin=%.2f\n",m,i,anint[m][i], i,annais[i], *agemin);
10334: }
10335: else if(agev[m][i] >*agemax){
10336: *agemax=agev[m][i];
1.156 brouard 10337: /* printf(" Max anint[%d][%d]=%.0f annais[%d]=%.0f, agemax=%.2f\n",m,i,anint[m][i], i,annais[i], *agemax);*/
1.136 brouard 10338: }
10339: /*agev[m][i]=anint[m][i]-annais[i];*/
10340: /* agev[m][i] = age[i]+2*m;*/
1.214 brouard 10341: } /* en if 9*/
1.136 brouard 10342: else { /* =9 */
1.214 brouard 10343: /* printf("Debug num[%d]=%ld s[%d][%d]=%d\n",i,num[i], m,i, s[m][i]); */
1.136 brouard 10344: agev[m][i]=1;
10345: s[m][i]=-1;
10346: }
10347: }
1.214 brouard 10348: else if(s[m][i]==0) /*= 0 Unknown */
1.136 brouard 10349: agev[m][i]=1;
1.214 brouard 10350: else{
10351: printf("Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10352: fprintf(ficlog, "Warning, num[%d]=%ld, s[%d][%d]=%d\n", i, num[i], m, i,s[m][i]);
10353: agev[m][i]=0;
10354: }
10355: } /* End for lastpass */
10356: }
1.136 brouard 10357:
10358: for (i=1; i<=imx; i++) {
10359: for(m=firstpass; (m<=lastpass); m++){
10360: if (s[m][i] > (nlstate+ndeath)) {
1.169 brouard 10361: (*nberr)++;
1.136 brouard 10362: 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);
10363: 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);
10364: return 1;
10365: }
10366: }
10367: }
10368:
10369: /*for (i=1; i<=imx; i++){
10370: for (m=firstpass; (m<lastpass); m++){
10371: printf("%ld %d %.lf %d %d\n", num[i],(covar[1][i]),agev[m][i],s[m][i],s[m+1][i]);
10372: }
10373:
10374: }*/
10375:
10376:
1.139 brouard 10377: printf("Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
10378: fprintf(ficlog,"Total number of individuals= %d, Agemin = %.2f, Agemax= %.2f\n\n", imx, *agemin, *agemax);
1.136 brouard 10379:
10380: return (0);
1.164 brouard 10381: /* endread:*/
1.136 brouard 10382: printf("Exiting calandcheckages: ");
10383: return (1);
10384: }
10385:
1.172 brouard 10386: #if defined(_MSC_VER)
10387: /*printf("Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10388: /*fprintf(ficlog, "Visual C++ compiler: %s \n;", _MSC_FULL_VER);*/
10389: //#include "stdafx.h"
10390: //#include <stdio.h>
10391: //#include <tchar.h>
10392: //#include <windows.h>
10393: //#include <iostream>
10394: typedef BOOL(WINAPI *LPFN_ISWOW64PROCESS) (HANDLE, PBOOL);
10395:
10396: LPFN_ISWOW64PROCESS fnIsWow64Process;
10397:
10398: BOOL IsWow64()
10399: {
10400: BOOL bIsWow64 = FALSE;
10401:
10402: //typedef BOOL (APIENTRY *LPFN_ISWOW64PROCESS)
10403: // (HANDLE, PBOOL);
10404:
10405: //LPFN_ISWOW64PROCESS fnIsWow64Process;
10406:
10407: HMODULE module = GetModuleHandle(_T("kernel32"));
10408: const char funcName[] = "IsWow64Process";
10409: fnIsWow64Process = (LPFN_ISWOW64PROCESS)
10410: GetProcAddress(module, funcName);
10411:
10412: if (NULL != fnIsWow64Process)
10413: {
10414: if (!fnIsWow64Process(GetCurrentProcess(),
10415: &bIsWow64))
10416: //throw std::exception("Unknown error");
10417: printf("Unknown error\n");
10418: }
10419: return bIsWow64 != FALSE;
10420: }
10421: #endif
1.177 brouard 10422:
1.191 brouard 10423: void syscompilerinfo(int logged)
1.292 brouard 10424: {
10425: #include <stdint.h>
10426:
10427: /* #include "syscompilerinfo.h"*/
1.185 brouard 10428: /* command line Intel compiler 32bit windows, XP compatible:*/
10429: /* /GS /W3 /Gy
10430: /Zc:wchar_t /Zi /O2 /Fd"Release\vc120.pdb" /D "WIN32" /D "NDEBUG" /D
10431: "_CONSOLE" /D "_LIB" /D "_USING_V110_SDK71_" /D "_UNICODE" /D
10432: "UNICODE" /Qipo /Zc:forScope /Gd /Oi /MT /Fa"Release\" /EHsc /nologo
1.186 brouard 10433: /Fo"Release\" /Qprof-dir "Release\" /Fp"Release\IMaCh.pch"
10434: */
10435: /* 64 bits */
1.185 brouard 10436: /*
10437: /GS /W3 /Gy
10438: /Zc:wchar_t /Zi /O2 /Fd"x64\Release\vc120.pdb" /D "WIN32" /D "NDEBUG"
10439: /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo /Zc:forScope
10440: /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Fo"x64\Release\" /Qprof-dir
10441: "x64\Release\" /Fp"x64\Release\IMaCh.pch" */
10442: /* Optimization are useless and O3 is slower than O2 */
10443: /*
10444: /GS /W3 /Gy /Zc:wchar_t /Zi /O3 /Fd"x64\Release\vc120.pdb" /D "WIN32"
10445: /D "NDEBUG" /D "_CONSOLE" /D "_LIB" /D "_UNICODE" /D "UNICODE" /Qipo
10446: /Zc:forScope /Oi /MD /Fa"x64\Release\" /EHsc /nologo /Qparallel
10447: /Fo"x64\Release\" /Qprof-dir "x64\Release\" /Fp"x64\Release\IMaCh.pch"
10448: */
1.186 brouard 10449: /* Link is */ /* /OUT:"visual studio
1.185 brouard 10450: 2013\Projects\IMaCh\Release\IMaCh.exe" /MANIFEST /NXCOMPAT
10451: /PDB:"visual studio
10452: 2013\Projects\IMaCh\Release\IMaCh.pdb" /DYNAMICBASE
10453: "kernel32.lib" "user32.lib" "gdi32.lib" "winspool.lib"
10454: "comdlg32.lib" "advapi32.lib" "shell32.lib" "ole32.lib"
10455: "oleaut32.lib" "uuid.lib" "odbc32.lib" "odbccp32.lib"
10456: /MACHINE:X86 /OPT:REF /SAFESEH /INCREMENTAL:NO
10457: /SUBSYSTEM:CONSOLE",5.01" /MANIFESTUAC:"level='asInvoker'
10458: uiAccess='false'"
10459: /ManifestFile:"Release\IMaCh.exe.intermediate.manifest" /OPT:ICF
10460: /NOLOGO /TLBID:1
10461: */
1.292 brouard 10462:
10463:
1.177 brouard 10464: #if defined __INTEL_COMPILER
1.178 brouard 10465: #if defined(__GNUC__)
10466: struct utsname sysInfo; /* For Intel on Linux and OS/X */
10467: #endif
1.177 brouard 10468: #elif defined(__GNUC__)
1.179 brouard 10469: #ifndef __APPLE__
1.174 brouard 10470: #include <gnu/libc-version.h> /* Only on gnu */
1.179 brouard 10471: #endif
1.177 brouard 10472: struct utsname sysInfo;
1.178 brouard 10473: int cross = CROSS;
10474: if (cross){
10475: printf("Cross-");
1.191 brouard 10476: if(logged) fprintf(ficlog, "Cross-");
1.178 brouard 10477: }
1.174 brouard 10478: #endif
10479:
1.191 brouard 10480: printf("Compiled with:");if(logged)fprintf(ficlog,"Compiled with:");
1.169 brouard 10481: #if defined(__clang__)
1.191 brouard 10482: printf(" Clang/LLVM");if(logged)fprintf(ficlog," Clang/LLVM"); /* Clang/LLVM. ---------------------------------------------- */
1.169 brouard 10483: #endif
10484: #if defined(__ICC) || defined(__INTEL_COMPILER)
1.191 brouard 10485: printf(" Intel ICC/ICPC");if(logged)fprintf(ficlog," Intel ICC/ICPC");/* Intel ICC/ICPC. ------------------------------------------ */
1.169 brouard 10486: #endif
10487: #if defined(__GNUC__) || defined(__GNUG__)
1.191 brouard 10488: printf(" GNU GCC/G++");if(logged)fprintf(ficlog," GNU GCC/G++");/* GNU GCC/G++. --------------------------------------------- */
1.169 brouard 10489: #endif
10490: #if defined(__HP_cc) || defined(__HP_aCC)
1.191 brouard 10491: printf(" Hewlett-Packard C/aC++");if(logged)fprintf(fcilog," Hewlett-Packard C/aC++"); /* Hewlett-Packard C/aC++. ---------------------------------- */
1.169 brouard 10492: #endif
10493: #if defined(__IBMC__) || defined(__IBMCPP__)
1.191 brouard 10494: printf(" IBM XL C/C++"); if(logged) fprintf(ficlog," IBM XL C/C++");/* IBM XL C/C++. -------------------------------------------- */
1.169 brouard 10495: #endif
10496: #if defined(_MSC_VER)
1.191 brouard 10497: printf(" Microsoft Visual Studio");if(logged)fprintf(ficlog," Microsoft Visual Studio");/* Microsoft Visual Studio. --------------------------------- */
1.169 brouard 10498: #endif
10499: #if defined(__PGI)
1.191 brouard 10500: printf(" Portland Group PGCC/PGCPP");if(logged) fprintf(ficlog," Portland Group PGCC/PGCPP");/* Portland Group PGCC/PGCPP. ------------------------------- */
1.169 brouard 10501: #endif
10502: #if defined(__SUNPRO_C) || defined(__SUNPRO_CC)
1.191 brouard 10503: printf(" Oracle Solaris Studio");if(logged)fprintf(ficlog," Oracle Solaris Studio\n");/* Oracle Solaris Studio. ----------------------------------- */
1.167 brouard 10504: #endif
1.191 brouard 10505: printf(" for "); if (logged) fprintf(ficlog, " for ");
1.169 brouard 10506:
1.167 brouard 10507: // http://stackoverflow.com/questions/4605842/how-to-identify-platform-compiler-from-preprocessor-macros
10508: #ifdef _WIN32 // note the underscore: without it, it's not msdn official!
10509: // Windows (x64 and x86)
1.191 brouard 10510: printf("Windows (x64 and x86) ");if(logged) fprintf(ficlog,"Windows (x64 and x86) ");
1.167 brouard 10511: #elif __unix__ // all unices, not all compilers
10512: // Unix
1.191 brouard 10513: printf("Unix ");if(logged) fprintf(ficlog,"Unix ");
1.167 brouard 10514: #elif __linux__
10515: // linux
1.191 brouard 10516: printf("linux ");if(logged) fprintf(ficlog,"linux ");
1.167 brouard 10517: #elif __APPLE__
1.174 brouard 10518: // Mac OS, not sure if this is covered by __posix__ and/or __unix__ though..
1.191 brouard 10519: printf("Mac OS ");if(logged) fprintf(ficlog,"Mac OS ");
1.167 brouard 10520: #endif
10521:
10522: /* __MINGW32__ */
10523: /* __CYGWIN__ */
10524: /* __MINGW64__ */
10525: // http://msdn.microsoft.com/en-us/library/b0084kay.aspx
10526: /* _MSC_VER //the Visual C++ compiler is 17.00.51106.1, the _MSC_VER macro evaluates to 1700. Type cl /? */
10527: /* _MSC_FULL_VER //the Visual C++ compiler is 15.00.20706.01, the _MSC_FULL_VER macro evaluates to 150020706 */
10528: /* _WIN64 // Defined for applications for Win64. */
10529: /* _M_X64 // Defined for compilations that target x64 processors. */
10530: /* _DEBUG // Defined when you compile with /LDd, /MDd, and /MTd. */
1.171 brouard 10531:
1.167 brouard 10532: #if UINTPTR_MAX == 0xffffffff
1.191 brouard 10533: printf(" 32-bit"); if(logged) fprintf(ficlog," 32-bit");/* 32-bit */
1.167 brouard 10534: #elif UINTPTR_MAX == 0xffffffffffffffff
1.191 brouard 10535: printf(" 64-bit"); if(logged) fprintf(ficlog," 64-bit");/* 64-bit */
1.167 brouard 10536: #else
1.191 brouard 10537: printf(" wtf-bit"); if(logged) fprintf(ficlog," wtf-bit");/* wtf */
1.167 brouard 10538: #endif
10539:
1.169 brouard 10540: #if defined(__GNUC__)
10541: # if defined(__GNUC_PATCHLEVEL__)
10542: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10543: + __GNUC_MINOR__ * 100 \
10544: + __GNUC_PATCHLEVEL__)
10545: # else
10546: # define __GNUC_VERSION__ (__GNUC__ * 10000 \
10547: + __GNUC_MINOR__ * 100)
10548: # endif
1.174 brouard 10549: printf(" using GNU C version %d.\n", __GNUC_VERSION__);
1.191 brouard 10550: if(logged) fprintf(ficlog, " using GNU C version %d.\n", __GNUC_VERSION__);
1.176 brouard 10551:
10552: if (uname(&sysInfo) != -1) {
10553: printf("Running on: %s %s %s %s %s\n",sysInfo.sysname, sysInfo.nodename, sysInfo.release, sysInfo.version, sysInfo.machine);
1.191 brouard 10554: 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 10555: }
10556: else
10557: perror("uname() error");
1.179 brouard 10558: //#ifndef __INTEL_COMPILER
10559: #if !defined (__INTEL_COMPILER) && !defined(__APPLE__)
1.174 brouard 10560: printf("GNU libc version: %s\n", gnu_get_libc_version());
1.191 brouard 10561: if(logged) fprintf(ficlog,"GNU libc version: %s\n", gnu_get_libc_version());
1.177 brouard 10562: #endif
1.169 brouard 10563: #endif
1.172 brouard 10564:
1.286 brouard 10565: // void main ()
1.172 brouard 10566: // {
1.169 brouard 10567: #if defined(_MSC_VER)
1.174 brouard 10568: if (IsWow64()){
1.191 brouard 10569: printf("\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
10570: if (logged) fprintf(ficlog, "\nThe program (probably compiled for 32bit) is running under WOW64 (64bit) emulation.\n");
1.174 brouard 10571: }
10572: else{
1.191 brouard 10573: printf("\nThe program is not running under WOW64 (i.e probably on a 64bit Windows).\n");
10574: if (logged) fprintf(ficlog, "\nThe programm is not running under WOW64 (i.e probably on a 64bit Windows).\n");
1.174 brouard 10575: }
1.172 brouard 10576: // printf("\nPress Enter to continue...");
10577: // getchar();
10578: // }
10579:
1.169 brouard 10580: #endif
10581:
1.167 brouard 10582:
1.219 brouard 10583: }
1.136 brouard 10584:
1.219 brouard 10585: int prevalence_limit(double *p, double **prlim, double ageminpar, double agemaxpar, double ftolpl, int *ncvyearp){
1.288 brouard 10586: /*--------------- Prevalence limit (forward period or forward stable prevalence) --------------*/
1.235 brouard 10587: int i, j, k, i1, k4=0, nres=0 ;
1.202 brouard 10588: /* double ftolpl = 1.e-10; */
1.180 brouard 10589: double age, agebase, agelim;
1.203 brouard 10590: double tot;
1.180 brouard 10591:
1.202 brouard 10592: strcpy(filerespl,"PL_");
10593: strcat(filerespl,fileresu);
10594: if((ficrespl=fopen(filerespl,"w"))==NULL) {
1.288 brouard 10595: printf("Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
10596: fprintf(ficlog,"Problem with forward period (stable) prevalence resultfile: %s\n", filerespl);return 1;
1.202 brouard 10597: }
1.288 brouard 10598: printf("\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
10599: fprintf(ficlog,"\nComputing forward period (stable) prevalence: result on file '%s' \n", filerespl);
1.202 brouard 10600: pstamp(ficrespl);
1.288 brouard 10601: fprintf(ficrespl,"# Forward period (stable) prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.202 brouard 10602: fprintf(ficrespl,"#Age ");
10603: for(i=1; i<=nlstate;i++) fprintf(ficrespl,"%d-%d ",i,i);
10604: fprintf(ficrespl,"\n");
1.180 brouard 10605:
1.219 brouard 10606: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
1.180 brouard 10607:
1.219 brouard 10608: agebase=ageminpar;
10609: agelim=agemaxpar;
1.180 brouard 10610:
1.227 brouard 10611: /* i1=pow(2,ncoveff); */
1.234 brouard 10612: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
1.219 brouard 10613: if (cptcovn < 1){i1=1;}
1.180 brouard 10614:
1.238 brouard 10615: for(k=1; k<=i1;k++){ /* For each combination k of dummy covariates in the model */
10616: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
1.253 brouard 10617: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10618: continue;
1.235 brouard 10619:
1.238 brouard 10620: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10621: /* for(cptcov=1,k=0;cptcov<=1;cptcov++){ */
10622: //for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){
10623: /* k=k+1; */
10624: /* to clean */
10625: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10626: fprintf(ficrespl,"#******");
10627: printf("#******");
10628: fprintf(ficlog,"#******");
10629: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10630: fprintf(ficrespl," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); /* Here problem for varying dummy*/
10631: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10632: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10633: }
10634: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10635: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10636: fprintf(ficrespl," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10637: fprintf(ficlog," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10638: }
10639: fprintf(ficrespl,"******\n");
10640: printf("******\n");
10641: fprintf(ficlog,"******\n");
10642: if(invalidvarcomb[k]){
10643: printf("\nCombination (%d) ignored because no case \n",k);
10644: fprintf(ficrespl,"#Combination (%d) ignored because no case \n",k);
10645: fprintf(ficlog,"\nCombination (%d) ignored because no case \n",k);
10646: continue;
10647: }
1.219 brouard 10648:
1.238 brouard 10649: fprintf(ficrespl,"#Age ");
10650: for(j=1;j<=cptcoveff;j++) {
10651: fprintf(ficrespl,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10652: }
10653: for(i=1; i<=nlstate;i++) fprintf(ficrespl," %d-%d ",i,i);
10654: fprintf(ficrespl,"Total Years_to_converge\n");
1.227 brouard 10655:
1.238 brouard 10656: for (age=agebase; age<=agelim; age++){
10657: /* for (age=agebase; age<=agebase; age++){ */
10658: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, ncvyearp, k, nres);
10659: fprintf(ficrespl,"%.0f ",age );
10660: for(j=1;j<=cptcoveff;j++)
10661: fprintf(ficrespl,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10662: tot=0.;
10663: for(i=1; i<=nlstate;i++){
10664: tot += prlim[i][i];
10665: fprintf(ficrespl," %.5f", prlim[i][i]);
10666: }
10667: fprintf(ficrespl," %.3f %d\n", tot, *ncvyearp);
10668: } /* Age */
10669: /* was end of cptcod */
10670: } /* cptcov */
10671: } /* nres */
1.219 brouard 10672: return 0;
1.180 brouard 10673: }
10674:
1.218 brouard 10675: 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 10676: /*--------------- Back Prevalence limit (backward stable prevalence) --------------*/
1.218 brouard 10677:
10678: /* Computes the back prevalence limit for any combination of covariate values
10679: * at any age between ageminpar and agemaxpar
10680: */
1.235 brouard 10681: int i, j, k, i1, nres=0 ;
1.217 brouard 10682: /* double ftolpl = 1.e-10; */
10683: double age, agebase, agelim;
10684: double tot;
1.218 brouard 10685: /* double ***mobaverage; */
10686: /* double **dnewm, **doldm, **dsavm; /\* for use *\/ */
1.217 brouard 10687:
10688: strcpy(fileresplb,"PLB_");
10689: strcat(fileresplb,fileresu);
10690: if((ficresplb=fopen(fileresplb,"w"))==NULL) {
1.288 brouard 10691: printf("Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
10692: fprintf(ficlog,"Problem with backward prevalence resultfile: %s\n", fileresplb);return 1;
1.217 brouard 10693: }
1.288 brouard 10694: printf("Computing backward prevalence: result on file '%s' \n", fileresplb);
10695: fprintf(ficlog,"Computing backward prevalence: result on file '%s' \n", fileresplb);
1.217 brouard 10696: pstamp(ficresplb);
1.288 brouard 10697: fprintf(ficresplb,"# Backward prevalence. Precision given by ftolpl=%g \n", ftolpl);
1.217 brouard 10698: fprintf(ficresplb,"#Age ");
10699: for(i=1; i<=nlstate;i++) fprintf(ficresplb,"%d-%d ",i,i);
10700: fprintf(ficresplb,"\n");
10701:
1.218 brouard 10702:
10703: /* prlim=matrix(1,nlstate,1,nlstate);*/ /* back in main */
10704:
10705: agebase=ageminpar;
10706: agelim=agemaxpar;
10707:
10708:
1.227 brouard 10709: i1=pow(2,cptcoveff);
1.218 brouard 10710: if (cptcovn < 1){i1=1;}
1.227 brouard 10711:
1.238 brouard 10712: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10713: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10714: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10715: continue;
10716: //printf("cptcov=%d cptcod=%d codtab=%d\n",cptcov, cptcod,codtabm(cptcod,cptcov));
10717: fprintf(ficresplb,"#******");
10718: printf("#******");
10719: fprintf(ficlog,"#******");
10720: for(j=1;j<=cptcoveff ;j++) {/* all covariates */
10721: fprintf(ficresplb," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10722: printf(" V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10723: fprintf(ficlog," V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10724: }
10725: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10726: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10727: fprintf(ficresplb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10728: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10729: }
10730: fprintf(ficresplb,"******\n");
10731: printf("******\n");
10732: fprintf(ficlog,"******\n");
10733: if(invalidvarcomb[k]){
10734: printf("\nCombination (%d) ignored because no cases \n",k);
10735: fprintf(ficresplb,"#Combination (%d) ignored because no cases \n",k);
10736: fprintf(ficlog,"\nCombination (%d) ignored because no cases \n",k);
10737: continue;
10738: }
1.218 brouard 10739:
1.238 brouard 10740: fprintf(ficresplb,"#Age ");
10741: for(j=1;j<=cptcoveff;j++) {
10742: fprintf(ficresplb,"V%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10743: }
10744: for(i=1; i<=nlstate;i++) fprintf(ficresplb," %d-%d ",i,i);
10745: fprintf(ficresplb,"Total Years_to_converge\n");
1.218 brouard 10746:
10747:
1.238 brouard 10748: for (age=agebase; age<=agelim; age++){
10749: /* for (age=agebase; age<=agebase; age++){ */
10750: if(mobilavproj > 0){
10751: /* bprevalim(bprlim, mobaverage, nlstate, p, age, ageminpar, agemaxpar, oldm, savm, doldm, dsavm, ftolpl, ncvyearp, k); */
10752: /* bprevalim(bprlim, mobaverage, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10753: bprevalim(bprlim, mobaverage, nlstate, p, age, ftolpl, ncvyearp, k, nres);
1.238 brouard 10754: }else if (mobilavproj == 0){
10755: 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);
10756: 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);
10757: exit(1);
10758: }else{
10759: /* bprevalim(bprlim, probs, nlstate, p, age, oldm, savm, dnewm, doldm, dsavm, ftolpl, ncvyearp, k); */
1.242 brouard 10760: bprevalim(bprlim, probs, nlstate, p, age, ftolpl, ncvyearp, k,nres);
1.266 brouard 10761: /* printf("TOTOT\n"); */
10762: /* exit(1); */
1.238 brouard 10763: }
10764: fprintf(ficresplb,"%.0f ",age );
10765: for(j=1;j<=cptcoveff;j++)
10766: fprintf(ficresplb,"%d %d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10767: tot=0.;
10768: for(i=1; i<=nlstate;i++){
10769: tot += bprlim[i][i];
10770: fprintf(ficresplb," %.5f", bprlim[i][i]);
10771: }
10772: fprintf(ficresplb," %.3f %d\n", tot, *ncvyearp);
10773: } /* Age */
10774: /* was end of cptcod */
1.255 brouard 10775: /*fprintf(ficresplb,"\n");*/ /* Seems to be necessary for gnuplot only if two result lines and no covariate. */
1.238 brouard 10776: } /* end of any combination */
10777: } /* end of nres */
1.218 brouard 10778: /* hBijx(p, bage, fage); */
10779: /* fclose(ficrespijb); */
10780:
10781: return 0;
1.217 brouard 10782: }
1.218 brouard 10783:
1.180 brouard 10784: int hPijx(double *p, int bage, int fage){
10785: /*------------- h Pij x at various ages ------------*/
10786:
10787: int stepsize;
10788: int agelim;
10789: int hstepm;
10790: int nhstepm;
1.235 brouard 10791: int h, i, i1, j, k, k4, nres=0;
1.180 brouard 10792:
10793: double agedeb;
10794: double ***p3mat;
10795:
1.201 brouard 10796: strcpy(filerespij,"PIJ_"); strcat(filerespij,fileresu);
1.180 brouard 10797: if((ficrespij=fopen(filerespij,"w"))==NULL) {
10798: printf("Problem with Pij resultfile: %s\n", filerespij); return 1;
10799: fprintf(ficlog,"Problem with Pij resultfile: %s\n", filerespij); return 1;
10800: }
10801: printf("Computing pij: result on file '%s' \n", filerespij);
10802: fprintf(ficlog,"Computing pij: result on file '%s' \n", filerespij);
10803:
10804: stepsize=(int) (stepm+YEARM-1)/YEARM;
10805: /*if (stepm<=24) stepsize=2;*/
10806:
10807: agelim=AGESUP;
10808: hstepm=stepsize*YEARM; /* Every year of age */
10809: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
1.218 brouard 10810:
1.180 brouard 10811: /* hstepm=1; aff par mois*/
10812: pstamp(ficrespij);
10813: fprintf(ficrespij,"#****** h Pij x Probability to be in state j at age x+h being in i at x ");
1.227 brouard 10814: i1= pow(2,cptcoveff);
1.218 brouard 10815: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10816: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10817: /* k=k+1; */
1.235 brouard 10818: for(nres=1; nres <= nresult; nres++) /* For each resultline */
10819: for(k=1; k<=i1;k++){
1.253 brouard 10820: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 10821: continue;
1.183 brouard 10822: fprintf(ficrespij,"\n#****** ");
1.227 brouard 10823: for(j=1;j<=cptcoveff;j++)
1.198 brouard 10824: fprintf(ficrespij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 10825: for (k4=1; k4<= nsq; k4++){ /* For each selected (single) quantitative value */
10826: printf(" V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10827: fprintf(ficrespij," V%d=%f ",Tvqresult[nres][k4],Tqresult[nres][k4]);
10828: }
1.183 brouard 10829: fprintf(ficrespij,"******\n");
10830:
10831: for (agedeb=fage; agedeb>=bage; agedeb--){ /* If stepm=6 months */
10832: nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /* Typically 20 years = 20*12/6=40 */
10833: nhstepm = nhstepm/hstepm; /* Typically 40/4=10 */
10834:
10835: /* nhstepm=nhstepm*YEARM; aff par mois*/
1.180 brouard 10836:
1.183 brouard 10837: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10838: oldm=oldms;savm=savms;
1.235 brouard 10839: hpxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k, nres);
1.183 brouard 10840: fprintf(ficrespij,"# Cov Agex agex+h hpijx with i,j=");
10841: for(i=1; i<=nlstate;i++)
10842: for(j=1; j<=nlstate+ndeath;j++)
10843: fprintf(ficrespij," %1d-%1d",i,j);
10844: fprintf(ficrespij,"\n");
10845: for (h=0; h<=nhstepm; h++){
10846: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10847: fprintf(ficrespij,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm );
1.180 brouard 10848: for(i=1; i<=nlstate;i++)
10849: for(j=1; j<=nlstate+ndeath;j++)
1.183 brouard 10850: fprintf(ficrespij," %.5f", p3mat[i][j][h]);
1.180 brouard 10851: fprintf(ficrespij,"\n");
10852: }
1.183 brouard 10853: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10854: fprintf(ficrespij,"\n");
10855: }
1.180 brouard 10856: /*}*/
10857: }
1.218 brouard 10858: return 0;
1.180 brouard 10859: }
1.218 brouard 10860:
10861: int hBijx(double *p, int bage, int fage, double ***prevacurrent){
1.217 brouard 10862: /*------------- h Bij x at various ages ------------*/
10863:
10864: int stepsize;
1.218 brouard 10865: /* int agelim; */
10866: int ageminl;
1.217 brouard 10867: int hstepm;
10868: int nhstepm;
1.238 brouard 10869: int h, i, i1, j, k, nres;
1.218 brouard 10870:
1.217 brouard 10871: double agedeb;
10872: double ***p3mat;
1.218 brouard 10873:
10874: strcpy(filerespijb,"PIJB_"); strcat(filerespijb,fileresu);
10875: if((ficrespijb=fopen(filerespijb,"w"))==NULL) {
10876: printf("Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10877: fprintf(ficlog,"Problem with Pij back resultfile: %s\n", filerespijb); return 1;
10878: }
10879: printf("Computing pij back: result on file '%s' \n", filerespijb);
10880: fprintf(ficlog,"Computing pij back: result on file '%s' \n", filerespijb);
10881:
10882: stepsize=(int) (stepm+YEARM-1)/YEARM;
10883: /*if (stepm<=24) stepsize=2;*/
1.217 brouard 10884:
1.218 brouard 10885: /* agelim=AGESUP; */
1.289 brouard 10886: ageminl=AGEINF; /* was 30 */
1.218 brouard 10887: hstepm=stepsize*YEARM; /* Every year of age */
10888: hstepm=hstepm/stepm; /* Typically 2 years, = 2/6 months = 4 */
10889:
10890: /* hstepm=1; aff par mois*/
10891: pstamp(ficrespijb);
1.255 brouard 10892: 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 10893: i1= pow(2,cptcoveff);
1.218 brouard 10894: /* for(cptcov=1,k=0;cptcov<=i1;cptcov++){ */
10895: /* /\*for(cptcod=1;cptcod<=ncodemax[cptcov];cptcod++){*\/ */
10896: /* k=k+1; */
1.238 brouard 10897: for(nres=1; nres <= nresult; nres++){ /* For each resultline */
10898: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 10899: if(i1 != 1 && TKresult[nres]!= k)
1.238 brouard 10900: continue;
10901: fprintf(ficrespijb,"\n#****** ");
10902: for(j=1;j<=cptcoveff;j++)
10903: fprintf(ficrespijb,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
10904: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
10905: fprintf(ficrespijb," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
10906: }
10907: fprintf(ficrespijb,"******\n");
1.264 brouard 10908: if(invalidvarcomb[k]){ /* Is it necessary here? */
1.238 brouard 10909: fprintf(ficrespijb,"\n#Combination (%d) ignored because no cases \n",k);
10910: continue;
10911: }
10912:
10913: /* for (agedeb=fage; agedeb>=bage; agedeb--){ /\* If stepm=6 months *\/ */
10914: for (agedeb=bage; agedeb<=fage; agedeb++){ /* If stepm=6 months and estepm=24 (2 years) */
10915: /* nhstepm=(int) rint((agelim-agedeb)*YEARM/stepm); /\* Typically 20 years = 20*12/6=40 *\/ */
1.297 brouard 10916: 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 */
10917: nhstepm = nhstepm/hstepm; /* Typically 40/4=10, because estepm=24 stepm=6 => hstepm=24/6=4 or 28*/
1.238 brouard 10918:
10919: /* nhstepm=nhstepm*YEARM; aff par mois*/
10920:
1.266 brouard 10921: p3mat=ma3x(1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm); /* We can't have it at an upper level because of nhstepm */
10922: /* and memory limitations if stepm is small */
10923:
1.238 brouard 10924: /* oldm=oldms;savm=savms; */
10925: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,nlstate,stepm,oldm,savm, k); */
1.267 brouard 10926: hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm, k, nres);
1.238 brouard 10927: /* hbxij(p3mat,nhstepm,agedeb,hstepm,p,prevacurrent,nlstate,stepm,oldm,savm, dnewm, doldm, dsavm, k); */
1.255 brouard 10928: fprintf(ficrespijb,"# Cov Agex agex-h hbijx with i,j=");
1.217 brouard 10929: for(i=1; i<=nlstate;i++)
10930: for(j=1; j<=nlstate+ndeath;j++)
1.238 brouard 10931: fprintf(ficrespijb," %1d-%1d",i,j);
1.217 brouard 10932: fprintf(ficrespijb,"\n");
1.238 brouard 10933: for (h=0; h<=nhstepm; h++){
10934: /*agedebphstep = agedeb + h*hstepm/YEARM*stepm;*/
10935: fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb - h*hstepm/YEARM*stepm );
10936: /* fprintf(ficrespijb,"%d %3.f %3.f",k, agedeb, agedeb + h*hstepm/YEARM*stepm ); */
10937: for(i=1; i<=nlstate;i++)
10938: for(j=1; j<=nlstate+ndeath;j++)
10939: fprintf(ficrespijb," %.5f", p3mat[i][j][h]);
10940: fprintf(ficrespijb,"\n");
10941: }
10942: free_ma3x(p3mat,1,nlstate+ndeath,1, nlstate+ndeath, 0,nhstepm);
10943: fprintf(ficrespijb,"\n");
10944: } /* end age deb */
10945: } /* end combination */
10946: } /* end nres */
1.218 brouard 10947: return 0;
10948: } /* hBijx */
1.217 brouard 10949:
1.180 brouard 10950:
1.136 brouard 10951: /***********************************************/
10952: /**************** Main Program *****************/
10953: /***********************************************/
10954:
10955: int main(int argc, char *argv[])
10956: {
10957: #ifdef GSL
10958: const gsl_multimin_fminimizer_type *T;
10959: size_t iteri = 0, it;
10960: int rval = GSL_CONTINUE;
10961: int status = GSL_SUCCESS;
10962: double ssval;
10963: #endif
10964: int movingaverage(double ***probs, double bage,double fage, double ***mobaverage, int mobilav);
1.290 brouard 10965: int i,j, k, iter=0,m,size=100, cptcod; /* Suppressing because nobs */
10966: /* int i,j, k, n=MAXN,iter=0,m,size=100, cptcod; */
1.209 brouard 10967: int ncvyear=0; /* Number of years needed for the period prevalence to converge */
1.164 brouard 10968: int jj, ll, li, lj, lk;
1.136 brouard 10969: int numlinepar=0; /* Current linenumber of parameter file */
1.197 brouard 10970: int num_filled;
1.136 brouard 10971: int itimes;
10972: int NDIM=2;
10973: int vpopbased=0;
1.235 brouard 10974: int nres=0;
1.258 brouard 10975: int endishere=0;
1.277 brouard 10976: int noffset=0;
1.274 brouard 10977: int ncurrv=0; /* Temporary variable */
10978:
1.164 brouard 10979: char ca[32], cb[32];
1.136 brouard 10980: /* FILE *fichtm; *//* Html File */
10981: /* FILE *ficgp;*/ /*Gnuplot File */
10982: struct stat info;
1.191 brouard 10983: double agedeb=0.;
1.194 brouard 10984:
10985: double ageminpar=AGEOVERFLOW,agemin=AGEOVERFLOW, agemaxpar=-AGEOVERFLOW, agemax=-AGEOVERFLOW;
1.219 brouard 10986: double ageminout=-AGEOVERFLOW,agemaxout=AGEOVERFLOW; /* Smaller Age range redefined after movingaverage */
1.136 brouard 10987:
1.165 brouard 10988: double fret;
1.191 brouard 10989: double dum=0.; /* Dummy variable */
1.136 brouard 10990: double ***p3mat;
1.218 brouard 10991: /* double ***mobaverage; */
1.164 brouard 10992:
10993: char line[MAXLINE];
1.197 brouard 10994: char path[MAXLINE],pathc[MAXLINE],pathcd[MAXLINE],pathtot[MAXLINE];
10995:
1.234 brouard 10996: char modeltemp[MAXLINE];
1.230 brouard 10997: char resultline[MAXLINE];
10998:
1.136 brouard 10999: char pathr[MAXLINE], pathimach[MAXLINE];
1.164 brouard 11000: char *tok, *val; /* pathtot */
1.290 brouard 11001: int firstobs=1, lastobs=10; /* nobs = lastobs-firstobs declared globally ;*/
1.195 brouard 11002: int c, h , cpt, c2;
1.191 brouard 11003: int jl=0;
11004: int i1, j1, jk, stepsize=0;
1.194 brouard 11005: int count=0;
11006:
1.164 brouard 11007: int *tab;
1.136 brouard 11008: int mobilavproj=0 , prevfcast=0 ; /* moving average of prev, If prevfcast=1 prevalence projection */
1.296 brouard 11009: /* double anprojd, mprojd, jprojd; /\* For eventual projections *\/ */
11010: /* double anprojf, mprojf, jprojf; */
11011: /* double jintmean,mintmean,aintmean; */
11012: int prvforecast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11013: int prvbackcast = 0; /* Might be 1 (date of beginning of projection is a choice or 2 is the dateintmean */
11014: double yrfproj= 10.0; /* Number of years of forward projections */
11015: double yrbproj= 10.0; /* Number of years of backward projections */
11016: int prevbcast=0; /* defined as global for mlikeli and mle, replacing backcast */
1.136 brouard 11017: int mobilav=0,popforecast=0;
1.191 brouard 11018: int hstepm=0, nhstepm=0;
1.136 brouard 11019: int agemortsup;
11020: float sumlpop=0.;
11021: double jprev1=1, mprev1=1,anprev1=2000,jprev2=1, mprev2=1,anprev2=2000;
11022: double jpyram=1, mpyram=1,anpyram=2000,jpyram1=1, mpyram1=1,anpyram1=2000;
11023:
1.191 brouard 11024: double bage=0, fage=110., age, agelim=0., agebase=0.;
1.136 brouard 11025: double ftolpl=FTOL;
11026: double **prlim;
1.217 brouard 11027: double **bprlim;
1.136 brouard 11028: double ***param; /* Matrix of parameters */
1.251 brouard 11029: double ***paramstart; /* Matrix of starting parameter values */
11030: double *p, *pstart; /* p=param[1][1] pstart is for starting values guessed by freqsummary */
1.136 brouard 11031: double **matcov; /* Matrix of covariance */
1.203 brouard 11032: double **hess; /* Hessian matrix */
1.136 brouard 11033: double ***delti3; /* Scale */
11034: double *delti; /* Scale */
11035: double ***eij, ***vareij;
11036: double **varpl; /* Variances of prevalence limits by age */
1.269 brouard 11037:
1.136 brouard 11038: double *epj, vepp;
1.164 brouard 11039:
1.273 brouard 11040: double dateprev1, dateprev2;
1.296 brouard 11041: double jproj1=1,mproj1=1,anproj1=2000,jproj2=1,mproj2=1,anproj2=2000, dateproj1=0, dateproj2=0, dateprojd=0, dateprojf=0;
11042: double jback1=1,mback1=1,anback1=2000,jback2=1,mback2=1,anback2=2000, dateback1=0, dateback2=0, datebackd=0, datebackf=0;
11043:
1.217 brouard 11044:
1.136 brouard 11045: double **ximort;
1.145 brouard 11046: char *alph[]={"a","a","b","c","d","e"}, str[4]="1234";
1.136 brouard 11047: int *dcwave;
11048:
1.164 brouard 11049: char z[1]="c";
1.136 brouard 11050:
11051: /*char *strt;*/
11052: char strtend[80];
1.126 brouard 11053:
1.164 brouard 11054:
1.126 brouard 11055: /* setlocale (LC_ALL, ""); */
11056: /* bindtextdomain (PACKAGE, LOCALEDIR); */
11057: /* textdomain (PACKAGE); */
11058: /* setlocale (LC_CTYPE, ""); */
11059: /* setlocale (LC_MESSAGES, ""); */
11060:
11061: /* gettimeofday(&start_time, (struct timezone*)0); */ /* at first time */
1.157 brouard 11062: rstart_time = time(NULL);
11063: /* (void) gettimeofday(&start_time,&tzp);*/
11064: start_time = *localtime(&rstart_time);
1.126 brouard 11065: curr_time=start_time;
1.157 brouard 11066: /*tml = *localtime(&start_time.tm_sec);*/
11067: /* strcpy(strstart,asctime(&tml)); */
11068: strcpy(strstart,asctime(&start_time));
1.126 brouard 11069:
11070: /* printf("Localtime (at start)=%s",strstart); */
1.157 brouard 11071: /* tp.tm_sec = tp.tm_sec +86400; */
11072: /* tm = *localtime(&start_time.tm_sec); */
1.126 brouard 11073: /* tmg.tm_year=tmg.tm_year +dsign*dyear; */
11074: /* tmg.tm_mon=tmg.tm_mon +dsign*dmonth; */
11075: /* tmg.tm_hour=tmg.tm_hour + 1; */
1.157 brouard 11076: /* tp.tm_sec = mktime(&tmg); */
1.126 brouard 11077: /* strt=asctime(&tmg); */
11078: /* printf("Time(after) =%s",strstart); */
11079: /* (void) time (&time_value);
11080: * printf("time=%d,t-=%d\n",time_value,time_value-86400);
11081: * tm = *localtime(&time_value);
11082: * strstart=asctime(&tm);
11083: * printf("tim_value=%d,asctime=%s\n",time_value,strstart);
11084: */
11085:
11086: nberr=0; /* Number of errors and warnings */
11087: nbwarn=0;
1.184 brouard 11088: #ifdef WIN32
11089: _getcwd(pathcd, size);
11090: #else
1.126 brouard 11091: getcwd(pathcd, size);
1.184 brouard 11092: #endif
1.191 brouard 11093: syscompilerinfo(0);
1.196 brouard 11094: printf("\nIMaCh version %s, %s\n%s",version, copyright, fullversion);
1.126 brouard 11095: if(argc <=1){
11096: printf("\nEnter the parameter file name: ");
1.205 brouard 11097: if(!fgets(pathr,FILENAMELENGTH,stdin)){
11098: printf("ERROR Empty parameter file name\n");
11099: goto end;
11100: }
1.126 brouard 11101: i=strlen(pathr);
11102: if(pathr[i-1]=='\n')
11103: pathr[i-1]='\0';
1.156 brouard 11104: i=strlen(pathr);
1.205 brouard 11105: if(i >= 1 && pathr[i-1]==' ') {/* This may happen when dragging on oS/X! */
1.156 brouard 11106: pathr[i-1]='\0';
1.205 brouard 11107: }
11108: i=strlen(pathr);
11109: if( i==0 ){
11110: printf("ERROR Empty parameter file name\n");
11111: goto end;
11112: }
11113: for (tok = pathr; tok != NULL; ){
1.126 brouard 11114: printf("Pathr |%s|\n",pathr);
11115: while ((val = strsep(&tok, "\"" )) != NULL && *val == '\0');
11116: printf("val= |%s| pathr=%s\n",val,pathr);
11117: strcpy (pathtot, val);
11118: if(pathr[0] == '\0') break; /* Dirty */
11119: }
11120: }
1.281 brouard 11121: else if (argc<=2){
11122: strcpy(pathtot,argv[1]);
11123: }
1.126 brouard 11124: else{
11125: strcpy(pathtot,argv[1]);
1.281 brouard 11126: strcpy(z,argv[2]);
11127: printf("\nargv[2]=%s z=%c\n",argv[2],z[0]);
1.126 brouard 11128: }
11129: /*if(getcwd(pathcd, MAXLINE)!= NULL)printf ("Error pathcd\n");*/
11130: /*cygwin_split_path(pathtot,path,optionfile);
11131: printf("pathtot=%s, path=%s, optionfile=%s\n",pathtot,path,optionfile);*/
11132: /* cutv(path,optionfile,pathtot,'\\');*/
11133:
11134: /* Split argv[0], imach program to get pathimach */
11135: printf("\nargv[0]=%s argv[1]=%s, \n",argv[0],argv[1]);
11136: split(argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11137: printf("\nargv[0]=%s pathimach=%s, \noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",argv[0],pathimach,optionfile,optionfilext,optionfilefiname);
11138: /* strcpy(pathimach,argv[0]); */
11139: /* Split argv[1]=pathtot, parameter file name to get path, optionfile, extension and name */
11140: split(pathtot,path,optionfile,optionfilext,optionfilefiname);
11141: printf("\npathtot=%s,\npath=%s,\noptionfile=%s \noptionfilext=%s \noptionfilefiname=%s\n",pathtot,path,optionfile,optionfilext,optionfilefiname);
1.184 brouard 11142: #ifdef WIN32
11143: _chdir(path); /* Can be a relative path */
11144: if(_getcwd(pathcd,MAXLINE) > 0) /* So pathcd is the full path */
11145: #else
1.126 brouard 11146: chdir(path); /* Can be a relative path */
1.184 brouard 11147: if (getcwd(pathcd, MAXLINE) > 0) /* So pathcd is the full path */
11148: #endif
11149: printf("Current directory %s!\n",pathcd);
1.126 brouard 11150: strcpy(command,"mkdir ");
11151: strcat(command,optionfilefiname);
11152: if((outcmd=system(command)) != 0){
1.169 brouard 11153: printf("Directory already exists (or can't create it) %s%s, err=%d\n",path,optionfilefiname,outcmd);
1.126 brouard 11154: /* fprintf(ficlog,"Problem creating directory %s%s\n",path,optionfilefiname); */
11155: /* fclose(ficlog); */
11156: /* exit(1); */
11157: }
11158: /* if((imk=mkdir(optionfilefiname))<0){ */
11159: /* perror("mkdir"); */
11160: /* } */
11161:
11162: /*-------- arguments in the command line --------*/
11163:
1.186 brouard 11164: /* Main Log file */
1.126 brouard 11165: strcat(filelog, optionfilefiname);
11166: strcat(filelog,".log"); /* */
11167: if((ficlog=fopen(filelog,"w"))==NULL) {
11168: printf("Problem with logfile %s\n",filelog);
11169: goto end;
11170: }
11171: fprintf(ficlog,"Log filename:%s\n",filelog);
1.197 brouard 11172: fprintf(ficlog,"Version %s %s",version,fullversion);
1.126 brouard 11173: fprintf(ficlog,"\nEnter the parameter file name: \n");
11174: fprintf(ficlog,"pathimach=%s\npathtot=%s\n\
11175: path=%s \n\
11176: optionfile=%s\n\
11177: optionfilext=%s\n\
1.156 brouard 11178: optionfilefiname='%s'\n",pathimach,pathtot,path,optionfile,optionfilext,optionfilefiname);
1.126 brouard 11179:
1.197 brouard 11180: syscompilerinfo(1);
1.167 brouard 11181:
1.126 brouard 11182: printf("Local time (at start):%s",strstart);
11183: fprintf(ficlog,"Local time (at start): %s",strstart);
11184: fflush(ficlog);
11185: /* (void) gettimeofday(&curr_time,&tzp); */
1.157 brouard 11186: /* printf("Elapsed time %d\n", asc_diff_time(curr_time.tm_sec-start_time.tm_sec,tmpout)); */
1.126 brouard 11187:
11188: /* */
11189: strcpy(fileres,"r");
11190: strcat(fileres, optionfilefiname);
1.201 brouard 11191: strcat(fileresu, optionfilefiname); /* Without r in front */
1.126 brouard 11192: strcat(fileres,".txt"); /* Other files have txt extension */
1.201 brouard 11193: strcat(fileresu,".txt"); /* Other files have txt extension */
1.126 brouard 11194:
1.186 brouard 11195: /* Main ---------arguments file --------*/
1.126 brouard 11196:
11197: if((ficpar=fopen(optionfile,"r"))==NULL) {
1.155 brouard 11198: printf("Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
11199: fprintf(ficlog,"Problem with optionfile '%s' with errno='%s'\n",optionfile,strerror(errno));
1.126 brouard 11200: fflush(ficlog);
1.149 brouard 11201: /* goto end; */
11202: exit(70);
1.126 brouard 11203: }
11204:
11205: strcpy(filereso,"o");
1.201 brouard 11206: strcat(filereso,fileresu);
1.126 brouard 11207: if((ficparo=fopen(filereso,"w"))==NULL) { /* opened on subdirectory */
11208: printf("Problem with Output resultfile: %s\n", filereso);
11209: fprintf(ficlog,"Problem with Output resultfile: %s\n", filereso);
11210: fflush(ficlog);
11211: goto end;
11212: }
1.278 brouard 11213: /*-------- Rewriting parameter file ----------*/
11214: strcpy(rfileres,"r"); /* "Rparameterfile */
11215: strcat(rfileres,optionfilefiname); /* Parameter file first name */
11216: strcat(rfileres,"."); /* */
11217: strcat(rfileres,optionfilext); /* Other files have txt extension */
11218: if((ficres =fopen(rfileres,"w"))==NULL) {
11219: printf("Problem writing new parameter file: %s\n", rfileres);goto end;
11220: fprintf(ficlog,"Problem writing new parameter file: %s\n", rfileres);goto end;
11221: fflush(ficlog);
11222: goto end;
11223: }
11224: fprintf(ficres,"#IMaCh %s\n",version);
1.126 brouard 11225:
1.278 brouard 11226:
1.126 brouard 11227: /* Reads comments: lines beginning with '#' */
11228: numlinepar=0;
1.277 brouard 11229: /* Is it a BOM UTF-8 Windows file? */
11230: /* First parameter line */
1.197 brouard 11231: while(fgets(line, MAXLINE, ficpar)) {
1.277 brouard 11232: noffset=0;
11233: if( line[0] == (char)0xEF && line[1] == (char)0xBB) /* EF BB BF */
11234: {
11235: noffset=noffset+3;
11236: printf("# File is an UTF8 Bom.\n"); // 0xBF
11237: }
1.302 brouard 11238: /* else if( line[0] == (char)0xFE && line[1] == (char)0xFF)*/
11239: else if( line[0] == (char)0xFF && line[1] == (char)0xFE)
1.277 brouard 11240: {
11241: noffset=noffset+2;
11242: printf("# File is an UTF16BE BOM file\n");
11243: }
11244: else if( line[0] == 0 && line[1] == 0)
11245: {
11246: if( line[2] == (char)0xFE && line[3] == (char)0xFF){
11247: noffset=noffset+4;
11248: printf("# File is an UTF16BE BOM file\n");
11249: }
11250: } else{
11251: ;/*printf(" Not a BOM file\n");*/
11252: }
11253:
1.197 brouard 11254: /* If line starts with a # it is a comment */
1.277 brouard 11255: if (line[noffset] == '#') {
1.197 brouard 11256: numlinepar++;
11257: fputs(line,stdout);
11258: fputs(line,ficparo);
1.278 brouard 11259: fputs(line,ficres);
1.197 brouard 11260: fputs(line,ficlog);
11261: continue;
11262: }else
11263: break;
11264: }
11265: if((num_filled=sscanf(line,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", \
11266: title, datafile, &lastobs, &firstpass,&lastpass)) !=EOF){
11267: if (num_filled != 5) {
11268: printf("Should be 5 parameters\n");
1.283 brouard 11269: fprintf(ficlog,"Should be 5 parameters\n");
1.197 brouard 11270: }
1.126 brouard 11271: numlinepar++;
1.197 brouard 11272: printf("title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.283 brouard 11273: fprintf(ficparo,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11274: fprintf(ficres,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
11275: fprintf(ficlog,"title=%s datafile=%s lastobs=%d firstpass=%d lastpass=%d\n", title, datafile, lastobs, firstpass,lastpass);
1.197 brouard 11276: }
11277: /* Second parameter line */
11278: while(fgets(line, MAXLINE, ficpar)) {
1.283 brouard 11279: /* while(fscanf(ficpar,"%[^\n]", line)) { */
11280: /* If line starts with a # it is a comment. Strangely fgets reads the EOL and fputs doesn't */
1.197 brouard 11281: if (line[0] == '#') {
11282: numlinepar++;
1.283 brouard 11283: printf("%s",line);
11284: fprintf(ficres,"%s",line);
11285: fprintf(ficparo,"%s",line);
11286: fprintf(ficlog,"%s",line);
1.197 brouard 11287: continue;
11288: }else
11289: break;
11290: }
1.223 brouard 11291: 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", \
11292: &ftol, &stepm, &ncovcol, &nqv, &ntv, &nqtv, &nlstate, &ndeath, &maxwav, &mle, &weightopt)) !=EOF){
11293: if (num_filled != 11) {
11294: 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 11295: printf("but line=%s\n",line);
1.283 brouard 11296: 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");
11297: fprintf(ficlog,"but line=%s\n",line);
1.197 brouard 11298: }
1.286 brouard 11299: if( lastpass > maxwav){
11300: printf("Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11301: fprintf(ficlog,"Error (lastpass = %d) > (maxwav = %d)\n",lastpass, maxwav);
11302: fflush(ficlog);
11303: goto end;
11304: }
11305: 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 11306: 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 11307: 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 11308: 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 11309: }
1.203 brouard 11310: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
1.209 brouard 11311: /*ftolpl=6.e-4; *//* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
1.197 brouard 11312: /* Third parameter line */
11313: while(fgets(line, MAXLINE, ficpar)) {
11314: /* If line starts with a # it is a comment */
11315: if (line[0] == '#') {
11316: numlinepar++;
1.283 brouard 11317: printf("%s",line);
11318: fprintf(ficres,"%s",line);
11319: fprintf(ficparo,"%s",line);
11320: fprintf(ficlog,"%s",line);
1.197 brouard 11321: continue;
11322: }else
11323: break;
11324: }
1.201 brouard 11325: if((num_filled=sscanf(line,"model=1+age%[^.\n]", model)) !=EOF){
1.279 brouard 11326: if (num_filled != 1){
1.302 brouard 11327: printf("ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
11328: fprintf(ficlog,"ERROR %d: Model should be at minimum 'model=1+age+' instead of '%s'\n",num_filled, line);
1.197 brouard 11329: model[0]='\0';
11330: goto end;
11331: }
11332: else{
11333: if (model[0]=='+'){
11334: for(i=1; i<=strlen(model);i++)
11335: modeltemp[i-1]=model[i];
1.201 brouard 11336: strcpy(model,modeltemp);
1.197 brouard 11337: }
11338: }
1.199 brouard 11339: /* printf(" model=1+age%s modeltemp= %s, model=%s\n",model, modeltemp, model);fflush(stdout); */
1.203 brouard 11340: printf("model=1+age+%s\n",model);fflush(stdout);
1.283 brouard 11341: fprintf(ficparo,"model=1+age+%s\n",model);fflush(stdout);
11342: fprintf(ficres,"model=1+age+%s\n",model);fflush(stdout);
11343: fprintf(ficlog,"model=1+age+%s\n",model);fflush(stdout);
1.197 brouard 11344: }
11345: /* 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); */
11346: /* numlinepar=numlinepar+3; /\* In general *\/ */
11347: /* 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 11348: /* 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); */
11349: /* 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 11350: fflush(ficlog);
1.190 brouard 11351: /* if(model[0]=='#'|| model[0]== '\0'){ */
11352: if(model[0]=='#'){
1.279 brouard 11353: printf("Error in 'model' line: model should start with 'model=1+age+' and end without space \n \
11354: 'model=1+age+' or 'model=1+age+V1.' or 'model=1+age+age*age+V1+V1*age' or \n \
11355: 'model=1+age+V1+V2' or 'model=1+age+V1+V2+V1*V2' etc. \n"); \
1.187 brouard 11356: if(mle != -1){
1.279 brouard 11357: 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 11358: exit(1);
11359: }
11360: }
1.126 brouard 11361: while((c=getc(ficpar))=='#' && c!= EOF){
11362: ungetc(c,ficpar);
11363: fgets(line, MAXLINE, ficpar);
11364: numlinepar++;
1.195 brouard 11365: if(line[1]=='q'){ /* This #q will quit imach (the answer is q) */
11366: z[0]=line[1];
11367: }
11368: /* printf("****line [1] = %c \n",line[1]); */
1.141 brouard 11369: fputs(line, stdout);
11370: //puts(line);
1.126 brouard 11371: fputs(line,ficparo);
11372: fputs(line,ficlog);
11373: }
11374: ungetc(c,ficpar);
11375:
11376:
1.290 brouard 11377: covar=matrix(0,NCOVMAX,firstobs,lastobs); /**< used in readdata */
11378: if(nqv>=1)coqvar=matrix(1,nqv,firstobs,lastobs); /**< Fixed quantitative covariate */
11379: if(nqtv>=1)cotqvar=ma3x(1,maxwav,1,nqtv,firstobs,lastobs); /**< Time varying quantitative covariate */
11380: if(ntv+nqtv>=1)cotvar=ma3x(1,maxwav,1,ntv+nqtv,firstobs,lastobs); /**< Time varying covariate (dummy and quantitative)*/
1.136 brouard 11381: cptcovn=0; /*Number of covariates, i.e. number of '+' in model statement plus one, indepently of n in Vn*/
11382: /* v1+v2+v3+v2*v4+v5*age makes cptcovn = 5
11383: v1+v2*age+v2*v3 makes cptcovn = 3
11384: */
11385: if (strlen(model)>1)
1.187 brouard 11386: 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 11387: else
1.187 brouard 11388: ncovmodel=2; /* Constant and age */
1.133 brouard 11389: nforce= (nlstate+ndeath-1)*nlstate; /* Number of forces ij from state i to j */
11390: npar= nforce*ncovmodel; /* Number of parameters like aij*/
1.131 brouard 11391: if(npar >MAXPARM || nlstate >NLSTATEMAX || ndeath >NDEATHMAX || ncovmodel>NCOVMAX){
11392: 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);
11393: 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);
11394: fflush(stdout);
11395: fclose (ficlog);
11396: goto end;
11397: }
1.126 brouard 11398: delti3= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11399: delti=delti3[1][1];
11400: /*delti=vector(1,npar); *//* Scale of each paramater (output from hesscov)*/
11401: if(mle==-1){ /* Print a wizard for help writing covariance matrix */
1.247 brouard 11402: /* We could also provide initial parameters values giving by simple logistic regression
11403: * only one way, that is without matrix product. We will have nlstate maximizations */
11404: /* for(i=1;i<nlstate;i++){ */
11405: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
11406: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
11407: /* } */
1.126 brouard 11408: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.191 brouard 11409: printf(" You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
11410: fprintf(ficlog," You chose mle=-1, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11411: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
11412: fclose (ficparo);
11413: fclose (ficlog);
11414: goto end;
11415: exit(0);
1.220 brouard 11416: } else if(mle==-5) { /* Main Wizard */
1.126 brouard 11417: prwizard(ncovmodel, nlstate, ndeath, model, ficparo);
1.192 brouard 11418: printf(" You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
11419: fprintf(ficlog," You chose mle=-3, look at file %s for a template of covariance matrix \n",filereso);
1.126 brouard 11420: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
11421: matcov=matrix(1,npar,1,npar);
1.203 brouard 11422: hess=matrix(1,npar,1,npar);
1.220 brouard 11423: } else{ /* Begin of mle != -1 or -5 */
1.145 brouard 11424: /* Read guessed parameters */
1.126 brouard 11425: /* Reads comments: lines beginning with '#' */
11426: while((c=getc(ficpar))=='#' && c!= EOF){
11427: ungetc(c,ficpar);
11428: fgets(line, MAXLINE, ficpar);
11429: numlinepar++;
1.141 brouard 11430: fputs(line,stdout);
1.126 brouard 11431: fputs(line,ficparo);
11432: fputs(line,ficlog);
11433: }
11434: ungetc(c,ficpar);
11435:
11436: param= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.251 brouard 11437: paramstart= ma3x(1,nlstate,1,nlstate+ndeath-1,1,ncovmodel);
1.126 brouard 11438: for(i=1; i <=nlstate; i++){
1.234 brouard 11439: j=0;
1.126 brouard 11440: for(jj=1; jj <=nlstate+ndeath; jj++){
1.234 brouard 11441: if(jj==i) continue;
11442: j++;
1.292 brouard 11443: while((c=getc(ficpar))=='#' && c!= EOF){
11444: ungetc(c,ficpar);
11445: fgets(line, MAXLINE, ficpar);
11446: numlinepar++;
11447: fputs(line,stdout);
11448: fputs(line,ficparo);
11449: fputs(line,ficlog);
11450: }
11451: ungetc(c,ficpar);
1.234 brouard 11452: fscanf(ficpar,"%1d%1d",&i1,&j1);
11453: if ((i1 != i) || (j1 != jj)){
11454: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n \
1.126 brouard 11455: It might be a problem of design; if ncovcol and the model are correct\n \
11456: run imach with mle=-1 to get a correct template of the parameter file.\n",numlinepar, i,j, i1, j1);
1.234 brouard 11457: exit(1);
11458: }
11459: fprintf(ficparo,"%1d%1d",i1,j1);
11460: if(mle==1)
11461: printf("%1d%1d",i,jj);
11462: fprintf(ficlog,"%1d%1d",i,jj);
11463: for(k=1; k<=ncovmodel;k++){
11464: fscanf(ficpar," %lf",¶m[i][j][k]);
11465: if(mle==1){
11466: printf(" %lf",param[i][j][k]);
11467: fprintf(ficlog," %lf",param[i][j][k]);
11468: }
11469: else
11470: fprintf(ficlog," %lf",param[i][j][k]);
11471: fprintf(ficparo," %lf",param[i][j][k]);
11472: }
11473: fscanf(ficpar,"\n");
11474: numlinepar++;
11475: if(mle==1)
11476: printf("\n");
11477: fprintf(ficlog,"\n");
11478: fprintf(ficparo,"\n");
1.126 brouard 11479: }
11480: }
11481: fflush(ficlog);
1.234 brouard 11482:
1.251 brouard 11483: /* Reads parameters values */
1.126 brouard 11484: p=param[1][1];
1.251 brouard 11485: pstart=paramstart[1][1];
1.126 brouard 11486:
11487: /* Reads comments: lines beginning with '#' */
11488: while((c=getc(ficpar))=='#' && c!= EOF){
11489: ungetc(c,ficpar);
11490: fgets(line, MAXLINE, ficpar);
11491: numlinepar++;
1.141 brouard 11492: fputs(line,stdout);
1.126 brouard 11493: fputs(line,ficparo);
11494: fputs(line,ficlog);
11495: }
11496: ungetc(c,ficpar);
11497:
11498: for(i=1; i <=nlstate; i++){
11499: for(j=1; j <=nlstate+ndeath-1; j++){
1.234 brouard 11500: fscanf(ficpar,"%1d%1d",&i1,&j1);
11501: if ( (i1-i) * (j1-j) != 0){
11502: printf("Error in line parameters number %d, %1d%1d instead of %1d%1d \n",numlinepar, i,j, i1, j1);
11503: exit(1);
11504: }
11505: printf("%1d%1d",i,j);
11506: fprintf(ficparo,"%1d%1d",i1,j1);
11507: fprintf(ficlog,"%1d%1d",i1,j1);
11508: for(k=1; k<=ncovmodel;k++){
11509: fscanf(ficpar,"%le",&delti3[i][j][k]);
11510: printf(" %le",delti3[i][j][k]);
11511: fprintf(ficparo," %le",delti3[i][j][k]);
11512: fprintf(ficlog," %le",delti3[i][j][k]);
11513: }
11514: fscanf(ficpar,"\n");
11515: numlinepar++;
11516: printf("\n");
11517: fprintf(ficparo,"\n");
11518: fprintf(ficlog,"\n");
1.126 brouard 11519: }
11520: }
11521: fflush(ficlog);
1.234 brouard 11522:
1.145 brouard 11523: /* Reads covariance matrix */
1.126 brouard 11524: delti=delti3[1][1];
1.220 brouard 11525:
11526:
1.126 brouard 11527: /* 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 11528:
1.126 brouard 11529: /* Reads comments: lines beginning with '#' */
11530: while((c=getc(ficpar))=='#' && c!= EOF){
11531: ungetc(c,ficpar);
11532: fgets(line, MAXLINE, ficpar);
11533: numlinepar++;
1.141 brouard 11534: fputs(line,stdout);
1.126 brouard 11535: fputs(line,ficparo);
11536: fputs(line,ficlog);
11537: }
11538: ungetc(c,ficpar);
1.220 brouard 11539:
1.126 brouard 11540: matcov=matrix(1,npar,1,npar);
1.203 brouard 11541: hess=matrix(1,npar,1,npar);
1.131 brouard 11542: for(i=1; i <=npar; i++)
11543: for(j=1; j <=npar; j++) matcov[i][j]=0.;
1.220 brouard 11544:
1.194 brouard 11545: /* Scans npar lines */
1.126 brouard 11546: for(i=1; i <=npar; i++){
1.226 brouard 11547: count=fscanf(ficpar,"%1d%1d%d",&i1,&j1,&jk);
1.194 brouard 11548: if(count != 3){
1.226 brouard 11549: printf("Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11550: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11551: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11552: fprintf(ficlog,"Error! Error in parameter file %s at line %d after line starting with %1d%1d%1d\n\
1.194 brouard 11553: This is probably because your covariance matrix doesn't \n contain exactly %d lines corresponding to your model line '1+age+%s'.\n\
11554: Please run with mle=-1 to get a correct covariance matrix.\n",optionfile,numlinepar, i1,j1,jk, npar, model);
1.226 brouard 11555: exit(1);
1.220 brouard 11556: }else{
1.226 brouard 11557: if(mle==1)
11558: printf("%1d%1d%d",i1,j1,jk);
11559: }
11560: fprintf(ficlog,"%1d%1d%d",i1,j1,jk);
11561: fprintf(ficparo,"%1d%1d%d",i1,j1,jk);
1.126 brouard 11562: for(j=1; j <=i; j++){
1.226 brouard 11563: fscanf(ficpar," %le",&matcov[i][j]);
11564: if(mle==1){
11565: printf(" %.5le",matcov[i][j]);
11566: }
11567: fprintf(ficlog," %.5le",matcov[i][j]);
11568: fprintf(ficparo," %.5le",matcov[i][j]);
1.126 brouard 11569: }
11570: fscanf(ficpar,"\n");
11571: numlinepar++;
11572: if(mle==1)
1.220 brouard 11573: printf("\n");
1.126 brouard 11574: fprintf(ficlog,"\n");
11575: fprintf(ficparo,"\n");
11576: }
1.194 brouard 11577: /* End of read covariance matrix npar lines */
1.126 brouard 11578: for(i=1; i <=npar; i++)
11579: for(j=i+1;j<=npar;j++)
1.226 brouard 11580: matcov[i][j]=matcov[j][i];
1.126 brouard 11581:
11582: if(mle==1)
11583: printf("\n");
11584: fprintf(ficlog,"\n");
11585:
11586: fflush(ficlog);
11587:
11588: } /* End of mle != -3 */
1.218 brouard 11589:
1.186 brouard 11590: /* Main data
11591: */
1.290 brouard 11592: nobs=lastobs-firstobs+1; /* was = lastobs;*/
11593: /* num=lvector(1,n); */
11594: /* moisnais=vector(1,n); */
11595: /* annais=vector(1,n); */
11596: /* moisdc=vector(1,n); */
11597: /* andc=vector(1,n); */
11598: /* weight=vector(1,n); */
11599: /* agedc=vector(1,n); */
11600: /* cod=ivector(1,n); */
11601: /* for(i=1;i<=n;i++){ */
11602: num=lvector(firstobs,lastobs);
11603: moisnais=vector(firstobs,lastobs);
11604: annais=vector(firstobs,lastobs);
11605: moisdc=vector(firstobs,lastobs);
11606: andc=vector(firstobs,lastobs);
11607: weight=vector(firstobs,lastobs);
11608: agedc=vector(firstobs,lastobs);
11609: cod=ivector(firstobs,lastobs);
11610: for(i=firstobs;i<=lastobs;i++){
1.234 brouard 11611: num[i]=0;
11612: moisnais[i]=0;
11613: annais[i]=0;
11614: moisdc[i]=0;
11615: andc[i]=0;
11616: agedc[i]=0;
11617: cod[i]=0;
11618: weight[i]=1.0; /* Equal weights, 1 by default */
11619: }
1.290 brouard 11620: mint=matrix(1,maxwav,firstobs,lastobs);
11621: anint=matrix(1,maxwav,firstobs,lastobs);
11622: s=imatrix(1,maxwav+1,firstobs,lastobs); /* s[i][j] health state for wave i and individual j */
1.126 brouard 11623: tab=ivector(1,NCOVMAX);
1.144 brouard 11624: ncodemax=ivector(1,NCOVMAX); /* Number of code per covariate; if O and 1 only, 2**ncov; V1+V2+V3+V4=>16 */
1.192 brouard 11625: 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 11626:
1.136 brouard 11627: /* Reads data from file datafile */
11628: if (readdata(datafile, firstobs, lastobs, &imx)==1)
11629: goto end;
11630:
11631: /* Calculation of the number of parameters from char model */
1.234 brouard 11632: /* modelsav=V2+V1+V4+age*V3 strb=age*V3 stra=V2+V1+V4
1.137 brouard 11633: k=4 (age*V3) Tvar[k=4]= 3 (from V3) Tag[cptcovage=1]=4
11634: k=3 V4 Tvar[k=3]= 4 (from V4)
11635: k=2 V1 Tvar[k=2]= 1 (from V1)
11636: k=1 Tvar[1]=2 (from V2)
1.234 brouard 11637: */
11638:
11639: Tvar=ivector(1,NCOVMAX); /* Was 15 changed to NCOVMAX. */
11640: TvarsDind=ivector(1,NCOVMAX); /* */
11641: TvarsD=ivector(1,NCOVMAX); /* */
11642: TvarsQind=ivector(1,NCOVMAX); /* */
11643: TvarsQ=ivector(1,NCOVMAX); /* */
1.232 brouard 11644: TvarF=ivector(1,NCOVMAX); /* */
11645: TvarFind=ivector(1,NCOVMAX); /* */
11646: TvarV=ivector(1,NCOVMAX); /* */
11647: TvarVind=ivector(1,NCOVMAX); /* */
11648: TvarA=ivector(1,NCOVMAX); /* */
11649: TvarAind=ivector(1,NCOVMAX); /* */
1.231 brouard 11650: TvarFD=ivector(1,NCOVMAX); /* */
11651: TvarFDind=ivector(1,NCOVMAX); /* */
11652: TvarFQ=ivector(1,NCOVMAX); /* */
11653: TvarFQind=ivector(1,NCOVMAX); /* */
11654: TvarVD=ivector(1,NCOVMAX); /* */
11655: TvarVDind=ivector(1,NCOVMAX); /* */
11656: TvarVQ=ivector(1,NCOVMAX); /* */
11657: TvarVQind=ivector(1,NCOVMAX); /* */
11658:
1.230 brouard 11659: Tvalsel=vector(1,NCOVMAX); /* */
1.233 brouard 11660: Tvarsel=ivector(1,NCOVMAX); /* */
1.226 brouard 11661: Typevar=ivector(-1,NCOVMAX); /* -1 to 2 */
11662: Fixed=ivector(-1,NCOVMAX); /* -1 to 3 */
11663: Dummy=ivector(-1,NCOVMAX); /* -1 to 3 */
1.137 brouard 11664: /* V2+V1+V4+age*V3 is a model with 4 covariates (3 plus signs).
11665: For each model-covariate stores the data-covariate id. Tvar[1]=2, Tvar[2]=1, Tvar[3]=4,
11666: Tvar[4=age*V3] is 3 and 'age' is recorded in Tage.
11667: */
11668: /* For model-covariate k tells which data-covariate to use but
11669: because this model-covariate is a construction we invent a new column
11670: ncovcol + k1
11671: If already ncovcol=4 and model=V2+V1+V1*V4+age*V3
11672: Tvar[3=V1*V4]=4+1 etc */
1.227 brouard 11673: Tprod=ivector(1,NCOVMAX); /* Gives the k position of the k1 product */
11674: Tposprod=ivector(1,NCOVMAX); /* Gives the k1 product from the k position */
1.137 brouard 11675: /* Tprod[k1=1]=3(=V1*V4) for V2+V1+V1*V4+age*V3
11676: if V2+V1+V1*V4+age*V3+V3*V2 TProd[k1=2]=5 (V3*V2)
1.227 brouard 11677: Tposprod[k]=k1 , Tposprod[3]=1, Tposprod[5]=2
1.137 brouard 11678: */
1.145 brouard 11679: Tvaraff=ivector(1,NCOVMAX); /* Unclear */
11680: 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 11681: * For V3*V2 (in V2+V1+V1*V4+age*V3+V3*V2), V3*V2 position is 2nd.
11682: * Tvard[k1=2][1]=3 (V3) Tvard[k1=2][2]=2(V2) */
1.145 brouard 11683: Tage=ivector(1,NCOVMAX); /* Gives the covariate id of covariates associated with age: V2 + V1 + age*V4 + V3*age
1.137 brouard 11684: 4 covariates (3 plus signs)
11685: Tage[1=V3*age]= 4; Tage[2=age*V4] = 3
11686: */
1.230 brouard 11687: Tmodelind=ivector(1,NCOVMAX);/** gives the k model position of an
1.227 brouard 11688: * individual dummy, fixed or varying:
11689: * Tmodelind[Tvaraff[3]]=9,Tvaraff[1]@9={4,
11690: * 3, 1, 0, 0, 0, 0, 0, 0},
1.230 brouard 11691: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1 ,
11692: * V1 df, V2 qf, V3 & V4 dv, V5 qv
11693: * Tmodelind[1]@9={9,0,3,2,}*/
11694: TmodelInvind=ivector(1,NCOVMAX); /* TmodelInvind=Tvar[k]- ncovcol-nqv={5-2-1=2,*/
11695: TmodelInvQind=ivector(1,NCOVMAX);/** gives the k model position of an
1.228 brouard 11696: * individual quantitative, fixed or varying:
11697: * Tmodelqind[1]=1,Tvaraff[1]@9={4,
11698: * 3, 1, 0, 0, 0, 0, 0, 0},
11699: * model=V5+V4+V3+V4*V3+V5*age+V2+V1*V2+V1*age+V1*/
1.186 brouard 11700: /* Main decodemodel */
11701:
1.187 brouard 11702:
1.223 brouard 11703: if(decodemodel(model, lastobs) == 1) /* In order to get Tvar[k] V4+V3+V5 p Tvar[1]@3 = {4, 3, 5}*/
1.136 brouard 11704: goto end;
11705:
1.137 brouard 11706: if((double)(lastobs-imx)/(double)imx > 1.10){
11707: nbwarn++;
11708: 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);
11709: 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);
11710: }
1.136 brouard 11711: /* if(mle==1){*/
1.137 brouard 11712: if (weightopt != 1) { /* Maximisation without weights. We can have weights different from 1 but want no weight*/
11713: for(i=1;i<=imx;i++) weight[i]=1.0; /* changed to imx */
1.136 brouard 11714: }
11715:
11716: /*-calculation of age at interview from date of interview and age at death -*/
11717: agev=matrix(1,maxwav,1,imx);
11718:
11719: if(calandcheckages(imx, maxwav, &agemin, &agemax, &nberr, &nbwarn) == 1)
11720: goto end;
11721:
1.126 brouard 11722:
1.136 brouard 11723: agegomp=(int)agemin;
1.290 brouard 11724: free_vector(moisnais,firstobs,lastobs);
11725: free_vector(annais,firstobs,lastobs);
1.126 brouard 11726: /* free_matrix(mint,1,maxwav,1,n);
11727: free_matrix(anint,1,maxwav,1,n);*/
1.215 brouard 11728: /* free_vector(moisdc,1,n); */
11729: /* free_vector(andc,1,n); */
1.145 brouard 11730: /* */
11731:
1.126 brouard 11732: wav=ivector(1,imx);
1.214 brouard 11733: /* dh=imatrix(1,lastpass-firstpass+1,1,imx); */
11734: /* bh=imatrix(1,lastpass-firstpass+1,1,imx); */
11735: /* mw=imatrix(1,lastpass-firstpass+1,1,imx); */
11736: 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.*/
11737: bh=imatrix(1,lastpass-firstpass+2,1,imx);
11738: mw=imatrix(1,lastpass-firstpass+2,1,imx);
1.126 brouard 11739:
11740: /* Concatenates waves */
1.214 brouard 11741: /* Concatenates waves: wav[i] is the number of effective (useful waves) of individual i.
11742: Death is a valid wave (if date is known).
11743: mw[mi][i] is the number of (mi=1 to wav[i]) effective wave out of mi of individual i
11744: dh[m][i] or dh[mw[mi][i]][i] is the delay between two effective waves m=mw[mi][i]
11745: and mw[mi+1][i]. dh depends on stepm.
11746: */
11747:
1.126 brouard 11748: concatwav(wav, dh, bh, mw, s, agedc, agev, firstpass, lastpass, imx, nlstate, stepm);
1.248 brouard 11749: /* Concatenates waves */
1.145 brouard 11750:
1.290 brouard 11751: free_vector(moisdc,firstobs,lastobs);
11752: free_vector(andc,firstobs,lastobs);
1.215 brouard 11753:
1.126 brouard 11754: /* Routine tricode is to calculate cptcoveff (real number of unique covariates) and to associate covariable number and modality */
11755: nbcode=imatrix(0,NCOVMAX,0,NCOVMAX);
11756: ncodemax[1]=1;
1.145 brouard 11757: Ndum =ivector(-1,NCOVMAX);
1.225 brouard 11758: cptcoveff=0;
1.220 brouard 11759: if (ncovmodel-nagesqr > 2 ){ /* That is if covariate other than cst, age and age*age */
11760: tricode(&cptcoveff,Tvar,nbcode,imx, Ndum); /**< Fills nbcode[Tvar[j]][l]; */
1.227 brouard 11761: }
11762:
11763: ncovcombmax=pow(2,cptcoveff);
11764: invalidvarcomb=ivector(1, ncovcombmax);
11765: for(i=1;i<ncovcombmax;i++)
11766: invalidvarcomb[i]=0;
11767:
1.211 brouard 11768: /* Nbcode gives the value of the lth modality (currently 1 to 2) of jth covariate, in
1.186 brouard 11769: V2+V1*age, there are 3 covariates Tvar[2]=1 (V1).*/
1.211 brouard 11770: /* 1 to ncodemax[j] which is the maximum value of this jth covariate */
1.227 brouard 11771:
1.200 brouard 11772: /* codtab=imatrix(1,100,1,10);*/ /* codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) */
1.198 brouard 11773: /*printf(" codtab[1,1],codtab[100,10]=%d,%d\n", codtab[1][1],codtabm(100,10));*/
1.186 brouard 11774: /* codtab gives the value 1 or 2 of the hth combination of k covariates (1 or 2).*/
1.211 brouard 11775: /* nbcode[Tvaraff[j]][codtabm(h,j)]) : if there are only 2 modalities for a covariate j,
11776: * codtabm(h,j) gives its value classified at position h and nbcode gives how it is coded
11777: * (currently 0 or 1) in the data.
11778: * In a loop on h=1 to 2**k, and a loop on j (=1 to k), we get the value of
11779: * corresponding modality (h,j).
11780: */
11781:
1.145 brouard 11782: h=0;
11783: /*if (cptcovn > 0) */
1.126 brouard 11784: m=pow(2,cptcoveff);
11785:
1.144 brouard 11786: /**< codtab(h,k) k = codtab[h,k]=( (h-1) - mod(k-1,2**(k-1) )/2**(k-1) + 1
1.211 brouard 11787: * For k=4 covariates, h goes from 1 to m=2**k
11788: * codtabm(h,k)= (1 & (h-1) >> (k-1)) + 1;
11789: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
1.186 brouard 11790: * h\k 1 2 3 4
1.143 brouard 11791: *______________________________
11792: * 1 i=1 1 i=1 1 i=1 1 i=1 1
11793: * 2 2 1 1 1
11794: * 3 i=2 1 2 1 1
11795: * 4 2 2 1 1
11796: * 5 i=3 1 i=2 1 2 1
11797: * 6 2 1 2 1
11798: * 7 i=4 1 2 2 1
11799: * 8 2 2 2 1
1.197 brouard 11800: * 9 i=5 1 i=3 1 i=2 1 2
11801: * 10 2 1 1 2
11802: * 11 i=6 1 2 1 2
11803: * 12 2 2 1 2
11804: * 13 i=7 1 i=4 1 2 2
11805: * 14 2 1 2 2
11806: * 15 i=8 1 2 2 2
11807: * 16 2 2 2 2
1.143 brouard 11808: */
1.212 brouard 11809: /* How to do the opposite? From combination h (=1 to 2**k) how to get the value on the covariates? */
1.211 brouard 11810: /* from h=5 and m, we get then number of covariates k=log(m)/log(2)=4
11811: * and the value of each covariate?
11812: * V1=1, V2=1, V3=2, V4=1 ?
11813: * h-1=4 and 4 is 0100 or reverse 0010, and +1 is 1121 ok.
11814: * h=6, 6-1=5, 5 is 0101, 1010, 2121, V1=2nd, V2=1st, V3=2nd, V4=1st.
11815: * In order to get the real value in the data, we use nbcode
11816: * nbcode[Tvar[3][2nd]]=1 and nbcode[Tvar[4][1]]=0
11817: * We are keeping this crazy system in order to be able (in the future?)
11818: * to have more than 2 values (0 or 1) for a covariate.
11819: * #define codtabm(h,k) (1 & (h-1) >> (k-1))+1
11820: * h=6, k=2? h-1=5=0101, reverse 1010, +1=2121, k=2nd position: value is 1: codtabm(6,2)=1
11821: * bbbbbbbb
11822: * 76543210
11823: * h-1 00000101 (6-1=5)
1.219 brouard 11824: *(h-1)>>(k-1)= 00000010 >> (2-1) = 1 right shift
1.211 brouard 11825: * &
11826: * 1 00000001 (1)
1.219 brouard 11827: * 00000000 = 1 & ((h-1) >> (k-1))
11828: * +1= 00000001 =1
1.211 brouard 11829: *
11830: * h=14, k=3 => h'=h-1=13, k'=k-1=2
11831: * h' 1101 =2^3+2^2+0x2^1+2^0
11832: * >>k' 11
11833: * & 00000001
11834: * = 00000001
11835: * +1 = 00000010=2 = codtabm(14,3)
11836: * Reverse h=6 and m=16?
11837: * cptcoveff=log(16)/log(2)=4 covariate: 6-1=5=0101 reversed=1010 +1=2121 =>V1=2, V2=1, V3=2, V4=1.
11838: * for (j=1 to cptcoveff) Vj=decodtabm(j,h,cptcoveff)
11839: * decodtabm(h,j,cptcoveff)= (((h-1) >> (j-1)) & 1) +1
11840: * decodtabm(h,j,cptcoveff)= (h <= (1<<cptcoveff)?(((h-1) >> (j-1)) & 1) +1 : -1)
11841: * V3=decodtabm(14,3,2**4)=2
11842: * h'=13 1101 =2^3+2^2+0x2^1+2^0
11843: *(h-1) >> (j-1) 0011 =13 >> 2
11844: * &1 000000001
11845: * = 000000001
11846: * +1= 000000010 =2
11847: * 2211
11848: * V1=1+1, V2=0+1, V3=1+1, V4=1+1
11849: * V3=2
1.220 brouard 11850: * codtabm and decodtabm are identical
1.211 brouard 11851: */
11852:
1.145 brouard 11853:
11854: free_ivector(Ndum,-1,NCOVMAX);
11855:
11856:
1.126 brouard 11857:
1.186 brouard 11858: /* Initialisation of ----------- gnuplot -------------*/
1.126 brouard 11859: strcpy(optionfilegnuplot,optionfilefiname);
11860: if(mle==-3)
1.201 brouard 11861: strcat(optionfilegnuplot,"-MORT_");
1.126 brouard 11862: strcat(optionfilegnuplot,".gp");
11863:
11864: if((ficgp=fopen(optionfilegnuplot,"w"))==NULL) {
11865: printf("Problem with file %s",optionfilegnuplot);
11866: }
11867: else{
1.204 brouard 11868: fprintf(ficgp,"\n# IMaCh-%s\n", version);
1.126 brouard 11869: fprintf(ficgp,"# %s\n", optionfilegnuplot);
1.141 brouard 11870: //fprintf(ficgp,"set missing 'NaNq'\n");
11871: fprintf(ficgp,"set datafile missing 'NaNq'\n");
1.126 brouard 11872: }
11873: /* fclose(ficgp);*/
1.186 brouard 11874:
11875:
11876: /* Initialisation of --------- index.htm --------*/
1.126 brouard 11877:
11878: strcpy(optionfilehtm,optionfilefiname); /* Main html file */
11879: if(mle==-3)
1.201 brouard 11880: strcat(optionfilehtm,"-MORT_");
1.126 brouard 11881: strcat(optionfilehtm,".htm");
11882: if((fichtm=fopen(optionfilehtm,"w"))==NULL) {
1.131 brouard 11883: printf("Problem with %s \n",optionfilehtm);
11884: exit(0);
1.126 brouard 11885: }
11886:
11887: strcpy(optionfilehtmcov,optionfilefiname); /* Only for matrix of covariance */
11888: strcat(optionfilehtmcov,"-cov.htm");
11889: if((fichtmcov=fopen(optionfilehtmcov,"w"))==NULL) {
11890: printf("Problem with %s \n",optionfilehtmcov), exit(0);
11891: }
11892: else{
11893: fprintf(fichtmcov,"<html><head>\n<title>IMaCh Cov %s</title></head>\n <body><font size=\"2\">%s <br> %s</font> \
11894: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11895: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n",\
1.126 brouard 11896: optionfilehtmcov,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model);
11897: }
11898:
1.213 brouard 11899: fprintf(fichtm,"<html><head>\n<head>\n<meta charset=\"utf-8\"/><meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\" />\n<title>IMaCh %s</title></head>\n <body><font size=\"7\"><a href=http:/euroreves.ined.fr/imach>IMaCh for Interpolated Markov Chain</a> </font><br>\n<font size=\"3\">Sponsored by Copyright (C) 2002-2015 <a href=http://www.ined.fr>INED</a>-EUROREVES-Institut de longévité-2013-2016-Japan Society for the Promotion of Sciences 日本å¦è¡“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 11900: <hr size=\"2\" color=\"#EC5E5E\"> \n\
11901: <font size=\"2\">IMaCh-%s <br> %s</font> \
1.126 brouard 11902: <hr size=\"2\" color=\"#EC5E5E\"> \n\
1.204 brouard 11903: Title=%s <br>Datafile=%s Firstpass=%d Lastpass=%d Stepm=%d Weight=%d Model=1+age+%s<br>\n\
1.126 brouard 11904: \n\
11905: <hr size=\"2\" color=\"#EC5E5E\">\
11906: <ul><li><h4>Parameter files</h4>\n\
11907: - Parameter file: <a href=\"%s.%s\">%s.%s</a><br>\n\
11908: - Copy of the parameter file: <a href=\"o%s\">o%s</a><br>\n\
11909: - Log file of the run: <a href=\"%s\">%s</a><br>\n\
11910: - Gnuplot file name: <a href=\"%s\">%s</a><br>\n\
11911: - Date and time at start: %s</ul>\n",\
11912: optionfilehtm,version,fullversion,title,datafile,firstpass,lastpass,stepm, weightopt, model,\
11913: optionfilefiname,optionfilext,optionfilefiname,optionfilext,\
11914: fileres,fileres,\
11915: filelog,filelog,optionfilegnuplot,optionfilegnuplot,strstart);
11916: fflush(fichtm);
11917:
11918: strcpy(pathr,path);
11919: strcat(pathr,optionfilefiname);
1.184 brouard 11920: #ifdef WIN32
11921: _chdir(optionfilefiname); /* Move to directory named optionfile */
11922: #else
1.126 brouard 11923: chdir(optionfilefiname); /* Move to directory named optionfile */
1.184 brouard 11924: #endif
11925:
1.126 brouard 11926:
1.220 brouard 11927: /* Calculates basic frequencies. Computes observed prevalence at single age
11928: and for any valid combination of covariates
1.126 brouard 11929: and prints on file fileres'p'. */
1.251 brouard 11930: freqsummary(fileres, p, pstart, agemin, agemax, s, agev, nlstate, imx, Tvaraff, invalidvarcomb, nbcode, ncodemax,mint,anint,strstart, \
1.227 brouard 11931: firstpass, lastpass, stepm, weightopt, model);
1.126 brouard 11932:
11933: fprintf(fichtm,"\n");
1.286 brouard 11934: 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 11935: ftol, stepm);
11936: fprintf(fichtm,"\n<li>Number of fixed dummy covariates: ncovcol=%d ", ncovcol);
11937: ncurrv=1;
11938: for(i=ncurrv; i <=ncovcol; i++) fprintf(fichtm,"V%d ", i);
11939: fprintf(fichtm,"\n<li> Number of fixed quantitative variables: nqv=%d ", nqv);
11940: ncurrv=i;
11941: for(i=ncurrv; i <=ncurrv-1+nqv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11942: fprintf(fichtm,"\n<li> Number of time varying (wave varying) dummy covariates: ntv=%d ", ntv);
1.274 brouard 11943: ncurrv=i;
11944: for(i=ncurrv; i <=ncurrv-1+ntv; i++) fprintf(fichtm,"V%d ", i);
1.290 brouard 11945: fprintf(fichtm,"\n<li>Number of time varying quantitative covariates: nqtv=%d ", nqtv);
1.274 brouard 11946: ncurrv=i;
11947: for(i=ncurrv; i <=ncurrv-1+nqtv; i++) fprintf(fichtm,"V%d ", i);
11948: 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", \
11949: nlstate, ndeath, maxwav, mle, weightopt);
11950:
11951: fprintf(fichtm,"<h4> Diagram of states <a href=\"%s_.svg\">%s_.svg</a></h4> \n\
11952: <img src=\"%s_.svg\">", subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"),subdirf2(optionfilefiname,"D_"));
11953:
11954:
11955: fprintf(fichtm,"\n<h4>Some descriptive statistics </h4>\n<br>Total number of observations=%d <br>\n\
1.126 brouard 11956: Youngest age at first (selected) pass %.2f, oldest age %.2f<br>\n\
11957: Interval (in months) between two waves: Min=%d Max=%d Mean=%.2lf<br>\n",\
1.274 brouard 11958: imx,agemin,agemax,jmin,jmax,jmean);
1.126 brouard 11959: pmmij= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
1.268 brouard 11960: oldms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11961: newms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11962: savms= matrix(1,nlstate+ndeath,1,nlstate+ndeath); /* creation */
11963: oldm=oldms; newm=newms; savm=savms; /* Keeps fixed addresses to free */
1.218 brouard 11964:
1.126 brouard 11965: /* For Powell, parameters are in a vector p[] starting at p[1]
11966: so we point p on param[1][1] so that p[1] maps on param[1][1][1] */
11967: p=param[1][1]; /* *(*(*(param +1)+1)+0) */
11968:
11969: globpr=0; /* To get the number ipmx of contributions and the sum of weights*/
1.186 brouard 11970: /* For mortality only */
1.126 brouard 11971: if (mle==-3){
1.136 brouard 11972: ximort=matrix(1,NDIM,1,NDIM);
1.248 brouard 11973: for(i=1;i<=NDIM;i++)
11974: for(j=1;j<=NDIM;j++)
11975: ximort[i][j]=0.;
1.186 brouard 11976: /* ximort=gsl_matrix_alloc(1,NDIM,1,NDIM); */
1.290 brouard 11977: cens=ivector(firstobs,lastobs);
11978: ageexmed=vector(firstobs,lastobs);
11979: agecens=vector(firstobs,lastobs);
11980: dcwave=ivector(firstobs,lastobs);
1.223 brouard 11981:
1.126 brouard 11982: for (i=1; i<=imx; i++){
11983: dcwave[i]=-1;
11984: for (m=firstpass; m<=lastpass; m++)
1.226 brouard 11985: if (s[m][i]>nlstate) {
11986: dcwave[i]=m;
11987: /* printf("i=%d j=%d s=%d dcwave=%d\n",i,j, s[j][i],dcwave[i]);*/
11988: break;
11989: }
1.126 brouard 11990: }
1.226 brouard 11991:
1.126 brouard 11992: for (i=1; i<=imx; i++) {
11993: if (wav[i]>0){
1.226 brouard 11994: ageexmed[i]=agev[mw[1][i]][i];
11995: j=wav[i];
11996: agecens[i]=1.;
11997:
11998: if (ageexmed[i]> 1 && wav[i] > 0){
11999: agecens[i]=agev[mw[j][i]][i];
12000: cens[i]= 1;
12001: }else if (ageexmed[i]< 1)
12002: cens[i]= -1;
12003: if (agedc[i]< AGESUP && agedc[i]>1 && dcwave[i]>firstpass && dcwave[i]<=lastpass)
12004: cens[i]=0 ;
1.126 brouard 12005: }
12006: else cens[i]=-1;
12007: }
12008:
12009: for (i=1;i<=NDIM;i++) {
12010: for (j=1;j<=NDIM;j++)
1.226 brouard 12011: ximort[i][j]=(i == j ? 1.0 : 0.0);
1.126 brouard 12012: }
12013:
1.302 brouard 12014: p[1]=0.0268; p[NDIM]=0.083;
12015: /* printf("%lf %lf", p[1], p[2]); */
1.126 brouard 12016:
12017:
1.136 brouard 12018: #ifdef GSL
12019: printf("GSL optimization\n"); fprintf(ficlog,"Powell\n");
1.162 brouard 12020: #else
1.126 brouard 12021: printf("Powell\n"); fprintf(ficlog,"Powell\n");
1.136 brouard 12022: #endif
1.201 brouard 12023: strcpy(filerespow,"POW-MORT_");
12024: strcat(filerespow,fileresu);
1.126 brouard 12025: if((ficrespow=fopen(filerespow,"w"))==NULL) {
12026: printf("Problem with resultfile: %s\n", filerespow);
12027: fprintf(ficlog,"Problem with resultfile: %s\n", filerespow);
12028: }
1.136 brouard 12029: #ifdef GSL
12030: fprintf(ficrespow,"# GSL optimization\n# iter -2*LL");
1.162 brouard 12031: #else
1.126 brouard 12032: fprintf(ficrespow,"# Powell\n# iter -2*LL");
1.136 brouard 12033: #endif
1.126 brouard 12034: /* for (i=1;i<=nlstate;i++)
12035: for(j=1;j<=nlstate+ndeath;j++)
12036: if(j!=i)fprintf(ficrespow," p%1d%1d",i,j);
12037: */
12038: fprintf(ficrespow,"\n");
1.136 brouard 12039: #ifdef GSL
12040: /* gsl starts here */
12041: T = gsl_multimin_fminimizer_nmsimplex;
12042: gsl_multimin_fminimizer *sfm = NULL;
12043: gsl_vector *ss, *x;
12044: gsl_multimin_function minex_func;
12045:
12046: /* Initial vertex size vector */
12047: ss = gsl_vector_alloc (NDIM);
12048:
12049: if (ss == NULL){
12050: GSL_ERROR_VAL ("failed to allocate space for ss", GSL_ENOMEM, 0);
12051: }
12052: /* Set all step sizes to 1 */
12053: gsl_vector_set_all (ss, 0.001);
12054:
12055: /* Starting point */
1.126 brouard 12056:
1.136 brouard 12057: x = gsl_vector_alloc (NDIM);
12058:
12059: if (x == NULL){
12060: gsl_vector_free(ss);
12061: GSL_ERROR_VAL ("failed to allocate space for x", GSL_ENOMEM, 0);
12062: }
12063:
12064: /* Initialize method and iterate */
12065: /* p[1]=0.0268; p[NDIM]=0.083; */
1.186 brouard 12066: /* gsl_vector_set(x, 0, 0.0268); */
12067: /* gsl_vector_set(x, 1, 0.083); */
1.136 brouard 12068: gsl_vector_set(x, 0, p[1]);
12069: gsl_vector_set(x, 1, p[2]);
12070:
12071: minex_func.f = &gompertz_f;
12072: minex_func.n = NDIM;
12073: minex_func.params = (void *)&p; /* ??? */
12074:
12075: sfm = gsl_multimin_fminimizer_alloc (T, NDIM);
12076: gsl_multimin_fminimizer_set (sfm, &minex_func, x, ss);
12077:
12078: printf("Iterations beginning .....\n\n");
12079: printf("Iter. # Intercept Slope -Log Likelihood Simplex size\n");
12080:
12081: iteri=0;
12082: while (rval == GSL_CONTINUE){
12083: iteri++;
12084: status = gsl_multimin_fminimizer_iterate(sfm);
12085:
12086: if (status) printf("error: %s\n", gsl_strerror (status));
12087: fflush(0);
12088:
12089: if (status)
12090: break;
12091:
12092: rval = gsl_multimin_test_size (gsl_multimin_fminimizer_size (sfm), 1e-6);
12093: ssval = gsl_multimin_fminimizer_size (sfm);
12094:
12095: if (rval == GSL_SUCCESS)
12096: printf ("converged to a local maximum at\n");
12097:
12098: printf("%5d ", iteri);
12099: for (it = 0; it < NDIM; it++){
12100: printf ("%10.5f ", gsl_vector_get (sfm->x, it));
12101: }
12102: printf("f() = %-10.5f ssize = %.7f\n", sfm->fval, ssval);
12103: }
12104:
12105: printf("\n\n Please note: Program should be run many times with varying starting points to detemine global maximum\n\n");
12106:
12107: gsl_vector_free(x); /* initial values */
12108: gsl_vector_free(ss); /* inital step size */
12109: for (it=0; it<NDIM; it++){
12110: p[it+1]=gsl_vector_get(sfm->x,it);
12111: fprintf(ficrespow," %.12lf", p[it]);
12112: }
12113: gsl_multimin_fminimizer_free (sfm); /* p *(sfm.x.data) et p *(sfm.x.data+1) */
12114: #endif
12115: #ifdef POWELL
12116: powell(p,ximort,NDIM,ftol,&iter,&fret,gompertz);
12117: #endif
1.126 brouard 12118: fclose(ficrespow);
12119:
1.203 brouard 12120: hesscov(matcov, hess, p, NDIM, delti, 1e-4, gompertz);
1.126 brouard 12121:
12122: for(i=1; i <=NDIM; i++)
12123: for(j=i+1;j<=NDIM;j++)
1.220 brouard 12124: matcov[i][j]=matcov[j][i];
1.126 brouard 12125:
12126: printf("\nCovariance matrix\n ");
1.203 brouard 12127: fprintf(ficlog,"\nCovariance matrix\n ");
1.126 brouard 12128: for(i=1; i <=NDIM; i++) {
12129: for(j=1;j<=NDIM;j++){
1.220 brouard 12130: printf("%f ",matcov[i][j]);
12131: fprintf(ficlog,"%f ",matcov[i][j]);
1.126 brouard 12132: }
1.203 brouard 12133: printf("\n "); fprintf(ficlog,"\n ");
1.126 brouard 12134: }
12135:
12136: printf("iter=%d MLE=%f Eq=%lf*exp(%lf*(age-%d))\n",iter,-gompertz(p),p[1],p[2],agegomp);
1.193 brouard 12137: for (i=1;i<=NDIM;i++) {
1.126 brouard 12138: printf("%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
1.193 brouard 12139: fprintf(ficlog,"%f [%f ; %f]\n",p[i],p[i]-2*sqrt(matcov[i][i]),p[i]+2*sqrt(matcov[i][i]));
12140: }
1.302 brouard 12141: lsurv=vector(agegomp,AGESUP);
12142: lpop=vector(agegomp,AGESUP);
12143: tpop=vector(agegomp,AGESUP);
1.126 brouard 12144: lsurv[agegomp]=100000;
12145:
12146: for (k=agegomp;k<=AGESUP;k++) {
12147: agemortsup=k;
12148: if (p[1]*exp(p[2]*(k-agegomp))>1) break;
12149: }
12150:
12151: for (k=agegomp;k<agemortsup;k++)
12152: lsurv[k+1]=lsurv[k]-lsurv[k]*(p[1]*exp(p[2]*(k-agegomp)));
12153:
12154: for (k=agegomp;k<agemortsup;k++){
12155: lpop[k]=(lsurv[k]+lsurv[k+1])/2.;
12156: sumlpop=sumlpop+lpop[k];
12157: }
12158:
12159: tpop[agegomp]=sumlpop;
12160: for (k=agegomp;k<(agemortsup-3);k++){
12161: /* tpop[k+1]=2;*/
12162: tpop[k+1]=tpop[k]-lpop[k];
12163: }
12164:
12165:
12166: printf("\nAge lx qx dx Lx Tx e(x)\n");
12167: for (k=agegomp;k<(agemortsup-2);k++)
12168: 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]);
12169:
12170:
12171: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.220 brouard 12172: ageminpar=50;
12173: agemaxpar=100;
1.194 brouard 12174: if(ageminpar == AGEOVERFLOW ||agemaxpar == AGEOVERFLOW){
12175: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12176: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12177: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
12178: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
12179: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12180: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12181: }else{
12182: printf("Warning! ageminpar %f and agemaxpar %f have been fixed because for simplification until it is fixed...\n\n",ageminpar,agemaxpar);
12183: 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 12184: printinggnuplotmort(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, pathc,p);
1.220 brouard 12185: }
1.201 brouard 12186: printinghtmlmort(fileresu,title,datafile, firstpass, lastpass, \
1.126 brouard 12187: stepm, weightopt,\
12188: model,imx,p,matcov,agemortsup);
12189:
1.302 brouard 12190: free_vector(lsurv,agegomp,AGESUP);
12191: free_vector(lpop,agegomp,AGESUP);
12192: free_vector(tpop,agegomp,AGESUP);
1.220 brouard 12193: free_matrix(ximort,1,NDIM,1,NDIM);
1.290 brouard 12194: free_ivector(dcwave,firstobs,lastobs);
12195: free_vector(agecens,firstobs,lastobs);
12196: free_vector(ageexmed,firstobs,lastobs);
12197: free_ivector(cens,firstobs,lastobs);
1.220 brouard 12198: #ifdef GSL
1.136 brouard 12199: #endif
1.186 brouard 12200: } /* Endof if mle==-3 mortality only */
1.205 brouard 12201: /* Standard */
12202: else{ /* For mle !=- 3, could be 0 or 1 or 4 etc. */
12203: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12204: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
1.132 brouard 12205: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
1.126 brouard 12206: printf("First Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12207: for (k=1; k<=npar;k++)
12208: printf(" %d %8.5f",k,p[k]);
12209: printf("\n");
1.205 brouard 12210: if(mle>=1){ /* Could be 1 or 2, Real Maximization */
12211: /* mlikeli uses func not funcone */
1.247 brouard 12212: /* for(i=1;i<nlstate;i++){ */
12213: /* /\*reducing xi for 1 to npar to 1 to ncovmodel; *\/ */
12214: /* mlikeli(ficres,p, ncovmodel, ncovmodel, nlstate, ftol, funcnoprod); */
12215: /* } */
1.205 brouard 12216: mlikeli(ficres,p, npar, ncovmodel, nlstate, ftol, func);
12217: }
12218: if(mle==0) {/* No optimization, will print the likelihoods for the datafile */
12219: globpr=0;/* Computes sum of likelihood for globpr=1 and funcone */
12220: /* Computes likelihood for initial parameters, uses funcone to compute gpimx and gsw */
12221: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12222: }
12223: globpr=1; /* again, to print the individual contributions using computed gpimx and gsw */
1.126 brouard 12224: likelione(ficres, p, npar, nlstate, &globpr, &ipmx, &sw, &fretone, funcone); /* Prints the contributions to the likelihood */
12225: printf("Second Likeli=%12.6f ipmx=%ld sw=%12.6f",fretone,ipmx,sw);
12226: for (k=1; k<=npar;k++)
12227: printf(" %d %8.5f",k,p[k]);
12228: printf("\n");
12229:
12230: /*--------- results files --------------*/
1.283 brouard 12231: /* 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 12232:
12233:
12234: fprintf(ficres,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12235: printf("# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12236: fprintf(ficlog,"# Parameters nlstate*nlstate*ncov a12*1 + b12 * age + ...\n");
12237: for(i=1,jk=1; i <=nlstate; i++){
12238: for(k=1; k <=(nlstate+ndeath); k++){
1.225 brouard 12239: if (k != i) {
12240: printf("%d%d ",i,k);
12241: fprintf(ficlog,"%d%d ",i,k);
12242: fprintf(ficres,"%1d%1d ",i,k);
12243: for(j=1; j <=ncovmodel; j++){
12244: printf("%12.7f ",p[jk]);
12245: fprintf(ficlog,"%12.7f ",p[jk]);
12246: fprintf(ficres,"%12.7f ",p[jk]);
12247: jk++;
12248: }
12249: printf("\n");
12250: fprintf(ficlog,"\n");
12251: fprintf(ficres,"\n");
12252: }
1.126 brouard 12253: }
12254: }
1.203 brouard 12255: if(mle != 0){
12256: /* Computing hessian and covariance matrix only at a peak of the Likelihood, that is after optimization */
1.126 brouard 12257: ftolhess=ftol; /* Usually correct */
1.203 brouard 12258: hesscov(matcov, hess, p, npar, delti, ftolhess, func);
12259: 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");
12260: 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");
12261: for(i=1,jk=1; i <=nlstate; i++){
1.225 brouard 12262: for(k=1; k <=(nlstate+ndeath); k++){
12263: if (k != i) {
12264: printf("%d%d ",i,k);
12265: fprintf(ficlog,"%d%d ",i,k);
12266: for(j=1; j <=ncovmodel; j++){
12267: printf("%12.7f W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk], p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
12268: fprintf(ficlog,"%12.7f W=%8.3f CI=[%12.7f ; %12.7f] ",p[jk], p[jk]/sqrt(matcov[jk][jk]), p[jk]-1.96*sqrt(matcov[jk][jk]),p[jk]+1.96*sqrt(matcov[jk][jk]));
12269: jk++;
12270: }
12271: printf("\n");
12272: fprintf(ficlog,"\n");
12273: }
12274: }
1.193 brouard 12275: }
1.203 brouard 12276: } /* end of hesscov and Wald tests */
1.225 brouard 12277:
1.203 brouard 12278: /* */
1.126 brouard 12279: fprintf(ficres,"# Scales (for hessian or gradient estimation)\n");
12280: printf("# Scales (for hessian or gradient estimation)\n");
12281: fprintf(ficlog,"# Scales (for hessian or gradient estimation)\n");
12282: for(i=1,jk=1; i <=nlstate; i++){
12283: for(j=1; j <=nlstate+ndeath; j++){
1.225 brouard 12284: if (j!=i) {
12285: fprintf(ficres,"%1d%1d",i,j);
12286: printf("%1d%1d",i,j);
12287: fprintf(ficlog,"%1d%1d",i,j);
12288: for(k=1; k<=ncovmodel;k++){
12289: printf(" %.5e",delti[jk]);
12290: fprintf(ficlog," %.5e",delti[jk]);
12291: fprintf(ficres," %.5e",delti[jk]);
12292: jk++;
12293: }
12294: printf("\n");
12295: fprintf(ficlog,"\n");
12296: fprintf(ficres,"\n");
12297: }
1.126 brouard 12298: }
12299: }
12300:
12301: 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 12302: if(mle >= 1) /* To big for the screen */
1.126 brouard 12303: 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");
12304: 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");
12305: /* # 121 Var(a12)\n\ */
12306: /* # 122 Cov(b12,a12) Var(b12)\n\ */
12307: /* # 131 Cov(a13,a12) Cov(a13,b12, Var(a13)\n\ */
12308: /* # 132 Cov(b13,a12) Cov(b13,b12, Cov(b13,a13) Var(b13)\n\ */
12309: /* # 212 Cov(a21,a12) Cov(a21,b12, Cov(a21,a13) Cov(a21,b13) Var(a21)\n\ */
12310: /* # 212 Cov(b21,a12) Cov(b21,b12, Cov(b21,a13) Cov(b21,b13) Cov(b21,a21) Var(b21)\n\ */
12311: /* # 232 Cov(a23,a12) Cov(a23,b12, Cov(a23,a13) Cov(a23,b13) Cov(a23,a21) Cov(a23,b21) Var(a23)\n\ */
12312: /* # 232 Cov(b23,a12) Cov(b23,b12) ... Var (b23)\n" */
12313:
12314:
12315: /* Just to have a covariance matrix which will be more understandable
12316: even is we still don't want to manage dictionary of variables
12317: */
12318: for(itimes=1;itimes<=2;itimes++){
12319: jj=0;
12320: for(i=1; i <=nlstate; i++){
1.225 brouard 12321: for(j=1; j <=nlstate+ndeath; j++){
12322: if(j==i) continue;
12323: for(k=1; k<=ncovmodel;k++){
12324: jj++;
12325: ca[0]= k+'a'-1;ca[1]='\0';
12326: if(itimes==1){
12327: if(mle>=1)
12328: printf("#%1d%1d%d",i,j,k);
12329: fprintf(ficlog,"#%1d%1d%d",i,j,k);
12330: fprintf(ficres,"#%1d%1d%d",i,j,k);
12331: }else{
12332: if(mle>=1)
12333: printf("%1d%1d%d",i,j,k);
12334: fprintf(ficlog,"%1d%1d%d",i,j,k);
12335: fprintf(ficres,"%1d%1d%d",i,j,k);
12336: }
12337: ll=0;
12338: for(li=1;li <=nlstate; li++){
12339: for(lj=1;lj <=nlstate+ndeath; lj++){
12340: if(lj==li) continue;
12341: for(lk=1;lk<=ncovmodel;lk++){
12342: ll++;
12343: if(ll<=jj){
12344: cb[0]= lk +'a'-1;cb[1]='\0';
12345: if(ll<jj){
12346: if(itimes==1){
12347: if(mle>=1)
12348: printf(" Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12349: fprintf(ficlog," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12350: fprintf(ficres," Cov(%s%1d%1d,%s%1d%1d)",ca,i,j,cb, li,lj);
12351: }else{
12352: if(mle>=1)
12353: printf(" %.5e",matcov[jj][ll]);
12354: fprintf(ficlog," %.5e",matcov[jj][ll]);
12355: fprintf(ficres," %.5e",matcov[jj][ll]);
12356: }
12357: }else{
12358: if(itimes==1){
12359: if(mle>=1)
12360: printf(" Var(%s%1d%1d)",ca,i,j);
12361: fprintf(ficlog," Var(%s%1d%1d)",ca,i,j);
12362: fprintf(ficres," Var(%s%1d%1d)",ca,i,j);
12363: }else{
12364: if(mle>=1)
12365: printf(" %.7e",matcov[jj][ll]);
12366: fprintf(ficlog," %.7e",matcov[jj][ll]);
12367: fprintf(ficres," %.7e",matcov[jj][ll]);
12368: }
12369: }
12370: }
12371: } /* end lk */
12372: } /* end lj */
12373: } /* end li */
12374: if(mle>=1)
12375: printf("\n");
12376: fprintf(ficlog,"\n");
12377: fprintf(ficres,"\n");
12378: numlinepar++;
12379: } /* end k*/
12380: } /*end j */
1.126 brouard 12381: } /* end i */
12382: } /* end itimes */
12383:
12384: fflush(ficlog);
12385: fflush(ficres);
1.225 brouard 12386: while(fgets(line, MAXLINE, ficpar)) {
12387: /* If line starts with a # it is a comment */
12388: if (line[0] == '#') {
12389: numlinepar++;
12390: fputs(line,stdout);
12391: fputs(line,ficparo);
12392: fputs(line,ficlog);
1.299 brouard 12393: fputs(line,ficres);
1.225 brouard 12394: continue;
12395: }else
12396: break;
12397: }
12398:
1.209 brouard 12399: /* while((c=getc(ficpar))=='#' && c!= EOF){ */
12400: /* ungetc(c,ficpar); */
12401: /* fgets(line, MAXLINE, ficpar); */
12402: /* fputs(line,stdout); */
12403: /* fputs(line,ficparo); */
12404: /* } */
12405: /* ungetc(c,ficpar); */
1.126 brouard 12406:
12407: estepm=0;
1.209 brouard 12408: 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 12409:
12410: if (num_filled != 6) {
12411: 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);
12412: 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);
12413: goto end;
12414: }
12415: printf("agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%lf\n",ageminpar,agemaxpar, bage, fage, estepm, ftolpl);
12416: }
12417: /* ftolpl=6*ftol*1.e5; /\* 6.e-3 make convergences in less than 80 loops for the prevalence limit *\/ */
12418: /*ftolpl=6.e-4;*/ /* 6.e-3 make convergences in less than 80 loops for the prevalence limit */
12419:
1.209 brouard 12420: /* fscanf(ficpar,"agemin=%lf agemax=%lf bage=%lf fage=%lf estepm=%d ftolpl=%\n",&ageminpar,&agemaxpar, &bage, &fage, &estepm); */
1.126 brouard 12421: if (estepm==0 || estepm < stepm) estepm=stepm;
12422: if (fage <= 2) {
12423: bage = ageminpar;
12424: fage = agemaxpar;
12425: }
12426:
12427: fprintf(ficres,"# agemin agemax for life expectancy, bage fage (if mle==0 ie no data nor Max likelihood).\n");
1.211 brouard 12428: fprintf(ficres,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
12429: fprintf(ficparo,"agemin=%.0f agemax=%.0f bage=%.0f fage=%.0f estepm=%d, ftolpl=%e\n",ageminpar,agemaxpar,bage,fage, estepm, ftolpl);
1.220 brouard 12430:
1.186 brouard 12431: /* Other stuffs, more or less useful */
1.254 brouard 12432: while(fgets(line, MAXLINE, ficpar)) {
12433: /* If line starts with a # it is a comment */
12434: if (line[0] == '#') {
12435: numlinepar++;
12436: fputs(line,stdout);
12437: fputs(line,ficparo);
12438: fputs(line,ficlog);
1.299 brouard 12439: fputs(line,ficres);
1.254 brouard 12440: continue;
12441: }else
12442: break;
12443: }
12444:
12445: 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){
12446:
12447: if (num_filled != 7) {
12448: 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);
12449: 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);
12450: goto end;
12451: }
12452: printf("begin-prev-date=%.lf/%.lf/%.lf end-prev-date=%.lf/%.lf/%.lf mov_average=%d\n",jprev1, mprev1,anprev1,jprev2, mprev2,anprev2,mobilav);
12453: 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);
12454: 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);
12455: 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 12456: }
1.254 brouard 12457:
12458: while(fgets(line, MAXLINE, ficpar)) {
12459: /* If line starts with a # it is a comment */
12460: if (line[0] == '#') {
12461: numlinepar++;
12462: fputs(line,stdout);
12463: fputs(line,ficparo);
12464: fputs(line,ficlog);
1.299 brouard 12465: fputs(line,ficres);
1.254 brouard 12466: continue;
12467: }else
12468: break;
1.126 brouard 12469: }
12470:
12471:
12472: dateprev1=anprev1+(mprev1-1)/12.+(jprev1-1)/365.;
12473: dateprev2=anprev2+(mprev2-1)/12.+(jprev2-1)/365.;
12474:
1.254 brouard 12475: if((num_filled=sscanf(line,"pop_based=%d\n",&popbased)) !=EOF){
12476: if (num_filled != 1) {
12477: 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);
12478: 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);
12479: goto end;
12480: }
12481: printf("pop_based=%d\n",popbased);
12482: fprintf(ficlog,"pop_based=%d\n",popbased);
12483: fprintf(ficparo,"pop_based=%d\n",popbased);
12484: fprintf(ficres,"pop_based=%d\n",popbased);
12485: }
12486:
1.258 brouard 12487: /* Results */
1.307 brouard 12488: endishere=0;
1.258 brouard 12489: nresult=0;
1.308 brouard 12490: parameterline=0;
1.258 brouard 12491: do{
12492: if(!fgets(line, MAXLINE, ficpar)){
12493: endishere=1;
1.308 brouard 12494: parameterline=15;
1.258 brouard 12495: }else if (line[0] == '#') {
12496: /* If line starts with a # it is a comment */
1.254 brouard 12497: numlinepar++;
12498: fputs(line,stdout);
12499: fputs(line,ficparo);
12500: fputs(line,ficlog);
1.299 brouard 12501: fputs(line,ficres);
1.254 brouard 12502: continue;
1.258 brouard 12503: }else if(sscanf(line,"prevforecast=%[^\n]\n",modeltemp))
12504: parameterline=11;
1.296 brouard 12505: else if(sscanf(line,"prevbackcast=%[^\n]\n",modeltemp))
1.258 brouard 12506: parameterline=12;
1.307 brouard 12507: else if(sscanf(line,"result:%[^\n]\n",modeltemp)){
1.258 brouard 12508: parameterline=13;
1.307 brouard 12509: }
1.258 brouard 12510: else{
12511: parameterline=14;
1.254 brouard 12512: }
1.308 brouard 12513: switch (parameterline){ /* =0 only if only comments */
1.258 brouard 12514: case 11:
1.296 brouard 12515: 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)){
12516: 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 12517: 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);
12518: 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);
12519: 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);
12520: /* day and month of proj2 are not used but only year anproj2.*/
1.273 brouard 12521: dateproj1=anproj1+(mproj1-1)/12.+(jproj1-1)/365.;
12522: dateproj2=anproj2+(mproj2-1)/12.+(jproj2-1)/365.;
1.296 brouard 12523: prvforecast = 1;
12524: }
12525: else if((num_filled=sscanf(line,"prevforecast=%d yearsfproj=%lf mobil_average=%d\n",&prevfcast,&yrfproj,&mobilavproj)) !=EOF){/* && (num_filled == 3))*/
1.313 brouard 12526: printf("prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12527: fprintf(ficlog,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
12528: fprintf(ficres,"prevforecast=%d yearsfproj=%.2lf mobil_average=%d\n",prevfcast,yrfproj,mobilavproj);
1.296 brouard 12529: prvforecast = 2;
12530: }
12531: else {
12532: 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);
12533: 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);
12534: goto end;
1.258 brouard 12535: }
1.254 brouard 12536: break;
1.258 brouard 12537: case 12:
1.296 brouard 12538: 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)){
12539: 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);
12540: 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);
12541: 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);
12542: 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);
12543: /* day and month of back2 are not used but only year anback2.*/
1.273 brouard 12544: dateback1=anback1+(mback1-1)/12.+(jback1-1)/365.;
12545: dateback2=anback2+(mback2-1)/12.+(jback2-1)/365.;
1.296 brouard 12546: prvbackcast = 1;
12547: }
12548: else if((num_filled=sscanf(line,"prevbackcast=%d yearsbproj=%lf mobil_average=%d\n",&prevbcast,&yrbproj,&mobilavproj)) ==3){/* && (num_filled == 3))*/
1.313 brouard 12549: printf("prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12550: fprintf(ficlog,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
12551: fprintf(ficres,"prevbackcast=%d yearsbproj=%.2lf mobil_average=%d\n",prevbcast,yrbproj,mobilavproj);
1.296 brouard 12552: prvbackcast = 2;
12553: }
12554: else {
12555: 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);
12556: 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);
12557: goto end;
1.258 brouard 12558: }
1.230 brouard 12559: break;
1.258 brouard 12560: case 13:
1.307 brouard 12561: num_filled=sscanf(line,"result:%[^\n]\n",resultline);
12562: nresult++; /* Sum of resultlines */
12563: printf("Result %d: result:%s\n",nresult, resultline);
12564: if(nresult > MAXRESULTLINES){
12565: 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. ",MAXRESULTLINES,nresult,rfileres);
12566: 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. ",MAXRESULTLINES,nresult,rfileres);
12567: goto end;
12568: }
1.310 brouard 12569: if(!decoderesult(resultline, nresult)){ /* Fills TKresult[nresult] combination and Tresult[nresult][k4+1] combination values */
1.314 ! brouard 12570: fprintf(ficparo,"result: %s\n",resultline);
! 12571: fprintf(ficres,"result: %s\n",resultline);
! 12572: fprintf(ficlog,"result: %s\n",resultline);
1.310 brouard 12573: } else
12574: goto end;
1.307 brouard 12575: break;
12576: case 14:
12577: printf("Error: Unknown command '%s'\n",line);
12578: fprintf(ficlog,"Error: Unknown command '%s'\n",line);
1.314 ! brouard 12579: if(line[0] == ' ' || line[0] == '\n'){
! 12580: printf("It should not be an empty line '%s'\n",line);
! 12581: fprintf(ficlog,"It should not be an empty line '%s'\n",line);
! 12582: }
1.307 brouard 12583: if(ncovmodel >=2 && nresult==0 ){
12584: printf("ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
12585: fprintf(ficlog,"ERROR: no result lines! It should be at minimum 'result: V2=0 V1=1 or result:.' %s\n",line);
1.258 brouard 12586: }
1.307 brouard 12587: /* goto end; */
12588: break;
1.308 brouard 12589: case 15:
12590: printf("End of resultlines.\n");
12591: fprintf(ficlog,"End of resultlines.\n");
12592: break;
12593: default: /* parameterline =0 */
1.307 brouard 12594: nresult=1;
12595: decoderesult(".",nresult ); /* No covariate */
1.258 brouard 12596: } /* End switch parameterline */
12597: }while(endishere==0); /* End do */
1.126 brouard 12598:
1.230 brouard 12599: /* freqsummary(fileres, agemin, agemax, s, agev, nlstate, imx,Tvaraff,nbcode, ncodemax,mint,anint); */
1.145 brouard 12600: /* ,dateprev1,dateprev2,jprev1, mprev1,anprev1,jprev2, mprev2,anprev2); */
1.126 brouard 12601:
12602: replace_back_to_slash(pathc,pathcd); /* Even gnuplot wants a / */
1.194 brouard 12603: if(ageminpar == AGEOVERFLOW ||agemaxpar == -AGEOVERFLOW){
1.230 brouard 12604: printf("Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12605: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12606: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.230 brouard 12607: fprintf(ficlog,"Warning! Error in gnuplot file with ageminpar %f or agemaxpar %f overflow\n\
1.194 brouard 12608: This is probably because your parameter file doesn't \n contain the exact number of lines (or columns) corresponding to your model line.\n\
12609: Please run with mle=-1 to get a correct covariance matrix.\n",ageminpar,agemaxpar);
1.220 brouard 12610: }else{
1.270 brouard 12611: /* printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,fage, prevfcast, backcast, pathc,p, (int)anproj1-(int)agemin, (int)anback1-(int)agemax+1); */
1.296 brouard 12612: /* It seems that anprojd which is computed from the mean year at interview which is known yet because of freqsummary */
12613: /* date2dmy(dateintmean,&jintmean,&mintmean,&aintmean); */ /* Done in freqsummary */
12614: if(prvforecast==1){
12615: dateprojd=(jproj1+12*mproj1+365*anproj1)/365;
12616: jprojd=jproj1;
12617: mprojd=mproj1;
12618: anprojd=anproj1;
12619: dateprojf=(jproj2+12*mproj2+365*anproj2)/365;
12620: jprojf=jproj2;
12621: mprojf=mproj2;
12622: anprojf=anproj2;
12623: } else if(prvforecast == 2){
12624: dateprojd=dateintmean;
12625: date2dmy(dateprojd,&jprojd, &mprojd, &anprojd);
12626: dateprojf=dateintmean+yrfproj;
12627: date2dmy(dateprojf,&jprojf, &mprojf, &anprojf);
12628: }
12629: if(prvbackcast==1){
12630: datebackd=(jback1+12*mback1+365*anback1)/365;
12631: jbackd=jback1;
12632: mbackd=mback1;
12633: anbackd=anback1;
12634: datebackf=(jback2+12*mback2+365*anback2)/365;
12635: jbackf=jback2;
12636: mbackf=mback2;
12637: anbackf=anback2;
12638: } else if(prvbackcast == 2){
12639: datebackd=dateintmean;
12640: date2dmy(datebackd,&jbackd, &mbackd, &anbackd);
12641: datebackf=dateintmean-yrbproj;
12642: date2dmy(datebackf,&jbackf, &mbackf, &anbackf);
12643: }
12644:
12645: printinggnuplot(fileresu, optionfilefiname,ageminpar,agemaxpar,bage, fage, prevfcast, prevbcast, pathc,p, (int)anprojd-bage, (int)anbackd-fage);
1.220 brouard 12646: }
12647: printinghtml(fileresu,title,datafile, firstpass, lastpass, stepm, weightopt, \
1.296 brouard 12648: model,imx,jmin,jmax,jmean,rfileres,popforecast,mobilav,prevfcast,mobilavproj,prevbcast, estepm, \
12649: jprev1,mprev1,anprev1,dateprev1, dateprojd, datebackd,jprev2,mprev2,anprev2,dateprev2,dateprojf, datebackf);
1.220 brouard 12650:
1.225 brouard 12651: /*------------ free_vector -------------*/
12652: /* chdir(path); */
1.220 brouard 12653:
1.215 brouard 12654: /* free_ivector(wav,1,imx); */ /* Moved after last prevalence call */
12655: /* free_imatrix(dh,1,lastpass-firstpass+2,1,imx); */
12656: /* free_imatrix(bh,1,lastpass-firstpass+2,1,imx); */
12657: /* free_imatrix(mw,1,lastpass-firstpass+2,1,imx); */
1.290 brouard 12658: free_lvector(num,firstobs,lastobs);
12659: free_vector(agedc,firstobs,lastobs);
1.126 brouard 12660: /*free_matrix(covar,0,NCOVMAX,1,n);*/
12661: /*free_matrix(covar,1,NCOVMAX,1,n);*/
12662: fclose(ficparo);
12663: fclose(ficres);
1.220 brouard 12664:
12665:
1.186 brouard 12666: /* Other results (useful)*/
1.220 brouard 12667:
12668:
1.126 brouard 12669: /*--------------- Prevalence limit (period or stable prevalence) --------------*/
1.180 brouard 12670: /*#include "prevlim.h"*/ /* Use ficrespl, ficlog */
12671: prlim=matrix(1,nlstate,1,nlstate);
1.209 brouard 12672: prevalence_limit(p, prlim, ageminpar, agemaxpar, ftolpl, &ncvyear);
1.126 brouard 12673: fclose(ficrespl);
12674:
12675: /*------------- h Pij x at various ages ------------*/
1.180 brouard 12676: /*#include "hpijx.h"*/
12677: hPijx(p, bage, fage);
1.145 brouard 12678: fclose(ficrespij);
1.227 brouard 12679:
1.220 brouard 12680: /* ncovcombmax= pow(2,cptcoveff); */
1.219 brouard 12681: /*-------------- Variance of one-step probabilities---*/
1.145 brouard 12682: k=1;
1.126 brouard 12683: varprob(optionfilefiname, matcov, p, delti, nlstate, bage, fage,k,Tvar,nbcode, ncodemax,strstart);
1.227 brouard 12684:
1.269 brouard 12685: /* Prevalence for each covariate combination in probs[age][status][cov] */
12686: probs= ma3x(AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12687: for(i=AGEINF;i<=AGESUP;i++)
1.219 brouard 12688: for(j=1;j<=nlstate+ndeath;j++) /* ndeath is useless but a necessity to be compared with mobaverages */
1.225 brouard 12689: for(k=1;k<=ncovcombmax;k++)
12690: probs[i][j][k]=0.;
1.269 brouard 12691: prevalence(probs, ageminpar, agemaxpar, s, agev, nlstate, imx, Tvar, nbcode,
12692: ncodemax, mint, anint, dateprev1, dateprev2, firstpass, lastpass);
1.219 brouard 12693: if (mobilav!=0 ||mobilavproj !=0 ) {
1.269 brouard 12694: mobaverages= ma3x(AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
12695: for(i=AGEINF;i<=AGESUP;i++)
1.268 brouard 12696: for(j=1;j<=nlstate+ndeath;j++)
1.227 brouard 12697: for(k=1;k<=ncovcombmax;k++)
12698: mobaverages[i][j][k]=0.;
1.219 brouard 12699: mobaverage=mobaverages;
12700: if (mobilav!=0) {
1.235 brouard 12701: printf("Movingaveraging observed prevalence\n");
1.258 brouard 12702: fprintf(ficlog,"Movingaveraging observed prevalence\n");
1.227 brouard 12703: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilav)!=0){
12704: fprintf(ficlog," Error in movingaverage mobilav=%d\n",mobilav);
12705: printf(" Error in movingaverage mobilav=%d\n",mobilav);
12706: }
1.269 brouard 12707: } else if (mobilavproj !=0) {
1.235 brouard 12708: printf("Movingaveraging projected observed prevalence\n");
1.258 brouard 12709: fprintf(ficlog,"Movingaveraging projected observed prevalence\n");
1.227 brouard 12710: if (movingaverage(probs, ageminpar, agemaxpar, mobaverage, mobilavproj)!=0){
12711: fprintf(ficlog," Error in movingaverage mobilavproj=%d\n",mobilavproj);
12712: printf(" Error in movingaverage mobilavproj=%d\n",mobilavproj);
12713: }
1.269 brouard 12714: }else{
12715: printf("Internal error moving average\n");
12716: fflush(stdout);
12717: exit(1);
1.219 brouard 12718: }
12719: }/* end if moving average */
1.227 brouard 12720:
1.126 brouard 12721: /*---------- Forecasting ------------------*/
1.296 brouard 12722: if(prevfcast==1){
12723: /* /\* if(stepm ==1){*\/ */
12724: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12725: /*This done previously after freqsummary.*/
12726: /* dateprojd=(jproj1+12*mproj1+365*anproj1)/365; */
12727: /* dateprojf=(jproj2+12*mproj2+365*anproj2)/365; */
12728:
12729: /* } else if (prvforecast==2){ */
12730: /* /\* if(stepm ==1){*\/ */
12731: /* /\* anproj1, mproj1, jproj1 either read explicitly or yrfproj *\/ */
12732: /* } */
12733: /*prevforecast(fileresu, dateintmean, anproj1, mproj1, jproj1, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, anproj2, p, cptcoveff);*/
12734: prevforecast(fileresu,dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2, mobilavproj, mobaverage, bage, fage, firstpass, lastpass, p, cptcoveff);
1.126 brouard 12735: }
1.269 brouard 12736:
1.296 brouard 12737: /* Prevbcasting */
12738: if(prevbcast==1){
1.219 brouard 12739: ddnewms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12740: ddoldms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12741: ddsavms=matrix(1,nlstate+ndeath,1,nlstate+ndeath);
12742:
12743: /*--------------- Back Prevalence limit (period or stable prevalence) --------------*/
12744:
12745: bprlim=matrix(1,nlstate,1,nlstate);
1.269 brouard 12746:
1.219 brouard 12747: back_prevalence_limit(p, bprlim, ageminpar, agemaxpar, ftolpl, &ncvyear, dateprev1, dateprev2, firstpass, lastpass, mobilavproj);
12748: fclose(ficresplb);
12749:
1.222 brouard 12750: hBijx(p, bage, fage, mobaverage);
12751: fclose(ficrespijb);
1.219 brouard 12752:
1.296 brouard 12753: /* /\* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, *\/ */
12754: /* /\* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); *\/ */
12755: /* prevbackforecast(fileresu, mobaverage, anback1, mback1, jback1, agemin, agemax, dateprev1, dateprev2, */
12756: /* mobilavproj, bage, fage, firstpass, lastpass, anback2, p, cptcoveff); */
12757: prevbackforecast(fileresu, mobaverage, dateintmean, dateprojd, dateprojf, agemin, agemax, dateprev1, dateprev2,
12758: mobilavproj, bage, fage, firstpass, lastpass, p, cptcoveff);
12759:
12760:
1.269 brouard 12761: varbprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, bprlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12762:
12763:
1.269 brouard 12764: free_matrix(bprlim,1,nlstate,1,nlstate); /*here or after loop ? */
1.219 brouard 12765: free_matrix(ddnewms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12766: free_matrix(ddsavms, 1, nlstate+ndeath, 1, nlstate+ndeath);
12767: free_matrix(ddoldms, 1, nlstate+ndeath, 1, nlstate+ndeath);
1.296 brouard 12768: } /* end Prevbcasting */
1.268 brouard 12769:
1.186 brouard 12770:
12771: /* ------ Other prevalence ratios------------ */
1.126 brouard 12772:
1.215 brouard 12773: free_ivector(wav,1,imx);
12774: free_imatrix(dh,1,lastpass-firstpass+2,1,imx);
12775: free_imatrix(bh,1,lastpass-firstpass+2,1,imx);
12776: free_imatrix(mw,1,lastpass-firstpass+2,1,imx);
1.218 brouard 12777:
12778:
1.127 brouard 12779: /*---------- Health expectancies, no variances ------------*/
1.218 brouard 12780:
1.201 brouard 12781: strcpy(filerese,"E_");
12782: strcat(filerese,fileresu);
1.126 brouard 12783: if((ficreseij=fopen(filerese,"w"))==NULL) {
12784: printf("Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12785: fprintf(ficlog,"Problem with Health Exp. resultfile: %s\n", filerese); exit(0);
12786: }
1.208 brouard 12787: printf("Computing Health Expectancies: result on file '%s' ...", filerese);fflush(stdout);
12788: fprintf(ficlog,"Computing Health Expectancies: result on file '%s' ...", filerese);fflush(ficlog);
1.238 brouard 12789:
12790: pstamp(ficreseij);
1.219 brouard 12791:
1.235 brouard 12792: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12793: if (cptcovn < 1){i1=1;}
12794:
12795: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12796: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12797: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12798: continue;
1.219 brouard 12799: fprintf(ficreseij,"\n#****** ");
1.235 brouard 12800: printf("\n#****** ");
1.225 brouard 12801: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12802: fprintf(ficreseij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12803: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12804: }
12805: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12806: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12807: fprintf(ficreseij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
1.219 brouard 12808: }
12809: fprintf(ficreseij,"******\n");
1.235 brouard 12810: printf("******\n");
1.219 brouard 12811:
12812: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12813: oldm=oldms;savm=savms;
1.235 brouard 12814: evsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, strstart, nres);
1.127 brouard 12815:
1.219 brouard 12816: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.127 brouard 12817: }
12818: fclose(ficreseij);
1.208 brouard 12819: printf("done evsij\n");fflush(stdout);
12820: fprintf(ficlog,"done evsij\n");fflush(ficlog);
1.269 brouard 12821:
1.218 brouard 12822:
1.227 brouard 12823: /*---------- State-specific expectancies and variances ------------*/
1.218 brouard 12824:
1.201 brouard 12825: strcpy(filerest,"T_");
12826: strcat(filerest,fileresu);
1.127 brouard 12827: if((ficrest=fopen(filerest,"w"))==NULL) {
12828: printf("Problem with total LE resultfile: %s\n", filerest);goto end;
12829: fprintf(ficlog,"Problem with total LE resultfile: %s\n", filerest);goto end;
12830: }
1.208 brouard 12831: printf("Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(stdout);
12832: fprintf(ficlog,"Computing Total Life expectancies with their standard errors: file '%s' ...\n", filerest); fflush(ficlog);
1.201 brouard 12833: strcpy(fileresstde,"STDE_");
12834: strcat(fileresstde,fileresu);
1.126 brouard 12835: if((ficresstdeij=fopen(fileresstde,"w"))==NULL) {
1.227 brouard 12836: printf("Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
12837: fprintf(ficlog,"Problem with State specific Exp. and std errors resultfile: %s\n", fileresstde); exit(0);
1.126 brouard 12838: }
1.227 brouard 12839: printf(" Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
12840: fprintf(ficlog," Computing State-specific Expectancies and standard errors: result on file '%s' \n", fileresstde);
1.126 brouard 12841:
1.201 brouard 12842: strcpy(filerescve,"CVE_");
12843: strcat(filerescve,fileresu);
1.126 brouard 12844: if((ficrescveij=fopen(filerescve,"w"))==NULL) {
1.227 brouard 12845: printf("Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
12846: fprintf(ficlog,"Problem with Covar. State-specific Exp. resultfile: %s\n", filerescve); exit(0);
1.126 brouard 12847: }
1.227 brouard 12848: printf(" Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
12849: fprintf(ficlog," Computing Covar. of State-specific Expectancies: result on file '%s' \n", filerescve);
1.126 brouard 12850:
1.201 brouard 12851: strcpy(fileresv,"V_");
12852: strcat(fileresv,fileresu);
1.126 brouard 12853: if((ficresvij=fopen(fileresv,"w"))==NULL) {
12854: printf("Problem with variance resultfile: %s\n", fileresv);exit(0);
12855: fprintf(ficlog,"Problem with variance resultfile: %s\n", fileresv);exit(0);
12856: }
1.227 brouard 12857: printf(" Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(stdout);
12858: fprintf(ficlog," Computing Variance-covariance of State-specific Expectancies: file '%s' ... ", fileresv);fflush(ficlog);
1.126 brouard 12859:
1.235 brouard 12860: i1=pow(2,cptcoveff); /* Number of combination of dummy covariates */
12861: if (cptcovn < 1){i1=1;}
12862:
12863: for(nres=1; nres <= nresult; nres++) /* For each resultline */
12864: for(k=1; k<=i1;k++){ /* For any combination of dummy covariates, fixed and varying */
1.253 brouard 12865: if(i1 != 1 && TKresult[nres]!= k)
1.235 brouard 12866: continue;
1.242 brouard 12867: printf("\n#****** Result for:");
12868: fprintf(ficrest,"\n#****** Result for:");
12869: fprintf(ficlog,"\n#****** Result for:");
1.227 brouard 12870: for(j=1;j<=cptcoveff;j++){
12871: printf("V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12872: fprintf(ficrest,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12873: fprintf(ficlog,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12874: }
1.235 brouard 12875: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12876: printf(" V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12877: fprintf(ficrest," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12878: fprintf(ficlog," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12879: }
1.208 brouard 12880: fprintf(ficrest,"******\n");
1.227 brouard 12881: fprintf(ficlog,"******\n");
12882: printf("******\n");
1.208 brouard 12883:
12884: fprintf(ficresstdeij,"\n#****** ");
12885: fprintf(ficrescveij,"\n#****** ");
1.225 brouard 12886: for(j=1;j<=cptcoveff;j++) {
1.227 brouard 12887: fprintf(ficresstdeij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
12888: fprintf(ficrescveij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.208 brouard 12889: }
1.235 brouard 12890: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12891: fprintf(ficresstdeij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12892: fprintf(ficrescveij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12893: }
1.208 brouard 12894: fprintf(ficresstdeij,"******\n");
12895: fprintf(ficrescveij,"******\n");
12896:
12897: fprintf(ficresvij,"\n#****** ");
1.238 brouard 12898: /* pstamp(ficresvij); */
1.225 brouard 12899: for(j=1;j<=cptcoveff;j++)
1.227 brouard 12900: fprintf(ficresvij,"V%d=%d ",Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]);
1.235 brouard 12901: for (j=1; j<= nsq; j++){ /* For each selected (single) quantitative value */
12902: fprintf(ficresvij," V%d=%f ",Tvqresult[nres][j],Tqresult[nres][j]);
12903: }
1.208 brouard 12904: fprintf(ficresvij,"******\n");
12905:
12906: eij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12907: oldm=oldms;savm=savms;
1.235 brouard 12908: printf(" cvevsij ");
12909: fprintf(ficlog, " cvevsij ");
12910: cvevsij(eij, p, nlstate, stepm, (int) bage, (int)fage, oldm, savm, k, estepm, delti, matcov, strstart, nres);
1.208 brouard 12911: printf(" end cvevsij \n ");
12912: fprintf(ficlog, " end cvevsij \n ");
12913:
12914: /*
12915: */
12916: /* goto endfree; */
12917:
12918: vareij=ma3x(1,nlstate,1,nlstate,(int) bage, (int) fage);
12919: pstamp(ficrest);
12920:
1.269 brouard 12921: epj=vector(1,nlstate+1);
1.208 brouard 12922: for(vpopbased=0; vpopbased <= popbased; vpopbased++){ /* Done for vpopbased=0 and vpopbased=1 if popbased==1*/
1.227 brouard 12923: oldm=oldms;savm=savms; /* ZZ Segmentation fault */
12924: cptcod= 0; /* To be deleted */
12925: printf("varevsij vpopbased=%d \n",vpopbased);
12926: fprintf(ficlog, "varevsij vpopbased=%d \n",vpopbased);
1.235 brouard 12927: 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 12928: 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 ");
12929: if(vpopbased==1)
12930: 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);
12931: else
1.288 brouard 12932: fprintf(ficrest,"the age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12933: fprintf(ficrest,"# Age popbased mobilav e.. (std) ");
12934: for (i=1;i<=nlstate;i++) fprintf(ficrest,"e.%d (std) ",i);
12935: fprintf(ficrest,"\n");
12936: /* printf("Which p?\n"); for(i=1;i<=npar;i++)printf("p[i=%d]=%lf,",i,p[i]);printf("\n"); */
1.288 brouard 12937: printf("Computing age specific forward period (stable) prevalences in each health state \n");
12938: fprintf(ficlog,"Computing age specific forward period (stable) prevalences in each health state \n");
1.227 brouard 12939: for(age=bage; age <=fage ;age++){
1.235 brouard 12940: prevalim(prlim, nlstate, p, age, oldm, savm, ftolpl, &ncvyear, k, nres); /*ZZ Is it the correct prevalim */
1.227 brouard 12941: if (vpopbased==1) {
12942: if(mobilav ==0){
12943: for(i=1; i<=nlstate;i++)
12944: prlim[i][i]=probs[(int)age][i][k];
12945: }else{ /* mobilav */
12946: for(i=1; i<=nlstate;i++)
12947: prlim[i][i]=mobaverage[(int)age][i][k];
12948: }
12949: }
1.219 brouard 12950:
1.227 brouard 12951: fprintf(ficrest," %4.0f %d %d",age, vpopbased, mobilav);
12952: /* fprintf(ficrest," %4.0f %d %d %d %d",age, vpopbased, mobilav,Tvaraff[j],nbcode[Tvaraff[j]][codtabm(k,j)]); */ /* to be done */
12953: /* printf(" age %4.0f ",age); */
12954: for(j=1, epj[nlstate+1]=0.;j <=nlstate;j++){
12955: for(i=1, epj[j]=0.;i <=nlstate;i++) {
12956: epj[j] += prlim[i][i]*eij[i][j][(int)age];
12957: /*ZZZ printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]);*/
12958: /* printf("%lf %lf ", prlim[i][i] ,eij[i][j][(int)age]); */
12959: }
12960: epj[nlstate+1] +=epj[j];
12961: }
12962: /* printf(" age %4.0f \n",age); */
1.219 brouard 12963:
1.227 brouard 12964: for(i=1, vepp=0.;i <=nlstate;i++)
12965: for(j=1;j <=nlstate;j++)
12966: vepp += vareij[i][j][(int)age];
12967: fprintf(ficrest," %7.3f (%7.3f)", epj[nlstate+1],sqrt(vepp));
12968: for(j=1;j <=nlstate;j++){
12969: fprintf(ficrest," %7.3f (%7.3f)", epj[j],sqrt(vareij[j][j][(int)age]));
12970: }
12971: fprintf(ficrest,"\n");
12972: }
1.208 brouard 12973: } /* End vpopbased */
1.269 brouard 12974: free_vector(epj,1,nlstate+1);
1.208 brouard 12975: free_ma3x(eij,1,nlstate,1,nlstate,(int) bage, (int)fage);
12976: free_ma3x(vareij,1,nlstate,1,nlstate,(int) bage, (int)fage);
1.235 brouard 12977: printf("done selection\n");fflush(stdout);
12978: fprintf(ficlog,"done selection\n");fflush(ficlog);
1.208 brouard 12979:
1.235 brouard 12980: } /* End k selection */
1.227 brouard 12981:
12982: printf("done State-specific expectancies\n");fflush(stdout);
12983: fprintf(ficlog,"done State-specific expectancies\n");fflush(ficlog);
12984:
1.288 brouard 12985: /* variance-covariance of forward period prevalence*/
1.269 brouard 12986: varprlim(fileresu, nresult, mobaverage, mobilavproj, bage, fage, prlim, &ncvyear, ftolpl, p, matcov, delti, stepm, cptcoveff);
1.268 brouard 12987:
1.227 brouard 12988:
1.290 brouard 12989: free_vector(weight,firstobs,lastobs);
1.227 brouard 12990: free_imatrix(Tvard,1,NCOVMAX,1,2);
1.290 brouard 12991: free_imatrix(s,1,maxwav+1,firstobs,lastobs);
12992: free_matrix(anint,1,maxwav,firstobs,lastobs);
12993: free_matrix(mint,1,maxwav,firstobs,lastobs);
12994: free_ivector(cod,firstobs,lastobs);
1.227 brouard 12995: free_ivector(tab,1,NCOVMAX);
12996: fclose(ficresstdeij);
12997: fclose(ficrescveij);
12998: fclose(ficresvij);
12999: fclose(ficrest);
13000: fclose(ficpar);
13001:
13002:
1.126 brouard 13003: /*---------- End : free ----------------*/
1.219 brouard 13004: if (mobilav!=0 ||mobilavproj !=0)
1.269 brouard 13005: free_ma3x(mobaverages,AGEINF, AGESUP,1,nlstate+ndeath, 1,ncovcombmax); /* We need to have a squared matrix with prevalence of the dead! */
13006: free_ma3x(probs,AGEINF,AGESUP,1,nlstate+ndeath, 1,ncovcombmax);
1.220 brouard 13007: free_matrix(prlim,1,nlstate,1,nlstate); /*here or after loop ? */
13008: free_matrix(pmmij,1,nlstate+ndeath,1,nlstate+ndeath);
1.126 brouard 13009: } /* mle==-3 arrives here for freeing */
1.227 brouard 13010: /* endfree:*/
13011: free_matrix(oldms, 1,nlstate+ndeath,1,nlstate+ndeath);
13012: free_matrix(newms, 1,nlstate+ndeath,1,nlstate+ndeath);
13013: free_matrix(savms, 1,nlstate+ndeath,1,nlstate+ndeath);
1.290 brouard 13014: if(ntv+nqtv>=1)free_ma3x(cotvar,1,maxwav,1,ntv+nqtv,firstobs,lastobs);
13015: if(nqtv>=1)free_ma3x(cotqvar,1,maxwav,1,nqtv,firstobs,lastobs);
13016: if(nqv>=1)free_matrix(coqvar,1,nqv,firstobs,lastobs);
13017: free_matrix(covar,0,NCOVMAX,firstobs,lastobs);
1.227 brouard 13018: free_matrix(matcov,1,npar,1,npar);
13019: free_matrix(hess,1,npar,1,npar);
13020: /*free_vector(delti,1,npar);*/
13021: free_ma3x(delti3,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13022: free_matrix(agev,1,maxwav,1,imx);
1.269 brouard 13023: free_ma3x(paramstart,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
1.227 brouard 13024: free_ma3x(param,1,nlstate,1, nlstate+ndeath-1,1,ncovmodel);
13025:
13026: free_ivector(ncodemax,1,NCOVMAX);
13027: free_ivector(ncodemaxwundef,1,NCOVMAX);
13028: free_ivector(Dummy,-1,NCOVMAX);
13029: free_ivector(Fixed,-1,NCOVMAX);
1.238 brouard 13030: free_ivector(DummyV,1,NCOVMAX);
13031: free_ivector(FixedV,1,NCOVMAX);
1.227 brouard 13032: free_ivector(Typevar,-1,NCOVMAX);
13033: free_ivector(Tvar,1,NCOVMAX);
1.234 brouard 13034: free_ivector(TvarsQ,1,NCOVMAX);
13035: free_ivector(TvarsQind,1,NCOVMAX);
13036: free_ivector(TvarsD,1,NCOVMAX);
13037: free_ivector(TvarsDind,1,NCOVMAX);
1.231 brouard 13038: free_ivector(TvarFD,1,NCOVMAX);
13039: free_ivector(TvarFDind,1,NCOVMAX);
1.232 brouard 13040: free_ivector(TvarF,1,NCOVMAX);
13041: free_ivector(TvarFind,1,NCOVMAX);
13042: free_ivector(TvarV,1,NCOVMAX);
13043: free_ivector(TvarVind,1,NCOVMAX);
13044: free_ivector(TvarA,1,NCOVMAX);
13045: free_ivector(TvarAind,1,NCOVMAX);
1.231 brouard 13046: free_ivector(TvarFQ,1,NCOVMAX);
13047: free_ivector(TvarFQind,1,NCOVMAX);
13048: free_ivector(TvarVD,1,NCOVMAX);
13049: free_ivector(TvarVDind,1,NCOVMAX);
13050: free_ivector(TvarVQ,1,NCOVMAX);
13051: free_ivector(TvarVQind,1,NCOVMAX);
1.230 brouard 13052: free_ivector(Tvarsel,1,NCOVMAX);
13053: free_vector(Tvalsel,1,NCOVMAX);
1.227 brouard 13054: free_ivector(Tposprod,1,NCOVMAX);
13055: free_ivector(Tprod,1,NCOVMAX);
13056: free_ivector(Tvaraff,1,NCOVMAX);
13057: free_ivector(invalidvarcomb,1,ncovcombmax);
13058: free_ivector(Tage,1,NCOVMAX);
13059: free_ivector(Tmodelind,1,NCOVMAX);
1.228 brouard 13060: free_ivector(TmodelInvind,1,NCOVMAX);
13061: free_ivector(TmodelInvQind,1,NCOVMAX);
1.227 brouard 13062:
13063: free_imatrix(nbcode,0,NCOVMAX,0,NCOVMAX);
13064: /* free_imatrix(codtab,1,100,1,10); */
1.126 brouard 13065: fflush(fichtm);
13066: fflush(ficgp);
13067:
1.227 brouard 13068:
1.126 brouard 13069: if((nberr >0) || (nbwarn>0)){
1.216 brouard 13070: printf("End of Imach with %d errors and/or %d warnings. Please look at the log file for details.\n",nberr,nbwarn);
13071: 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 13072: }else{
13073: printf("End of Imach\n");
13074: fprintf(ficlog,"End of Imach\n");
13075: }
13076: printf("See log file on %s\n",filelog);
13077: /* gettimeofday(&end_time, (struct timezone*)0);*/ /* after time */
1.157 brouard 13078: /*(void) gettimeofday(&end_time,&tzp);*/
13079: rend_time = time(NULL);
13080: end_time = *localtime(&rend_time);
13081: /* tml = *localtime(&end_time.tm_sec); */
13082: strcpy(strtend,asctime(&end_time));
1.126 brouard 13083: printf("Local time at start %s\nLocal time at end %s",strstart, strtend);
13084: fprintf(ficlog,"Local time at start %s\nLocal time at end %s\n",strstart, strtend);
1.157 brouard 13085: printf("Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
1.227 brouard 13086:
1.157 brouard 13087: printf("Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
13088: fprintf(ficlog,"Total time used %s\n", asc_diff_time(rend_time -rstart_time,tmpout));
13089: fprintf(ficlog,"Total time was %.0lf Sec.\n", difftime(rend_time,rstart_time));
1.126 brouard 13090: /* printf("Total time was %d uSec.\n", total_usecs);*/
13091: /* if(fileappend(fichtm,optionfilehtm)){ */
13092: fprintf(fichtm,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13093: fclose(fichtm);
13094: fprintf(fichtmcov,"<br>Local time at start %s<br>Local time at end %s<br>\n</body></html>",strstart, strtend);
13095: fclose(fichtmcov);
13096: fclose(ficgp);
13097: fclose(ficlog);
13098: /*------ End -----------*/
1.227 brouard 13099:
1.281 brouard 13100:
13101: /* Executes gnuplot */
1.227 brouard 13102:
13103: printf("Before Current directory %s!\n",pathcd);
1.184 brouard 13104: #ifdef WIN32
1.227 brouard 13105: if (_chdir(pathcd) != 0)
13106: printf("Can't move to directory %s!\n",path);
13107: if(_getcwd(pathcd,MAXLINE) > 0)
1.184 brouard 13108: #else
1.227 brouard 13109: if(chdir(pathcd) != 0)
13110: printf("Can't move to directory %s!\n", path);
13111: if (getcwd(pathcd, MAXLINE) > 0)
1.184 brouard 13112: #endif
1.126 brouard 13113: printf("Current directory %s!\n",pathcd);
13114: /*strcat(plotcmd,CHARSEPARATOR);*/
13115: sprintf(plotcmd,"gnuplot");
1.157 brouard 13116: #ifdef _WIN32
1.126 brouard 13117: sprintf(plotcmd,"\"%sgnuplot.exe\"",pathimach);
13118: #endif
13119: if(!stat(plotcmd,&info)){
1.158 brouard 13120: printf("Error or gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13121: if(!stat(getenv("GNUPLOTBIN"),&info)){
1.158 brouard 13122: printf("Error or gnuplot program not found: '%s' Environment GNUPLOTBIN not set.\n",plotcmd);fflush(stdout);
1.126 brouard 13123: }else
13124: strcpy(pplotcmd,plotcmd);
1.157 brouard 13125: #ifdef __unix
1.126 brouard 13126: strcpy(plotcmd,GNUPLOTPROGRAM);
13127: if(!stat(plotcmd,&info)){
1.158 brouard 13128: printf("Error gnuplot program not found: '%s'\n",plotcmd);fflush(stdout);
1.126 brouard 13129: }else
13130: strcpy(pplotcmd,plotcmd);
13131: #endif
13132: }else
13133: strcpy(pplotcmd,plotcmd);
13134:
13135: sprintf(plotcmd,"%s %s",pplotcmd, optionfilegnuplot);
1.158 brouard 13136: printf("Starting graphs with: '%s'\n",plotcmd);fflush(stdout);
1.292 brouard 13137: strcpy(pplotcmd,plotcmd);
1.227 brouard 13138:
1.126 brouard 13139: if((outcmd=system(plotcmd)) != 0){
1.292 brouard 13140: printf("Error in gnuplot, command might not be in your path: '%s', err=%d\n", plotcmd, outcmd);
1.154 brouard 13141: printf("\n Trying if gnuplot resides on the same directory that IMaCh\n");
1.152 brouard 13142: sprintf(plotcmd,"%sgnuplot %s", pathimach, optionfilegnuplot);
1.292 brouard 13143: if((outcmd=system(plotcmd)) != 0){
1.153 brouard 13144: printf("\n Still a problem with gnuplot command %s, err=%d\n", plotcmd, outcmd);
1.292 brouard 13145: strcpy(plotcmd,pplotcmd);
13146: }
1.126 brouard 13147: }
1.158 brouard 13148: printf(" Successful, please wait...");
1.126 brouard 13149: while (z[0] != 'q') {
13150: /* chdir(path); */
1.154 brouard 13151: printf("\nType e to edit results with your browser, g to graph again and q for exit: ");
1.126 brouard 13152: scanf("%s",z);
13153: /* if (z[0] == 'c') system("./imach"); */
13154: if (z[0] == 'e') {
1.158 brouard 13155: #ifdef __APPLE__
1.152 brouard 13156: sprintf(pplotcmd, "open %s", optionfilehtm);
1.157 brouard 13157: #elif __linux
13158: sprintf(pplotcmd, "xdg-open %s", optionfilehtm);
1.153 brouard 13159: #else
1.152 brouard 13160: sprintf(pplotcmd, "%s", optionfilehtm);
1.153 brouard 13161: #endif
13162: printf("Starting browser with: %s",pplotcmd);fflush(stdout);
13163: system(pplotcmd);
1.126 brouard 13164: }
13165: else if (z[0] == 'g') system(plotcmd);
13166: else if (z[0] == 'q') exit(0);
13167: }
1.227 brouard 13168: end:
1.126 brouard 13169: while (z[0] != 'q') {
1.195 brouard 13170: printf("\nType q for exiting: "); fflush(stdout);
1.126 brouard 13171: scanf("%s",z);
13172: }
1.283 brouard 13173: printf("End\n");
1.282 brouard 13174: exit(0);
1.126 brouard 13175: }
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